Velomobile Knowledge

Work in progress

Translation notes:

  • In the automobile world the steering rods are attached to the hub on the steering knuckle. I have used this term in preference to “steering plate” – often in a velobobile the steering knuckle is a flat plate of metal.

  • Some of the terms are the ones used by Google Translate / DeepL – they have either escaped my attention or I didn’t have a better idea. Please feel free to suggest better translations on the forum thread.

General Notes

What is a Velomobile?

A fully covered recumbent bicycle, typically with three wheels. The fairing provides weather protection and improves aerodynamics significantly, but also provokes nicknames such as „egg on wheels“, „torpeedo“ or „pill“.

How much does it cost?

The prices for new velomobiles usually range between 5000 and 10000 EUR. Of course they are cheaper second hand but because of, not only low volume production, but also increasing demand, velomobiles are relatively stable in value – often up to very high mileage.

Second-hand prices 2019

Fig. 1 Second-hand prices for velomobiles in 2019 (without consideration of original price, configuration and condition). As one can see, the Quest, one of the most produced velomobiles (see Fig. 6), has been offered accordingly frequently and at various prices. In contrast, newer models like DF or Quatrevelo (formerly: Quattrovelo) are still correspondingly expensive used.

Höchst- und Medianalter der Velomobil-Modelle

Fig. 2 Age of velomobiles produced so far, broken down by model. Since production is constantly increasing, mainly due to additional new models, the median age is correspondingly low. (Status 09/2022; Milan only from MK6)

How much does one weigh?

Most velomobiles weigh between 20 and 30 kg. A few older models or those with an electric assist motor are often well above that weight. There are also special racing designs weighing from about 16 kg. See Fig. 30 for velomobile models and their respective weight ranges.

Who makes such a thing?

There are a handful of manufacturers that make velomobiles, most are built by hand – most of these companies are based in the Netherlands (three of them in Dronten), but there are also manufacturers in Australia, Germany, Denmark, France and Switzerland. The principle area of distribution for velomobiles is generally speaking the Netherlands, Benelux and Germany.

Who drives one?

Frequent drivers can be seen in the Graph of monthly driving performance. The velomobile is often used for daily commuting.

The main areas of distribution of velomobiles are the Netherlands and Germany. Sales data of the Dutch manufacturers (Velomobiel.nl, Intercitybike) are publicly available, see Fig. 3.

Countries to which the Dutch velomobiles were sold

Fig. 3 Countries to which Dutch velomobiles (i.e. Velomobiel.nl and Intercitybike) were sold

Age distribution, according to a survey in the Velomobile Forum, see Fig. 4.

age distribution of velomobile riders from a survey in the Velomobil-Forum

Fig. 4 Age distribution of velomobile riders from a survey in the Velomobil-Forum (as of: 2017–2019)

Where can I find more information?

In general, there is little literature and few websites. Most manufacturers have a website, but the information there is often poor and/or outdated. But at least some dealers have very informative websites with tips on the models they sell.

The main sources of information, however, are internet forums:

The German velomobile forum, with 4730 registered members (of which about 25% have specified a velomobile model as of March 2021), is the leading forum, at least in Europe, where riders from other countries also participate. The wiki there is rather poorly maintained, most of the information is in the forum, often buried deep in long discussion threads – you have to search for it.

More information can be found in the social media and Youtube channels, here Saukki’s channel http://www.saukki.com/ is especially worth mentioning.

The scene is also well networked, there are regulars’ meetings in most of the larger German cities. The big meeting place for the scene, however, is the annual Spezialradmesse SPEZI in Germersheim – practically all manufacturers and dealers are present here, even if they are not exhibitors.

Where can I buy one?

Apart from at the manufacturers, there are also a few dealers spread across Europe (see Fig. 5). The best way to see who sells which models is checking dealers listed on the manufacturer’s website.

Map of dealers and manufacturers

Fig. 5 Map of velomobile manufacturers and dealers (online interactive)

Who are the most important manufacturers and what are their models called?

Manufacturers and models (alphabetical, incomplete):

Interactive model comparison

Photo galleries:

Relative market shares of Dutch velomobiles

Fig. 6 Relative market shares of Dutch velomobiles (Velomobiel.nl and Intercitybike as of: 02/2020)

Production figures from Velomobile World

Fig. 7 Production figures from Velomobile World incl. earlier production (before 2020) in the Netherlands (as of November 2021). No earlier figures are available for the Milans, therefore only from MK6.

Why are three manufacturers based in Dronten?

As you can see in Fig. 8, current velomobile production goes back almost entirely to the Alleweder. Today’s manufacturers were either directly involved in the production of the Alleweder or its successors, or were involved as suppliers. That’s why the local proximity is no wonder. And for the same reason, many models use the same shock absorbers and the same tiller – they fall back on tried and tested components from friendly companies. Since several manufacturers have now moved their production to Velomobile World in Romania, there will probably also be some shared components and similar production processes in the future.

Velomobile family tree

Fig. 8 Velomobile family tree (large version online), there are close personal links between the Dutch and also partially some German manufacturers.

Are there velomobiles for more than one person?

At least the Leiba Cargo can carry a passenger, and a very small person or child fits in the Quattrovelo. However, they cannot join in and pedal. Real tandems exist but only as prototypes, not as a series. Only the DuoQuest is now being built as a small series, but it’s more of a fun vehicle and not particularly fast.

But the point of having a tandem is also somewhat questionable: a velomobile tandem is much heavier than a solo, and if you drive it alone also much slower. In addition, it is not cheaper than two separate velomobiles and also not more efficient. (A correspondingly highly optimised tandem velomobile is conceivable, but has not yet been built, and would only have minor advantages and significant disadvantages with much larger external dimensions and a larger turning circle).

Which model is the best?

It depends on your requirements:

  • First of all, it has to fit you and you have to feel comfortable in it. For an average sized velomobile driver (approx. 1.70–1.90 m) this is much easier than for very tall or short people.

  • The seating position also differs, in some one sits more upright and in others almost lying down. So it depends on your prefered riding style.

  • Some models are particularly fast because they are as light and stiff as possible. But they also tend to be more uncomfortable (less space, firmer chassis).

  • Some models offer a particularly large amount of storage space for luggage or for bulky items.

  • Some models have a particularly large opening for entry.

  • Depending on the route, models with larger widths can be impractical.

  • Depending on your garage space, particularly long models can be hard to park.

So there are many criteria, you can (and should) look at the technical data and tests and reviews, but whether it really fits you or not you will only notice during a test drive. Even if your body length fits, it may be too tight at the shoulders, for example.

How complete should a test drive be?

Hard to say, you will see if you fit in it, if you can see out and can pedal without problems. But whether a Velomobile is really the right thing for you, you won’t know after a short test drive – your muscles have to adapt over several 100 km, and an efficient and correct driving style takes weeks to months to master. And finally can a velomobile work for you on your everyday routes? For example in city traffic where it hardly shows its qualities. If you are unsure you should take into account that you may have to resell the velomobile – or it maybe best to buy a cheap used one to start off with to gain your first experiences.

Can you build a velomobile yourself?

Of course, most of the parts can easily be made in the home workshop, and the special components such as struts can be obtained from the manufacturers. However, a velomobile made of GRP/CRP is very tedious because you have to produce a master model and a negative mold - that’s why most hobby projects use other materials such as plywood or aluminum.

There are also some blueprints/building instructions; E.g. for the Agilo (made of plywood; design by Bodo Sitko) or the Meufl asphalt pedal boat (frame in positive carbon construction, body made of foam; by Harald Winkler).

A few velomobile models were also sold as kits that only required assembly; Eg various Alleweder models (A2, A4, A4, FAW) and infrequiently some Go-One models.

If velomobiles are so fast, why don’t you see them in bike races?

Competitive cycling, especially in the professional sector, is almost entirely run by the UCI (Union Cycliste Internationale). This organisation has proven to be very conservative in the past. Not only are recumbent bikes forbidden in their set of rules, but even triathlon bikes and all sorts of other innovations and optimizations - only classic diamond frames are allowed. And professional cyclists only ride the races where money can be made - namely those of the UCI - and use the appropriate material.

But there is another reason: in conventional cycle races, a lot of energy can be saved by slipstreaming because the air resistance is much lower. On the other hand, in a velomobile race, drag is always low, so there is no reason to form a peloton. Accordingly, the team, management work and tactics play a much smaller role, the performance of the individual is much more important - a race would therefore have more of the character of an individual time trial and would therefore be more predictable and less exciting.

Body and geometry

Why do Velomobiles have three wheels?

A two-wheeler without fairing is significantly more compact than an unclad tricycle, and a whole lot lighter because a wheel and the steering and possibly suspension mechanisms are omitted. This is no longer so clear with bodywork, the vehicle is wide, regardless of whether it has two or three wheels, and together with the weight of the fairing, the relative weight difference is significantly smaller. Another reason is the wind sensitivity. A fairing means a lot of exposed side area for cross winds, and a two-wheeler is more sensitive to this because it is narrower and taller, so it offers even more side surface. In addition, you can not easily put your feet on the ground to catch a fall on a faired two-wheeler. That is why multi-track velomobiles ultimately prevailed because they are not that much heavier and slower than single-track velomobiles, but are more relaxed to drive, especially at high speeds and on long journeys.

Why are there usually two wheels in front and one behind?

There are two reasons for this:

  • First, driving stability. In a curve, the force of inertia pulls the vehicle tangentially out of the curve. With two front wheels (so-called tadpole arrangement), the front wheel on the outside of the curve is exactly in the direction in which the inertia is acting – the vehicle cannot tip over so easily. If there are two rear wheels (so-called delta arrangement), there is no wheel there, so it tilts more easily.

  • Second, the aerodynamics. It is easier to build a fast vehicle with the front wheels forming the widest part and close to the front or in the middle. If you had the two wheels in the back, you would have a long, unused space behind the rear wheels – while the rear wheel fills this space in a tadpole configuration.

Why are the front wheels small and the rear wheel big?

The front wheels are small for space reasons – if you wanted big wheels in closed wheel wells then the vehicle would have be very tall, and since you want to steer with the front wheels, the wheel wells would need to be deep enough, which in turn increases the overall width. So you make the front wheels no higher than the interior space that the recumbent cyclist needs for pedaling.

The rear wheel, on the other hand, has enough space because behind the head of the driver the vehicle is high anyway (exception: four-wheeler <Why a velomobile with four wheels?>). A large bike wheel has slightly lower rolling resistance, and secondly, it transfers the pedaling force of the driver to the road over a longer distance (larger circumference) than with a small rear wheel, there the wheel needs to turn faster for the same speed, so the chainring should have about 100 teeth or you will need an intermediate gear box. Small rear wheels do provide more torque for climbing steep hills.

Why is the drive wheel usually in the rear?

Because the steering is in front. It would be very cumbersome to drive a steered wheel because you would have to steer the chain somehow. Driving both front wheels would require, for example, a cardan shaft and a differential, and there is no space for this because the driver sits between the wheels. It is therefore the lesser evil to run a long chain to the back wheel, the two idler pulleys under the seat only take up a few centimeters extra height.

Why a velomobile with four wheels?

A four-wheeler is a velomobile that has two rear wheels instead of one. In order to not have to make the hood extremely wide, small wheels are used as rear wheels so that you can place them closer to the cente of gravity.

Four-wheelers have a few advantages:

  • The tipping stability is higher.

  • There is a larger luggage compartment between the rear wheels under the hood which can accommodate bulky items.

  • The rear breaks out less easily because, firstly, two wheels are hardly ever in the air at the same time, and, for other reasons they don’t both lose grip at the same time (see: Why is it dangerous when the rear wheel slides?) And secondly the rear wheels are closer to the cente of gravity so that they carry more weight.

  • If both rear wheels are driven then traction is better, also because they carry more load. For the same reason, a single driven small rear wheel is only slightly worse than a large driven rear wheel on a tricycle.

  • You have four tyres of the same size, so you only have to take one spare tube and tyre with you.

But there are also disadvantages:

  • An additional wheel with an additional or larger swing arm means more weight.

  • Two driven rear wheels are problematic when cornering if you use a rigid axle, they will wash out in curves. You can solve this with a differential, but this in turn means more weight and complexity and causes additional drive losses.

  • If you only drive one rear wheel, the traction is less than with a tricycle.

  • A small drive wheel needs a higher gear ratio, an intermediate gearbox may be necessary, which in turn means more weight.

  • The chain runs in the middle, but the drive wheel is on the side so you need a heavy, torsionally stiff axle.

  • Small wheels have higher rolling resistance.

Why is a four-wheeler more stable?

Tipping stability is determined by how wide the vehicle is and how low the centre of gravity is. A four-wheeler is no wider than a three-wheeler, the centre of gravity is the same height, and the tipping line (i.e. the line connecting the front wheel and the rear wheel on the side of tipping) is hardly further away from the centre of gravity, namely only at the rear – but the centre of gravity is closer to the front wheels than to the rear wheel. Why is a four-wheeler nevertheless noticeably more stable against tipping?

Firstly, the position of the centre of gravity relative to the tipping line determines the balance:

  • If the centre of gravity is above the tipping line, the equilibrium is unstable – any shift of the centre of gravity across the tipping line lowers it, i.e. it tips to one side or the other.

  • If, on the other hand, the centre of gravity is next to the tipping line, the equilibrium is stable – in order to cross the tipping line, the centre of gravity must first be raised.

  • The wider the vehicle, the higher the centre of gravity must be raised to cross the tipping line.

  • The lower the centre of gravity, the steeper it must be lifted initially, the lifting corresponds to the cosine of the angle – first it goes upwards, then increasingly to the side. I.e. a low centre of gravity not only has to be lifted higher when tilting, but the lifting is also steeper at the beginning.

This means that a multitrack is stable because the centre of gravity is located on the inside between the tipping lines and thus has to be lifted first when tipping. With a three-wheeler, however, this only applies to the front wheels – the rear wheel is in the middle of the vehicle, and the mass there is above its point of contact, so the three-wheeler is in an unstable equilibrium at the rear wheel. On a four-wheeler, on the other hand, the mass is in stable equilibrium along its entire length.

Secondly, there is the suspension. A vehicle with suspension does not tilt immediately, but before that the wheels spring up or down – the centrifugal force is not compensated by the lifting of the centre of gravity, but by the spring force of the chassis. This causes the body to rotate around the longitudinal axis between the wheels. However, this only works with the wheels on the outside – in the case of a three-wheeler, this means only the front wheels, while the rear wheel continues to tilt over its point of contact. With a four-wheeler, on the other hand, all four wheels can compress in the curve before the vehicle starts to tip – the front wheels then do not have to counteract the roll alone, so their suspension can be softer. In addition, with a four-wheeler, the rear suspension can be designed so that the centre of roll during compression is higher up compared to an independent suspension (as with the front wheels) – so the roll axis is not at road level as with the rear wheel of the three-wheeler, but closer to the centre of gravity, which makes the vehicle roll less.

And thirdly, a four-wheeler is wider at the rear, which means that the luggage there does not have to be stacked as high, which brings the centre of gravity further down, moreover, the luggage there is next to the tipping line instead of above it.

Open or closed wheel wells?

Both have advantages and disadvantages:

  • Closed wheel wells are slightly better aerodynamically.

  • With closed wheel wells it doesn’t matter aerodynamically how far inside the wheel sits, with open wheel wells, it must be as flush as possible on the outside, i.e. the adjustment is more flexible.

  • Closed wheel wells are often slightly larger, so wider tyres can be used (these are also correspondingly taller). With open wheel wells, on the other hand, you try not to make them as large because the gap between the wheel and bodywork disturbs airflow.

  • Velomobiles with closed wheel wells are not necessarily wider, which means that they have a smaller track width and are therefore somewhat more sensitive to tipping.

  • And of course, with open wheel wells, removing and installing the wheels and adjusting the brakes is much easier. However, there are now velomobiles with closed wheel wells, where you can access the axles from outside via maintenance hatches, there at least the removal and installation of the wheels is similarly simple.

One-sided or two-sided swing arm?

  • A double-sided swing arm is the classic design, normal rear hubs fit. Special hubs are required for one-sided swingarms.

  • Wheel hub motors are only available with two-sided mounting (except for one direct drive model from Canada). With one-sided swingarms you can only easily install a bottom bracket motor.

  • In the case of a double-sided swing arm, the forces act on the cente, this makes torsion significantly lower. A one-sided swingarm must be built much more solid to be similarly torsionally rigid.

  • With a one-sided swingarm, you can change the tyre without having to remove the wheel.

  • With a one-sided swingarm there is more space on the left side of the wheel arch for baggage. There is less space on the right-hand side because the swing arm is much more voluminous, however, the space for luggage would hardly be usable anyway, because the oily chain is there.

  • In the case of a one-sided swingarm, there is only one hole in the wheel well on the right side through which dirt can penetrate into the interior. The left side remains completely closed and clean.

  • With a one-sided swing arm it is possible to remove the wheel without touching the sprocket and chain. With a double-sided swing arm, it will be more difficult to disassemble the hub because there is not much space on the sides.

Why build velomobiles from carbon fibre or fiberglass?

Fiber-reinforced plastic can withstand roughly the same amount of force as steel – but is significantly lighter. However, the fibers must always be layed in the pulling direction. So you always need parts with relatively large diameters and smooth transitions so that the forces that occur can always be absorbed somewhere by fibers under tension. This is not the case with metal, this can withstand the same amount of force in all directions and can also be subjected to pressure.

A thin diamond frame with highly stressed parts can therefore be built from steel, carbon offers few advantages, unless you build a much more voluminous frame. A velomobile, on the other hand, deals with large areas that are lightly loaded. That would be much too heavy to build from sheet steel – but with carbon you can build a thin, load-bearing shell, which is neverthelessis stiff enough to transmit driving forces. So you can do without an additional supporting inner frame, the components combine both aerodynamic and load-bearing functions.

Why curved surfaces?

Because of the rigidity. This depends, among other things, on the thickness of the material. When you bend the material, it is stretched on the outside and compressed on the inside. The farther apart the inside and outside are, the greater the stretch/compression can be and the stiffer the material.

This is the reason why heavily loaded components need to be built quite voluminously – a tube twice as large is twice as thick, so it needs twice as much material and is twice as heavy. However, since the inside and outside of the bend are twice as far apart, it is four times as stiff, or can be constructed with correspondingly thin walls with the same stiffness and is lighter overall.

This does not only apply to closed forms such as tubes, but also to open surfaces. Bending a material in one direction increases the thickness for a bend perpendicular to it. A 1 mm thick material that is curved down 1 cm is therefore as stiff in the other direction as a material that is 1 cm thick. This is exactly what makes corrugated iron or corrugated cardboard so rigid compared to the flat material. Or: If you fold a piece of pizza in the transverse direction, it no longer hangs down in the longitudinal direction, but becomes stiffer there.

And that’s the reason why the curved surfaces of the body of a velomobile can make a significant contribution to rigidity, for example, the chain channel is a tightly curved shape in the transverse direction, and therefore it is incredibly stiff in the longitudinal direction and can withstand the chain forces pulling in that direction.

Why don’t many velomobiles have foot holes?

  • Because they are aerodynamically disadvantageous.

  • Because they too impair torsional rigidity, in such a velomobile either more material is needed to acheive the same rigidity, or power transmission is less efficient.

  • Because dirt, water and winter snow can penetrate from below.

Glass dome or motorcycle visor?

Admittedly, a rolling egg with a glass dome looks very chic, and from inside you have a much nicer panoramic view than from inside a dark, closed velomobile with small window openings in the :term`hood`. But in the rain, cold and dark it has massive disadvantages, because the raindrops worsen the view, especially with oncoming headlights, and such a window also mists up. You have all that with a motorcycle visor too of course, but the area you have to keep clear is much smaller. You can also easily replace a visor if it is scratched after wiping it too much. A visor is also relatively vertical compared to looking through a glass dome at a much flatter angle. These problems can be solved in cars, namely with a powerful fan and thousands of watts of engine waste heat and large windshield wipers. Ultimately, such a velomobile is more of a fair weather vehicle. But it shouldn’t be too sunny either, because otherwise it heats up under the glass dome like in a greenhouse.

Do you see enough with a hood and their tiny windows?

Actually you do.

Of course, you have a much smaller field of vision than if you were driving in convertible mode, overhead the view is of course completely restricted. But you can see enough to the front and to the sides – about as much as in a car. After all, you sit a lot closer to the windows so that they offer the same angle of view, even though they are smaller. And 95% of the time you just have to look through the front visor to see traffic, only at intersections do you need to look over your shoulders to the side. With rear view mirrors you can see the traffic behind you.

Steering

Tank steering or Tiller steering?

This is largely a matter of taste, but both mechanisms have advantages and disadvantages: **

  • With the tiller steering , the position of the hands is flexible, you can steer no matter where you hold the handlebar.

  • The position of the tank steering lever is fixed, some prefer this because it allows them more precise and quicker control, which is particularly interesting in races.

  • Since tank steering is often very sensitive, tiny finger movements are sufficient. However, the arms need to be supported for this. Armrests must be fitted so that every bump does not lead to an involuntary steering movement.

  • With a Tiller mechanically coupled brake levers are easy to implement – the brake levers are next to each other anyway and only need to be connected. With tank steering you can only install hydraulic brakes and couple the hydraulic hoses, and then you would need a fallback so that a hydraulic leak does not lead to the failure of both brakes.

  • For this reason, good-natured steering is necessary for tank steering, i.e. without or with negative scrub radius. With a Tiller, you either have coupled brake levers anyway, or at least can operate the individual brake levers together, so you have much more predictable behavior.

  • With a Tiller the arms are higher up, which means that in hot weather you can hang your arms out of the cockpit in a double manta position and still steer with them. This is completely impossible with tank steering, because at least one hand must be down on the steering lever and you would have to sit wrongly to bring the other hand up.

  • With tank steering, the arms must be next to the body, this space cannot be used for luggage. With a tiller, the elbows are much higher.

  • When it comes to tank steering, the braking and shift cables are shorter and thus there is less friction.

  • With tank steering, the upper body is better ventilated because the hands are not in front of it.

  • A Tiller is mechanically more complicated, with a universal joint on the steering bridge and a split tie rod.

  • The tank steering tie rod is not for steering, the steering levers are directly connected to the steering knuckles so the tie rod only couples the two wheels.

  • With tank steering, holes are required at the bottom of the steering lever through which water can drain when driving through a deep puddle.

Why is there no rear steering?

Rear wheel steering is more difficult to control. When cornering the centrifugal force pulls the vehicle to the outside of the curve. Front wheel steering steers the front wheels towards the inside of the bend, i.e. both forces act in opposite directions, the vehicle must be actively steered into the bend. With rear wheel steering it is the other way around, to go into the curve, it has to swing the tail outwards, and as the centrifugal force increases and it is drawn outwards – it tends to oversteer. Countermeasures are therefore needed to retain control at high speeds. Example: The kleine Schwarze by Harald Winkler has a counterweight that is pulled outwards by centrifugal force in the curve and straightens the steering.

What’s the deal with the scrub radius?

In the case of a single-track, the front wheel is located in the longitudinal axis of the bicycle and the steering bearing above the front wheel, i.e. the steering axis goes through the wheel. This is not necessarily the case with a velomobile: since the front wheels are suspended on one side, and it is not like a steering bearing at all, the steering axis does not necessarily go through the front wheel, but hits the ground next to it. So if you turn the steering, the front wheel does not turn on the spot, but rolls on a circular path around the intersection between the steering axis and the road. And the radius of this circular path is the scrub radius. Incidentally, this can be positive or negative:

  • If the steering axis runs next to the front wheel, i.e. points to a point inside the front wheel, then one speaks of a positive scrub radius.

  • If the steering axle is at an angle so that its extension goes through the front wheel and hits the ground on the outside, this is called a negative scrub radius.

The scrub radius is important when braking: when a front wheel brakes, it pulls the shock absorber backwards – not directly, but tangentially via a lever, the length of which is the scrub radius. This lever creates torque on the shock absorber. If the scrub radius is positive, the wheel pulls the shock absorber on the outside backwards – this steers the wheel outwards, i.e. the velomobile moves in the direction in which the brakes are applied. If the steering wheel radius is negative, it is the other way round, the front wheel steers inwards, the velomobile drives in the direction in which the brakes are not applied.

Now which is better? Ideally, the velomobile behaves neutrally and does not steer when braking. However, one-sided braking always ensures a change of direction even without a scrub radius – the braked wheel covers a shorter distance, so it is the inside of the curve. Therefore, one tends to have a slightly negative scrub radius, which has the opposite effect, and thus compensates for the behavior mentioned.

Where is the steering axis with a MacPherson strut?

It is not so easy to say. At first glance, the suspension strut rotates around itself when steering. But this only applies to the upper end: the suspension strut is attached to the wheel well, and rotates around this attachment point. On the other hand, there is no fixed pivot at the bottom.

The suspension strut is held at the bottom by trailing arm and wishbone (see: What are the names of all the parts on the chassis?), And this results in a virtual pivot point. However, this is not fixed, but changes somewhat with the steering angle, it is roughly where the extension of the trailing arms and wishbones would cross. And so it is also possible that the pivot point is somewhere between the spokes or even outside next to the wheel, although the shock absorber and the steering knuckle are completely inside next to it.

You can have a negative scrub radius by making the angle between the trailing arm and wishbone smaller. Here’s an example:

  • The trailing arm is not attached to the front of the wheel well so that it would be approximately parallel to the longitudinal axis of the vehicle, but inside the wheel well so that it runs obliquely outwards.

  • The steering knuckle is extended to the front and the trailing arm shortened so that it hits the steering knuckle less parallel.

What is caster and what is its function?

With a conventional bicycle headset, bearings are usually not vertical. This so-called steering head angle is usually around 70 °, i.e. the head tube points 20 ° forward. The tracking point is accordingly in front of the front wheel. This results in self straightening torque – the front wheel is pulled behind the steering axle, which means that the steering adjusts itself automatically. The larger the caster, i.e. the distance between tracking point and wheel contact patch, the more pronounced is this behavior.

This behavior has an important function for a single-tracker: If the bicycle tilts to the side (i.e. the cente of gravity moves to the side) while driving slowly, it automatically steers in this direction, i.e. the wheel moves under the cente of gravity again and thus counteracts the tipping. Cornering and inclination are thus inseparable – with a lean you can trigger cornering (e.g. when driving or pushing hands-free), and due to the unstable position of the cente of gravity (the steering head bearing is the highest above the ground), a steering lock also leads to opposite tilting.

With multiple track is it is fundamentally different, the steering and inclination are completely decoupled there – cornering does not push the cente of gravity back into the cente, but it always stays the same, so it does not trigger turning. The caster is therefore ineffective when driving slowly. When cornering quickly, there is at least the centrifugal force to which the caster reacts, the caster simply counteracts any fast cornering. A velomobile can lean very well – namely if it tips over when cornering at high speed. Then the wheel on the inside of the curve lifts and the velomobile becomes a single-tracker – but only when it is tilted so far that the cente of gravity is beyond the wheel on the outside of the curve does the weight also contribute to counter-steering.

Thus caster in the velomobile could at least improve straight-ahead running at high speeds. But firstly, the space in the wheel well is limited, you can only slightly tilt the shock absorber. Secondly, the steering on a tadpole multi-track cannot turn as freely as on a single-track, because the entire steering linkage (i.e. steering lever or tiller) still has to be moved, which would require more power or would require more caster.

The caster is also relevant for external forces. For example, it is the reason why you can push a two-wheeler on the saddle and steer it by tipping it sideways. But wind is also a side force, especially with the velomobile with its large lateral contact surface. When the latter pushes the velomobile to the side, the caster ensures that the wheels turn in accordingly, that improves straight-line running but it increases wind sensitivity.

What about the Ackermann Condition?

When driving straight ahead, all wheels of a multi-lane vehicle are parallel. It’s different in a curve, because not all wheels are the same distance from the cente of the curve, they have to drive on different narrow circular arcs, the inside wheel must be turned more than the outside wheel. This is done by a so-called steering trapeze – the tie rod is shorter than the distance between the pivot points of the front wheels. Ultimately, the extensions of all axis must intersect at one point when cornering. If this is not the case, the wheels scrub on the road.

This is not a big problem with velomobiles, as tight bends only make up a negligible part of the total distance.

Ackermann condition

Fig. 9 Ackermann condition: In a curve, all wheels steer around a common cente. (Source:)

Chassis

Why are the front wheels at an angle? Doesn’t that slow you down?

Of course it does. Imagine that a wheel is extremely sloped, almost horizontal – it would almost no longer roll, but would rub against the tyre flanks tangentially to the rim.

However, smooth running is not the only goal. Another is to achieve the largest possible track width – this means that the velomobile does not tip over so easily and you can drive through curves faster. A wide velomobile would have a larger cross-sectional area. While the space in the middle is needed to accommodate the driver’s legs as well as the shock absorber, drum brake and hub, significantly less space is needed below and above. If the wheels now provide negative camber, with the same hub spacing, you get a larger track width and a smaller width at the top, i.e. the Velomobile is narrower at the top (= less air resistance) and wider at the bottom (= higher cornering speed), and you can get this with a little more rolling resistance. For this reason, in most velomobiles, the axles of the front wheels are not perpendicular to the struts, but at an angle of 86°, that tilts the wheels closer to the strut at the top. However, the negative camber does not have to be 4°, and the struts themselves do not have to be vertical. Their angle depends on the setting of the tie rod, trailing arm and wishbone.

What are the names of the chassis parts?

How this looks in reality can be seen in the next picture.

left front wheel, from above

Fig. 10 Sketch of the left front wheel (top view)

How it looks in reality can bee seen in Fig. 11.

  • The velomobile is shown in gray on the sketch, the wheel in black. Direction of travel: to the right.

  • The trailing arm is shown in green, it is usually screwed to the front of the wheel well. Together with the wishbone, it guides the steering knuckle. The dashed line shows how it can move (the steering knuckle and wishbone must follow the movement).

  • The wishbone is shown in blue, it is typically attached in the middle of the vehicle under the steering bridge. The radius of movement also shown as a dashed line.

  • The tie rod is shown in red, it connects the steering knuckles of both front wheels, and the tiller may be attached to it. The track is adjusted by changing the length of the tie rod, see: How do you know if tracking is set correctly?

  • The steering knuckle is shown in orange, it is located at the lower end of the shock absorber, and the three steering linkages above are attached to it on ball ends. Since their distance is constant, the rotation of the steering knuckle is determined by the fact that the attachment points of the trailing and transverse links can only move along the dashed lines. See: Where is the steering axis with a MacPherson strut?

With tank steering, the rods leading to the tank steering levers are attached to the rear of the steering knuckle.

Mango steering linkage

Fig. 11 Steering linkage on a Mango (view from the front on the underside, parts removed and placed on the outside of the body in the appropriate places). From front to rear: trailing arm, wishbone, tie rod (in two parts and connected to a Tiller)

Toe-in, toe-out – which is correct?

Absolutely none. The wheels should be completely parallel. Anything else increases rolling resistance and above all tyre wear.

Cars sometimes use a slight toe-in or toe-out, You can afford that because the tyres are less sensitive and the engine has enough power. The advantages:

  • Steering play: A worn steering with play can flutter. A slight toe-in or toe-out forces the steering into a fixed position.

  • Turn in: In the curve, the weight shifts to the outer wheels. A toe-in on the front wheel supports the steering accordingly, because the loaded outer wheel then turns a little in the straight-ahead position, and the opposite influence of the inner wheel is reduced because it is relieved.

  • The situation is reversed on the rear wheel, here a toe-in supports the steering. However, this then leads to oversteer – when cornering, weight is shifted to the outer wheel, this steers further outwards through the toe-in, which increases the steering movement and brings even more weight to the outer wheel.

  • The driving force has an effect on the track, For example, a slight toe-in for driven rear wheels or a slight toe-in for driven front wheels can ensure that the wheels stay parallel under load.

With velomobiles, on the other hand, it is really only the rolling resistance that is interesting and here it depends on how the track is under load and in motion:

  • On some velomobiles the toe changes due to bump steer see: What is “Bump Steer”?.

  • The wheels have a camber (see: Why are the front wheels at an angle? Doesn’t that slow you down?), which corresponds to toe-in/toe-out perpendicular to the direction of travel. So if the velomobile tilts around the transverse axis, for example because of a change in tyre size or suspension, the camber becomes a slight toe-in or toe-out, which has to be corrected by means of toe adjustment.

For example, it may be necessary to set a slight toe-in when the vehicle is stationary so that the wheels roll optimally. Therefore, always take measurements with the driver and the load, and only after you have rolled a bit.

How do you know if tracking is set correctly?

If the front tyres after a few 100 km are worn out, the track was set incorrectly.

But the track can be measured relatively easily. Two approaches have proven their worth:

  • Measuring the position of the wheels: There are various methods, for example, place two slats on the inside of the wheels and measure their distance at the front and rear, or attach a laser pointer to the hubs. It is advisable to sit in the velomobile during the measurement and have it read by a second person so that the chassis is loaded just as it is while driving. The wheels must be as parallel as possible.

  • Roll test: You roll down with the Velomobile and different settings of the tie rod and measure how far you can get. It is important here that the starting point is always identical and the surface is as smooth as possible (e.g. without potholes) and straight (without curves). The route does not have to be long, for example, you can roll off a ramp that is only a few centimeters high in an underground garage or cellar.

In a roll test, the distances of the best settings are quite close, the worse the track is set, the more the rolled distances deviate from each other.

What is “Bump Steer”?

The suspension reacts to uneven surfaces, i.e. the front wheel moves up and down relative to the vehicle, and with it the wishbone. This is attached to the vehicle on the other side, i.e. the end of the wishbone attached to the front wheel scribes a circular path. If it is not nearly horizontal and/or is not very long, then this movement also has a noteworthy lateral component – i.e. the front wheel is pulled closer to the vehicle or pushed away. And there the control arm does not necessarily run in the direction of the virtual pivot point, which leads to a slight steering movement.

Ideally however, the wishbone is as long as possible and is as horizontal as possible in the loaded condition. In addition, the tie rod is ideally divided so that the parts are as long as the wishbones and are as horizontal as possible. A steering movement can thus be prevented during compression and rebound.

How should the suspension be adjusted?

  • When the wheel wells are open, the shock absorber should be inclined so that the wheel is flush with the edge of the wheel arch.

  • The scrub radius should be slightly negative (see: What’s the deal with the scrub radius?).

  • The tracking should be neutral, i.e. the wheels are parallel (apart from the camber, see Toe-in, toe-out – which is correct?).

  • The wishbones are as long as possible and as horizontal as possible when the vehicle is loaded (see: What is “Bump Steer”? ).

  • The Ackermann condition is fulfilled (see: What is the Ackermann condition? ).

  • The steering is ideally set in such a way that it is insensitive near the neutral position, because even small steering movements have a big impact at high speeds. In contrast, the steering can react more in the event of full steering locks, since these only occur when driving slowly.

Why is good straight-line stability not sufficient when it is windy?

Some drivers are surprised that their vehicle, which is normally very well behaved and runs like on rails even at high speeds, is noticeably deflected by gusts of wind. The reason for this is:

So basically they are two different things, a track-stable vehicle is not necessarily insensitive to gusts of wind, and a wind-insensitive vehicle does not necessarily have good straight-line stability.

Wheels and tyres

Which wheel size?

The front wheels are simple: 406 mm (according to ETRTO; corresponds to 20 inches). All other sizes of small wheels are much more unusual, and the choice of tyres and rims is correspondingly small. Only 16 inches (349 mm and 305 mm) are still reasonably common, but would not bring any real advantages. A few hobbyists have built smaller wheels for their 20-inch velomobiles in order to fit wider tyres (e.g. winter tyres to accommodate spikes), but this is only necessary if the space in the wheel well is limited.

With the rear wheel you have more choice, with many velomobiles possible sizes are:

  • 559 mm (26 inch, mountain bike)

  • 571 mm (27 inches, triathlon)

  • 584 mm (27.5 inches)

  • 622 mm (28 inch, racing bike)

The influence on development is rather small and can be matched by using suitable chainrings. Therefore, the most important criterion is the size of the best tyres. The fastest tyres are basically for racing bikes (i.e. 622 mm), but they are quite narrow (and wide ones also often do not fit in the wheel well). For mountain bikes there are also a lot of tyres, but almost only wide and with studs. Narrow, fast tyres in 559 mm are less and less common, so the best compromise (fast, not too narrow tyres) could be 584 mm.

Why use such wide rims in a velomobile?

The rim width depends on the tyre width:

  • Too narrow: the tyre does not hold properly on the rim because it is not vertical to the rim flange (i.e. orthogonal to the pressure direction), but spreads outwards. In addition, the tread is far from the rim, i.e. side forces pull a larger lever force on the rim flange.

  • Too wide: the tread is close to the rim, which means that the tyre cannot deflect sideways very much and will run faster. If the tube is squeezed, it gets the classic snakebite holes.

Snakebites are not a big problem in the Velomobile because the weight is distributed over an additional wheel. In addition, you typically drive on smooth roads and not over curbs and stones. On the other hand, lateral forces in curves are a big problem – since you don’t lean in the curve, the forces do not always only act perpendicular to the tread, but often also from the side. Therefore one tends to use wide rims. With open wheel wells, they also have the advantage that the tyre does not protrude as far over the rim, i.e. the profile is smoother and therefore more aerodynamic.

What tyre width?

The tyre width is limited by the size of the wheel wells and even if a wide tyre would basically fit in on the front wheels, that often reduces steering angle, i.e. the circumference becomes significantly larger.

Otherwise, the weight is distributed over at least three tyres, i.e. a single wheel has to withstand less load than with a conventional bicycle. So there is nothing to be said against very narrow tyres, which are more prone to snakebites. But since the tyres run inside the wheel wells, a narrow tyre is not necessary good for aerodynamic reasons.

Ultimately, you want tyres that are as fast as possible, and that seems to be the best with a width of approx. 28 mm. But that also depends on the condition of the road – on bad roads, wide tyres that have higher rolling resistance than the narrow racing tyres, can still be faster because there are fewer vibrations.

What tyre pressure?

In principle, the minimum and maximum pressures specified by the manufacturer are useful points of reference. But it also depends on the following things:

  • Suspension travel of the tyre: A wider tyre is also taller, i.e. has more suspension travel. To avoid a snakebite, a narrow tyre must be inflated harder so that it does not sag despite low spring travel.

  • Tyre spring stiffness: A wide tyre will bounce less than a narrow tyre at the same pressure and load, i.e. to achieve the same spring travel as a narrow tyre, a wide tyre must be ridden with less pressure. (It goes with this that a wider tyre can also withstand less pressure, see: What about Barlow’s formula?)

  • Load: The higher the system weight, the higher the pressure must be so that the tyre deflects the same as with a lower load.

  • Number of wheels: A velomobile is heavier than other bicycles, but it also has more wheels. The load per wheel is therefore lower and the tyre deforms less – the pressure can therefore be lower.

  • Lateral forces: Since a multi-track does not tilt into a curve, the forces do not only act radially, but there are also strong lateral forces. If the pressure is too low, the tyre can deform strongly sideways, which in the long run damages the tyre or in extreme cases pulls the tyre off the rim. Therefore, even with little load and wide tyres, the pressure must not be too low to keep the tyre securely on the rim.

How high should the pressure be then?

  • The higher the pressure, the less the tyre deforms, this flexing work costs energy, i.e. high pressure reduces rolling resistance. However, this effect decreases with increasing pressure, if a hard tyre hardly deforms at all, then even more pressure can only reduce the deformation very little further (see Fig. 12).

  • If the tyre does not have any spring at all, then not only the tyre is deflected in the event of bumps, but the entire vehicle with the rider. This of course costs much more energy, because this acceleration can hardly be converted back into propulsion and is lost (see: Do you need suspension? and Doesn’t suspension cost valuable energy?). This speaks for a lower pressure.

For the lowest possible rolling resistance, a high pressure is therefore generally recommended, but low enough so that the tyre’s suspension travel swallows the high-frequency bumps of a rough road, where the suspension does not yet respond, instead of jolting the vehicle and driver. Here, a flexible tyre (i.e. thin-walled, soft rubber compound, thin tube or latex tube or tubeless) is at an advantage because it not only causes less loss when deformed, but also adapts better to the bumps or a lower air pressure costs less extra power (see: Fig. 22).

Rolling resistance of road bike tyres at different pressures

Fig. 12 Histogram of the rolling resistance of road bike tyres at different pressures, measured on a roller dynamometer. The median value drops from 19.6 W at 4.1 bar to 14.2 W at 8.3 bar, as can be seen, an increase to 5.5 bar improves the median by a full 3 W, but further increases improve it less and less.

How good are latex tubes?

Latex tubes are more flexible than butyl tubes and therefore reduce rolling resistance (see: How do you reduce rolling resistance? and Why are good tyres so important on a velomobile?). According to a measurement, the rolling resistance of fast tyres (Continental GP 5000) is 20% lower than that of thin butyl tubes. With slow tyres the effect should be lower, with low temperatures higher.

On the other hand, latex tubes are not as tight, you have to pump them up more often than with butyl tubes. In addition, latex tubes stretch more at certain points and become much thinner at these points. Therefore, latex tubes have to fit the tyre size exactly, you have to be more careful when fitting them than with butyl tubes (the tubes have to fit absolutely evenly), and small damages in the casing are more critical because latex tubes already swell outwards through small holes in the carcass and then burst. On the other hand, small punctures are less of a problem, at low pressure they close up again and it takes longer for the tyre to go completely flat. However, it is correspondingly difficult to find and patch such tiny punctures.

What is tubeless?

Normally, a tire only provides grip on the road and absorbs the resulting forces with its carcass; airtightness is ensured by a tube inside. With a tubeless wheel you do without the tube, as the tire is made airtight and also sits particularly firmly on the rim. This has a flat rim base and often also Humps to fix the tire better; In addition, a special rim tape seals the rim base airtight. The valve is screwed directly into the rim. A sealing milk is usually also added to the tires to seal small leaks. This brings the following advantages:

  • By doing without the tube, rolling resistance is reduced and driving comfort is increased (compared to butyl).

  • Compared to latex tubes, rolling resistance is similar or slightly lower.

  • You usually have to pump air less often than with latex tubes.

  • In addition, lower pressures are possible because no Snakebite can occur, which is why tubeless first became popular in the mountain bike sector.

  • In contrast to tubes, a tire does not burst suddenly, but the air loss occurs more slowly and is often stopped by the sealing milk, so that the tire often does not go completely flat, but retains a small residual pressure. This increases security.

But repair is more difficult:

  • Small holes can often be sealed with the sealant. However, this often only works up to a certain pressure - until a proper repair is carried out, you have to drive with reduced pressure, but at least you can make progress.

  • Slightly larger holes can be partially sealed with a tubeless repair kit by stuffing rubber strips into the hole with a special tool. The sealant does the rest.

  • For bigger punctures you have to remove the tire. This is more difficult because a tubeless tire sits tighter on the rim. Inside, you first have to clean the tire from the sealing milk before you can stick on a patch or insert a tube.

  • In the event of a total loss of air, you need a compressor or pressure tank to inflate the tire sufficiently quickly until it is tight against the rim. Otherwise you have to use a tube.

What about Barlow’s formula?

It’s actually quite simple: pressure is force per surface, i.e. if the inner surface of the tyre is larger, a higher force acts on the tyre at the same pressure.

If a tyre is twice as wide, then it is approximately twice as high, i.e. the circumference of the tyre cross-section is also twice as large. Since the circumference of the wheel is almost the same, the inside of the tyre has twice the area – so double the tensile force acts on the carcass of the tyre.

That is why wide tyres can withstand much less pressure than narrow tyres. But wide tyres also need less pressure because, firstly, the tyre contact patch is wider – therefore, with the same load and pressure, the contact patch is shorter, i.e. you need less pressure so that the tyre contact patch is as short as a narrow tyre. Secondly, a wide tyre is taller, i.e. it has a lot more travel until there is a puncture (“snakebite”).

Of course, you could build a wide tyre correspondingly more solid so that it can withstand more pressure. But this also makes it stiffer, i.e. the rolling resistance increases – which you actually want to reduce with higher pressure. But high pressure does not only make the tyre, but also the rim load greater. Lightweight rims in particular can only withstand a certain pressure that is specified by the manufacturer, so if you fit a wider tyre, you have to consider not only its maximum pressure, but also that of the rim (which is always related to a specific tyre width) – corrected using the Barlow formula. If the tyre is twice as wide, you can only load the rim with half of its maximum pressure.

Suspension

Do you need suspension?

Normally yes. The main characteristic of a velomobile is its speed. And high speed means that driving over an obstacle and the resulting change in speed happens in a shorter time, i.e. the acceleration is higher. And acceleration times mass is power, the former is constant, i.e. the vehicle and driver have to endure stronger forces at higher acceleration. In addition, at high speeds there is less time to correct a change of direction due to an obstacle. Therefore, it makes sense to first reduce the force peaks that affect the vehicle and driver with the help of suspension, and also to make the vehicle more controllable.

Another effect is added to the front wheels: since the wheels are side by side, but rarely the bumps, the vehicle tilts around the longitudinal axis. The driver is not only accelerated up and down, but also thrown from side to side, which can be very uncomfortable.

For the rear wheel, suspension is important for another reason: if an undamped rear wheel encounters bumps with a suitable frequency, it can jump so much that the vehicle overturns sideways (see: Why is it dangerous when the rear wheel slides?),

Doesn’t suspension cost valuable energy?

Yes, of course. Or rather: not the suspension, but the associated damping is nothing other than a dissipation of energy. But what would happen to the energy otherwise?

  • The driver has laboriously built up kinetic energy, and bumps in the road ensure that a small part of the kinetic energy is directed in a different direction – for example, from forward to upward, the vehicle is braked and lifted. But only rarely is there a suitable counter-evenness afterwards, which accelerates the velomobile, which is falling down again, forwards like a small ramp. So most of this energy is lost anyway, and so it is better to destroy it specifically in the damping element instead of putting a load on other parts of the vehicle.

  • With suspension, ideally only the unsprung part of the vehicle is accelerated on bumps, and only its kinetic energy is lost. In contrast, without suspension, the entire vehicle with driver has to react to the bump, so a much larger mass has to be accelerated – so more energy is lost. A well-responsive suspension therefore makes you faster.

  • A suspension costs weight, but an unsprung, fast vehicle has to be built quite a bit more massive to be able to withstand the higher force peaks.

  • A suspended vehicle also offers better control through more grip (and thus higher speed becomes possible on bad sections).

  • Without suspension, the driver (who, after all, makes up the largest part of the total mass) is also accelerated, and he also does not return the energy 1:1 to the road via the vehicle, but dampens it away via his muscles and tendons. And that is not only unpleasant, but also strains the muscles – i.e. costs energy that should rather be invested in propulsion.

And then there is the fear that a suspension system bobs along with the cadence and eats up part of the pedalling power. Of course, this can happen in principle, but it is negligible in a well-tuned vehicle. Firstly, the drive can be designed in such a way that the chain pull is almost perpendicular to the direction of the suspension, and the chain passes almost through the pivot point of the swingarm joint, so that the force component acting on the suspension is very small. This cannot be completely avoided because the sprocket used or even the load changes the chain line slightly, and also the leverage ratios are different in each gear. Secondly, the rider is quite fixed in a recumbent, and the cente of gravity does not change as much as with out of the saddle on an upright bike. Accordingly, less acceleration occurs, which is then eaten up by the damping.

What does a good suspension have to do?

Travel: You want to go fast with a velomobile and have little ground clearance. Accordingly, you drive mainly on the road, where there are no huge obstacles to overcome, but at worst potholes. Accordingly, you don’t need a lot of travel. A large suspension travel would even be disadvantageous because, for example, the wheel wells would have to be correspondingly more voluminous, which ultimately worsens the external dimensions and the aerodynamics. In addition, the negative spring travel must be larger, i.e. the vehicle should be higher, which increases the tendency to tip over in bends.

Responsiveness: At high speeds, the accelerations are high, so good suspension must respond very quickly and even with small forces. This affects above all damping. In many damping mechanisms sliding friction occurs, which is unfavorable, because static friction must first be overcome before sliding friction. Therefore, friction dampers or poorly constructed piston mechanisms with oil dampers are unfavorable, while spring-damper systems that are based on bending (e.g. leaf springs), torsion (e.g. torsion element in the joint) or compression (e.g. elastomers) do not have this problem. (they do have other disadvantages, such as temperature dependency, or lack of adjustability of the suspension/damping characteristics.)

Strength of damping: This must be neither too strong nor too weak, as can be seen in Fig. 13, an oscillation decays fastest when the damping corresponds to the aperiodic limit case. So both the spring stiffness must be matched to the mass to achieve the correct rest position, and the damping must be matched to the resonant frequency resulting from spring stiffness and mass. This allows the velomobile to compensate for uneven ground as well as possible and then come to rest as quickly as possible.

damped harmonic oscillator

Fig. 13 Damped harmonic oscillator, with varying degrees of damping. In all cases the oscillation decays exponentially. If the damping is too low (= oscillation fall), the equilibrium position is reached fastest, but overshooting occurs. With too much damping (= creep case) there is no overshooting, but the equilibrium position is reached only slowly. In the aperiodic limit case (= critical damping) the amplitude decays fastest because it is the lowest damping without overshooting.

Couldn’t the suspension be replaced by low pressure wide tyres?

Wide tyres with little pressure can be surprisingly comfortable on rough ground. But they cannot replace real suspension:

  • Travel: To offer several centimetres of travel, a tyre would have to be extremely bulky – and thus hardly lighter than a suspension, with less lateral support and a larger turning circle.

  • Damping: A suspension must also dampen (see: What does a good suspension have to do?) – but a good tyre should have no damping at all, because this would eat up energy not only during bumps, but during the deformation that occurs in every single rotation. (That’s why you can judge a tyre’s rolling resistance by how high it bounces back when you drop it).

  • Adjustability: it’s much easier to adjust a suspension than a tyre – it’s relatively easy to change springs or adjust damping without affecting travel, whereas with a tyre you can only change air pressure, the shape and rubber compound is fixed.

In the end, it is better if a well-tuned suspension compensates for and dampens away rough bumps, and high-frequency vibrations (with low amplitude) are swallowed by a supple tyre.

Why is it dangerous when the rear wheel slides?

It’s a combination of several things:

  • The front wheels do not easily lose cornering grip, because they are rarely in the air at the same time and rarely is one suddenly in the air. It’s different with the rear wheel, if it jumps over bumps undamped or if the tyre bursts, the cornering is suddenly sharply reduced.

  • Since the centre of gravity is much closer to the front wheels than to the rear wheel, there is little weight on the rear wheel. Correspondingly, the cornering force is normally significantly lower than that of the front wheels, and it is therefore easily critical if it is reduced again.

  • When turning, not only the load plays a role, but also the position relative to the cente of gravity: the front wheels are almost next to the cente of gravity – they roll tangentially to a rotation about the vertical axis and there is practically no resistance. The rear wheel, however, rolls radially, it opposes the tangential movement with high resistance. If the centre of gravity is further back, the direction of the cornering force is more favourable for the front wheels (they no longer roll tangentially to it), and the rear wheel has more load (see Fig. 14).

  • With a velomobile it’s steering angle is quite small, because it is limited by the width of the wheel wells. So if the rear wheel loses lateral grip and breaks out to the side, there are only limited countermeasures available.

  • You often drive fast with a velomobile. The reaction time is correspondingly short in order to be able to react to a rear wheel breaking out.

If for some reason the rear wheel no longer has cornering power – regardless of whether it jumps or blocks, or a burst tyre no longer sits on the rim – and there is also lateral force – because you are driving through a curve or it gets a side blow from a pothole – then it breaks out. And if the velomobile turns faster than you can counteract it, it can turn sideways and roll over. This requires surprisingly little, one example is a velomobile with an unsprung rear wheel that hits the vibrating strip in a straight line with the rear wheel and is tipped over in just a few seconds at full speed.

What can I do about it?

  • Luggage to the rear: firstly increase the load on the rear wheel and secondly moves the cente of gravity further away from the front wheels.

  • Check tyres: tyres rarely burst unannounced, there is usually visible pre-damage.

  • Use reliable tyres with a sticky rubber compound.

  • Tubeless rim: Makes it more difficult for the tyre to come off if it blows out.

  • Wide rim: Makes the tyre less tall, which means that as long as it doesn’t bounce off, it bounces less.

Direction of the forces applied to the wheels

Fig. 14 Direction of the forces on the wheels as the velomobile tilts around its vertical axis, with different centes of gravity. Since the rear wheel is located behind the cente of gravity, the forces are always applied perpendicularly to it when the velomobile tilts, tracking is thus good. Green: If the cente of gravity is further back (here: 70% of the distance between the front and rear wheels), then there is a lot of load on the rear wheel (here: 70%), and in addition, the forces on the front wheels are almost perpendicular to them. Red: If the cente of gravity is further forward (here: 10% of the distance), then the forces are almost parallel to the front wheels – so they can’t provide much lateral support. The rear wheel can, but only 10% of the load is on it. In reality, the centre of gravity is usually located between these extremes, somewhere near the orange point, i.e. about one third of the distance between the front wheels and the rear wheel, and accordingly load distribution on the wheels is relatively balanced.

Gears

What role do gears have to fulfill?

The challenge with velomobiles is that on the one hand you drive at a very constant speed – the gearing there must be very finely graded – but on the other hand you cover a much wider speed range than with all other bicycles, and you have a heavy vehicle which, uphill, must still be pedalled at a reasonable cadence without going out of the saddle.

A fine gradation could also be achieved with a half-step gradation of the chainrings. However, it is very impractical to need two gear changes for each gear at high speeds, instead of simply shifting on a pinion.

Is a hub gear possible?

Rather not – hardly any IGH has a sufficiently large range of at least 500%. Only the Rohloff Speedhub and the Pinion gearboxes are in that area. In addition, IGHs inherently have constant or symmetrical gear changes, a flexible gradation is only possible with derailleur gears.

But what is possible and popular: combinations of derailleur and IGH. First, there are combined derailleur and hub gears, here the hub gear extends the gear ratio range, and also enables gear changs to be made quickly at traffic lights, for example, and the combination of both gearshifts to mitigate rough gear changes through intermediate gears. Unfortunately, such a hub requires a two-sided swing arm. Second, there is the Schlumpf Mountain Drive, although these only offer two gears and are less quick to shift with the heel than with a gear lever, they offer an enormous gear jump, with which you can significantly expand the range of the derailleur and do without the second chainring. Third, you can with a velomobiles install an additional gearbox in the drive train, which can also be a hub gear. The disadvantage of all IGH systems is the high weight.

Mountain bike or road bike cassette?

Neither nor. You cannot buy off-the-shelf an optimal cassette. Cassettes are no longer made from individual sprockets, but are riveted together in groups. And for racing cyclists there are finely graduated cassettes with unfortunately too little gear range and for mountain bikers there is enough range, but incorrectly spaced.

Therefore, some velomobile drivers facing this problem assemble the cassette themselves – either from a mountain bike cassette, which you supplement with individual sprockets or entire sprocket groups from other cassettes, or from special sets of individual sprockets. However, this rarely works without tinkering, for example, spacers have to be adjusted or protruding parts have to be ground off. For optimal gear range, it is often even necessary to drill out the rivets from riveted cassettes and rivet them back together with new sprockets. Since the pinion wear on a velomobile is low thanks to little dirt, such a cassette will last for many years.

Why can’t you change through all the gears?

Even a mountain bike derailleur with a long chain tensioner has a capacity of only 40 teeth, if the chainrings and sprockets have a larger tooth difference (i.e. teeth of the largest chainring + the largest sprocket minus the teeth of the smallest chainring + the smallest sprocket), then the chain can no longer be tensioned and sags. And an extended chain tensioner on the rear derailleur is rarely possible because the space within the bodywork is too small. In practice, this means that only the largest sprockets can be used on the small chainring. But you only lose “double” gears, and since these are mostly the small pinions, which are ideally spaced anyway, it doesn’t matter.

Which chainring or cassette sizes work?

The smallest sprockets are freewheel body specific, smaller than 11 teeth does not fit on a normal Shimano freewheel body. If you have a large rear wheel, you need a chainring with approx. 60 teeth in order to be able to comfortably pedal up to approx. 60 km / h. And you want to be able to pedal uphill at 5 km/h, that calls for a gear ratio of approx. 1:1. Because with well spaced cassettes the sprocket sizes only go up to approx. 40 teeth, and with a bolt circle of 130 mm the smallest chainring must have 38 teeth, there is much to be said for compact cranksets, i.e. with 110 mm bolt circle, where you can also fit smaller chainrings. The huge jump that occurs between the small and large chainring is customary, front derailleurs are actually still changeable if they are carefully aligned (i.e. both the radial distance and the angle).

Cassette spacing?

Most of the time you drive on the flat, at speeds between 40 and 50 km/h. Accordingly, the sprockets should be very finely spaced – a tooth difference is a much larger percentage difference with small sprocket sizes than with large sprockets, and at high speeds and thus pedaling power, this also corresponds to a big leap in performance that has to be mastered. You should ignore speeds higher than approx. 65 km/h. They occur relatively often, but only downhill. No normal person can pedal at this speed for a long time on the flat. And such speeds are usually over after a few seconds, you don’t need an extra gear for that, you just let it roll a bit. Steep mountain climbs are not more common than steep descents – but it doesn’t take seconds, but often many minutes to climb. A suitable gear is worth it here much sooner. So the bottom line for the cassette is: the small sprockets finely spaced, towards the large sprockets always wider spacing.

Big or small chainring/sprocket?

At the same speed and cadence:

  • With a small chainring and small sprocket, the chain speed is low, but the chain tension high. This means: high losses on the traction run (with chain deflection), and a strong tensile load/deformation of the drive.

  • With a large chainring and large sprocket, the chain tension is low and the chain speed is high. This means: higher losses where there is little traction, for example on the empty strand.

In total, it is therefore advisable to drive on large chainrings – even if it feels as if the idle is less with smaller chainrings. That’s true, but it’s the other way around under load.

Small chainring or large largest sprocket?

This is actually a matter of taste, the latter, however, has the attraction that you can then drive a lot with the large chainring and rarely need the front derailleur. The efficiency is also somewhat higher there. However, getting a well spaced cassette is more difficult if the largest sprocket is to be particularly large (over 34 teeth).

Drivetrain

How big are the losses from chain idlers?

The following things happen on an idler:

  • The chain links are angled. The smaller the idler pulley, the stronger it needs to be – because the chain links are distributed over fewer teeth.

  • The idler loads the bearing. If, for example, the chain is deflected by 90 °, it presses against the idler pulley at a 45 ° angle.

The forces that occur are not negligible: with a deflection of, for example, 90 °, a force of sine(45°) = approx. 70% of the chain pull acting on the idler. Crank and chainring size can easily double the pedal force. That can then correspond to a weight of 100 kg.

This is not so bad when the bearing is loaded: a ball bearing has very little friction, i.e. the losses are low despite the high force. It is different with the force on the chain links. The deflection per chain link is much less, the 90 ° deflection is distributed, for example, to 10 links on a large idler, and the force is correspondingly low (here, for example, sine(9°) = 16% of the chain pull). But chain links are not ball bearings, but plain bearings that have much higher friction – especially if they are also dirty.

Therefore, the most important measure is to keep the losses from idlers small: choose large idler pulleys!

One or two idlers?

Good question. Basically you can get the chain to the rear with an idler under the seat. But if it is also supposed to be large so that the losses are low, it protrudes further down, and the chain runs obliquely at the front and back. With two idlers you can lower the seat (i.e. lower cente of gravity ), but still have more ground clearance, and the chain can come off the chainring more steeply so that you do not touch it with your calves. In total, two idlers are slightly less efficient, but offer these other advantages.

Are ceramic bearings worth it?

Ball bearings made of ceramic are very wear-resistant and do not need any lubricant. Therefore, the stiff ball bearing grease and seals can be dispensed with, which means that the bearings have lower losses. However, even normal steel ball bearings have very low losses, in terms of magnitude, less than one watt for all bearings in the drive train and wheels combined. That is why ceramic bearings can only save a fraction of a watt at best – that is hardly measurable, and certainly not noticeable.

Are chain tubes inefficient?

A chain that grinds in a pipe naturally causes friction. And if the chain is oily and there is a lot of dirt stuck to it, it also settles in the chain tube, the inside diameter becomes smaller and the chain grinds more often. Wet dirt cannot dry and fall off as easily, but the wet dirt stays in the tubes forever.

But: in a velomobile, the chain is fairly clean because it has no direct contact with street dirt. Only the dirt that the driver brings in, for example through dirty shoes, can land on the chain. And there a chain tube is rather positive because it protects the chain from dirt. It is always damp in a velomobile and dirt collects at the lowest point, i.e. happily below in the chain channel. It would not be a good thing if the chain were permanently pulled through this damp dirt, but at least there tubes can help against the pollution.

The laying of the tubes is also important. Frictional force is the product of contact pressure and sliding friction coefficient, where the chain is under tension, it does not follow the course of the tube, but is pressed against the tube, the friction is correspondingly high. It is therefore important in the tension strand that the chain does not follow the tube, but vice versa the tube follows the chain. If the chain does not touch the tube under tension, there is no friction.

It is different in the empty strand. The chain sags there – in order to be frictionless, the chain tube would have to follow the course of the sagging chain. But the contact pressure there is also low (namely only approximately the weight of the chain), i.e. the friction is low. And in the Velomobile there is no space anyway for the chain to sag freely – whether it is dragging in the chain tube or on the floor doesn’t make much difference. In the first case, it is protected from dirt.

Alternative drive concepts?

Alternative drive concepts are discussed again and again, but so far none have been able to prevail over the conventional combination of pedal crank + bicycle chain - mostly because their disadvantages outweigh the advantages:

  • Belt Drive: A belt normally cannot be reversed, making it impossible to route backwards under the seat. In addition, it is not compatible with a derailleur (which is superior to the vast majority of gearboxes because of its range of gear ratios), cannot be opened (which prohibits all closed bushings), and the efficiency is somewhat lower.

  • Linear drive: A pedal drive without cranks would have geometric advantages, because the front of the velomobile does not have to accommodate the extensive pedal circuit - especially for riders with long legs and correspondingly long cranks and large feet, the front could be flatter and the angle of departure steeper. But the ergonomics are worse because the legs don’t describe a circular movement, but have to come to a standstill at the dead points. This prevents high cadences.

  • Cardan shaft: A high torque must be transmitted on a bicycle. While a chain is only subjected to tensile loads, a cardan shaft has to withstand torsional forces - and in order to be sufficiently stiff, it has to be very thick and solid. In addition, the two bevel gears at the ends cause additional losses.

  • Electric drive: With a combination of pedal generator and electric motor, you can design the power transmission geometrics completely freely and with the appropriate electronics you can even do without a mechanical gear shift. However, the double conversion between mechanical and electrical power generates high losses.

  • Hand crank drive: The leg muscles are significantly stronger than the arm muscles. Therefore, a drive with only the arms does not make much sense. If, on the other hand, you use both arms and legs for propulsion, you have significantly more power at your disposal. However, pure muscle power is only decisive for short sprints, over time, the limiting factor is the supply of oxygen and nutrients to the muscles – i.e. circulation, lung function and digestion. Therefore, a hand crank drive only adds a lot of additional mechanics and thus weight, but that does not make it faster.

Why a stiff drivetrain?

Whenever you pedal hard, a stiff velomobile is important, because then the pedaling force ends up on the road instead of in the deformation of the vehicle. This deformation is of course elastic, i.e. as if you were tensioning a spring. However, such a deformation naturally has hysteresis, that means you don’t get back all the energy you put into it.

But that’s not even the main problem: when you pedal on a stiff bike, the pedaling energy is immediately converted into speed – if you stop pedaling afterwards, the bike is faster and the pressure on the pedals is in the correct path. If, on the other hand, the frame is deformed by the pedaling energy, it is then under tension – i.e. it pulls on the chain and accelerates the bike, but it also presses on the pedals and thus on the muscles. But they don’t like that at all. For example, when you lift a weight, the muscles do lifting work, but if you keep a weight at the same height, physically no work is done, but it still feels tiring, and the muscles are consuming nutrients. Muscles therefore consume energy when they only have to exert a static counterforce – a screw in the ceiling, on the other hand, can carry the weight of a lamp for centuries without tiring. And that is why this counteraction against an elastically deformed drive train costs force that is then missing for acceleration..

What needs to be stiff?

The forces that occur are primarily the chain pull and the pressure of the driver’s back in the seat shell. Everything that goes with it has to be rigid:

  • The bottom bracket is attached to the bottom bracket mast.

  • This is above the steering bridge with the front wheel connected to the wheel wells. There are many curved surfaces and relatively thick material here, which means that there is enough rigidity.

  • The chain movement pulls the idlers up, the front one is therefore usually connected to the stiff steering bridge.

  • The chain idler also wants to pull the vehicle floor between the two idlers together like a bowstring. That is why in many velomobiles the chain tunnel also has a static function, because it serves as a U-profile that stiffens the vehicle floor in the longitudinal direction.

  • The chain pull goes on to the rear wheel, whose swing arm must be correspondingly rigid and also must be well supported by the bodywork.

  • Also the seat must be securely attached. For this reason, for example, some seats are bolted to the front of the front wheel wells and are supported at the rear against the rear wheel arch, which often holds the swing arm bearing in the front and is accordingly strong.

  • The body between the rear swing arm bearing and front wheel housings must also be sufficiently rigid. But this is basically the case because it is a tube with a large diameter. However, openings that are too large are disadvantageous, the velomobiles, which have huge entry openings, are correspondingly massive and heavy. (There were also convertible cars, for example, which had to be reinforced in comparison to their closed coupé counterparts, because otherwise they would not have been stiff enough.)

The rest is not so important and can be made thin-walled accordingly.

It also depends on the direction of the deformation: It is bad if the deformation creates a counter-pressure to leg strength - muscles are strained without the power going into the drive. On the other hand, it is not so bad if slight torsions occur in the bottom bracket mast - the force component in the pedaling direction is only small. Therefore, the goal is not rigidity at any price, but a good compromise between lightweight construction and rigidity in the pedaling direction.

Brakes

What are the requirements for a velomobile brake?

In any case, not particularly high braking forces. Firstly, the Velomobile has two front wheels that can be easily braked, secondly, the front wheels are small, i.e. the lever between the radius of the brakes and the radius of the wheels is smaller than with large wheels, and thirdly, you drive more smoothly, not downhill like on a mountain bike, on a surface which has surprises in store. In practice, the limiting factor is the tyres, a well adjusted brake can usually be blocked.

To do this, the brake must be able to withstand high heat output (see Fig. 15). Firstly, there is little drag on the Velomobile that slows it, while a mountain bike hardly ever sees speeds faster than 60 km/h, a velomobile easily reaches over 100 km/h. Second, with the Velomobile, the usual speeds are higher, with the same gradient, a higher speed means that more altitude energy has to be braked away per time. And thirdly, a velomobile is heavier and the brakes in the wheel wells are less well cooled.

So the problem is not short braking maneuvers, but long descents. Larger drum brakes help here, simply because they are larger their heat disapation capacity is higher – there is more material that can be heated, and the surface area of the hub shell is also slightly larger.

And what happens if the brakes are too hot? First, they fade i.e. the braking effect decreases noticeably. Second, the hub shell can expand so far that the spokes lose their tension and get loose when the wheel rolls, and soon break through material fatigue due to the alternating load.

Braking power required to maintain speed

Fig. 15 Braking power needed to maintain speed on various gradients: the solid curves show a velomobile, the dashed ones a racing bike. While on the latter, drag increases so much that, at higher speeds, the necessary braking power decreases again, and on a gradient below 5%, you don’t have to brake at all to keep from going over 50 km/h, on the velomobile, this maximum power required is much higher and at much higher speeds. While a road bike brake does not have to withstand more than 500 W on a 10% incline, with a velomobile it is almost twice as much at the same speed and at its highest at three times the speed, almost four times as much is needed.

How can I cool the brakes or make them less hot?

One way is to direct more of the airstream onto the brakes. This was done, for example, with the Velomobile that Marcel Graber used to do the Trans America Bike Race 2018 – he used wheel well covers (“pants”) and, correspondingly, front wheels without a wheel disc, and air channels were cut into the wheel well linings pointing towards the hub.

Another option is water cooling, which was implemented by Patrick Flé (Velomobilize): with a plastic bottle and a hose, water is sprayed into the brake drum, the water evaporates immediately because of the heat and the vaporization means a very high amount of energy can be dissipated. The brakes can withstand this temperature shock mechanically.

Then there were also experiments with continuous braking. Firstly, with an additional rear brake, owever, this has the disadvantage that the rear wheel locks slightly and then tends to break out. Another method is to use braking parachutes, it has proven useful to use two parachutes, each with a diameter of approx. 40 cm. However, this is only of limited practical use because firstly the parachutes have to be ejected manually at the start of the descent, secondly you have to stop at the end of the descent to repack them, and thirdly the parachutes dancing around are probably not recommended in heavy traffic.

The most practical methods are, however cooling fins and cooling towers of brakes sold by Ginkgo bike parts. These are milled into the brake drums or glued to them. This increases the surface area of the brake drums and air can cool them more effectively.

And last but not least, you can make the brakes heat less with your driving style: the worst thing is to drive downhill at high speed and brake constantly. Intermittent braking means that higher top speeds are achieved so braking drag is higher when the brakes are applied, and then higher temperatures are reached when braking, which makes heat conduction to the air and heat radiation more effective. During the following pause, the brakes can cool down more than if the brakes were applied permanently.

Why does a Velomobile have drum brakes?

Because no other brakes are suitable.

It has no disc brakes because they are attached next to the wheel. Exactly where there is the least space, because the wheel well is insufficient, and a brake would also have to be housed next to the driver’s thighs. That would only be possible if the velomobile were made wider. A brake disc itself is not particularly wide, but additional space is required when turning, which must be available in the wheel well. There are velomobiles with disc brakes, they do not have normal spoked wheels, but disc wheels – these are concave in shape so that the brake disc can be accommodated further inside the wheel. However, these wheels have other disadvantages, such as higher weight, and cannot be centeed. Another argument against disc brakes is the poorer cooling in the wheel well, the brake disc and brake pad are relatively small and therefore heat up quickly. Since the wheels are in wheel wells, the rainwater does not simply run off, but is repeatedly thrown by the wheel into the wheel arch and drips down in there, onto the brakes together with street dirt. An unprotected disc brake would therefore wear out quickly in a wheel well.

There are no rim brakes because that would be difficult to achieve with front wheels suspended on one side, cantilever bosses on both sides are fundamentally not possible, and even a side-pull brake does not fit easily into the wheel well if the wheel must also be able to turn. In addition, a rim brake protrudes outwards, which would be aerodynamically very unfavorable. A rim brake also heats the rim, this would quickly overheat the rim when going downhill due to the high driving speeds and small wheel sizes.

So there are only drum brakes.

Why no brake on the rear wheel?

For two reasons:

  • The rear wheel is relatively far away from the centre of gravity – while the driver sits almost between the front wheels, he sits quite a way from the rear wheel. Therefore, there is little weight on the rear wheel – you can tell by how easily it spins on slippery roads. In addition, the braking deceleration ensures dynamic shifting of the wheel load to the front, which further relieves the load on the rear wheel. You could only transfer small braking forces with the rear wheel before it slips.

  • A slipping rear wheel is dangerous because it slides equally well in all directions – while a rolling wheel moves much easier in the direction of roll than across it. That means a slipping wheel doesn’t have opposing lateral grip anymore and lateral forces accordingly, which can be very dangerous (see: Why is it dangerous when the rear wheel slides? ).

What is this about shims for drum brakes?

The Sturmey Archer drum brakes consist of two brake pads, which are attached to a common bearing on one side and are pressed apart by a rotating cam on the other side. With new brake pads, this cam only has to rotate a little until the brake pads touch the brake drum. Because of the small actuation angle, the rotating cam mostly pushes the pads apart and hardly ever moves sideways.

It is different with worn brake pads. The braking force remains approximately the same since the brake cable has been adjusted accordingly. However, the cam now has to rotate much more and also moves sideways quite a bit. When the brake lever is released, it is reset by means of a spring that pulls the brake pads back together, this spring no longer manages to press the cam hard enough so that it also slides along the side of the brake pad holder – the brake no longer resets to off and locks up.

As a remedy, you can increase the distance between the brake pads and cams so that the worn thickness of the brake pad is compensated for. This is achieved by using a thin metal shim under the sliding shoes on which the brake pad holders slide on the cam (shim thickness e.g. 0.8 mm). Then the cam no longer has to turn so far and accordingly moves less sideways. This is not critical for the brake pads, their thickness is ample even if the cam causes problems. So you can realign the brakes several times, and even then the problem is that the shoe is not holding, but there is still plenty enough brake pad.

On this occasion you should also lubricate the cam and the sliding shoes, it is best to use copper paste because it is heat-resistant.

Why not apply the parking brake after a long downhill ride?

On long descents, a brake gets very hot – including the housing, which is made of aluminium. This becomes soft due to the heat, and if the wheel no longer turns but the brake is permanently applied, it can deform slightly. This is not immediately dangerous – but the brake is then out of round and can no longer be adjusted so well. Therefore, the parking brake should only be applied after the brake has cooled down.

Electrical

Headlights: which ones? Where?

There is never enough light on a velomobile. First, you are often traveling at high speed – possibly up cycle paths that are not made for this and have no painted lines or reflectors down the sides – secondly, you sit low and therefore have a poor view of the street and are often dazzled by car headlights (especially when the left-hand cycle path is lower than the street), and thirdly, the headlights are often mounted quite low – and bicycle headlights are usually designed for a significantly higher mounting position.

While bright lighting is sufficient on a dry road, you will come up against limits on wet roads and in dazzling oncoming traffic with normal bicycle lights. Here at least it helps to mount the headlight as high as possible. But being on the hood is often counterproductive because you blind yourself. Therefore, an internal headlight is recommended, but as high as possible, without ruining the aerodynamics.

Turn signals: which ones? Where?

Since you can’t practically make hand signals, turn signals are highly recommended. These should be far enough apart that it is clearly visible in the dark which side is flashing. Since turn signals are only used very briefly, the power consumption is irrelevant – LEDs, as bright as possible (e.g. 3 W, 700 mA, with largest possible beam angle), and in the front for aerodynamic reasons inserted into the bodywork. Suitable flashing relays are available for motorcycles.

Electronics inside are weatherproof and protected right?

Yes and no. Of course, wind and weather do not hit the inside of the velomobile. However, you sweat a lot more, and in cold weather the air you breathe condenses on the cold bodywork surfaces. The velomobile has a climate like that of a tropical stalactite cave, and the electronics have to be able to withstand this. Anything that is not properly sealed or protected corrodes quickly.

Dynamo yes or no?

A bicycle which is more than a fair weather bike should have a good lighting system powered by a hub dynamo. This is a bit more difficult with velomobiles: on the one hand these are everyday bikes with high mileage – on the other hand there is no good dynamo solution yet. There are high quality hub dynamos for front wheel hubs suspended on one side but these are for trikes that have disc brakes. Velomobiles usually have drum brakes, i.e. the space in the hub is used for those.

It also is difficult on the rear wheel, there are dynamos for a cassette rear hub, however, they have rather poor performance. And they are only available as double-sided hubs. Velomobile drivers with one-sided swing arms abstain!

Then there would also be cross-country dynamos. However, you can only run these on the rim, since the thin flanks of faster tyres would not be able to withstand the stress of a dynamo roller. In addition, you would have to attach the dynamo somewhere in the wheel arch, where there is a lot of water and dirt when it rains – so the dynamo doesn’t have an easy life and would probably slip easily.

And last but not least there is a combination of hub dynamo and drum brake from Sturmey Archer. However, this is not really convincing, because firstly, the efficiency is not great, secondly, the permanent magnets will not tolerate the heat of the brakes for a long time (Curie temperature!), and thirdly, the dismantling of the wheel is nowhere near as easy, because not only one screw has to be loosened, but also the wires have to be removed.

Finally, there is also the problem that a velomobile uses a relatively large amount of electricity. For a headlight, the typical 3 W of a dynamo is absolutely the bottom limit, it can sometimes be triple that. At least one buffer battery is required. And that’s why most velomobile drivers stick to pure battery lighting.

Which voltage for the vehicle electrical system?

It used to be easy, since the dynamo always delivered 6 V at moderate driving speed, with which the incandescent light bulb shone passably, and at high speeds the dynamo made sure that not too much current flowed despite possibly higher voltages. With a light bulb you only have to apply a voltage, the resistance of the filament increases with the temperature, so that the current can only increase until it glows brightly.

It is different with today’s on-board electrical systems. A battery can deliver almost any current, and even a LED no longer has this negative feedback that limits the current, rather the opposite. This is why electronics are needed to regulate current and voltage. Many devices have such a thing built-in and tolerate a wide voltage range, for example from 6 to 16 V. And naked LEDs do not really care about the voltage anyway, it has to reach a certain minimum value, but otherwise the current has to be limited, which is so-called constant current source (CCS).

Accordingly, some people prefer 12v battery systems because they can easily install motorcycle parts. Others use 6v to install bicycle components. As long as all devices can cope with the voltage, and all LEDs are connected to constant current sources and, for example, a USB cable is connected to a 5 V voltage converter, the voltage of the vehicle electrical system does not matter much.

How to treat a lithium-ion battery?

  • Do not overcharge: When the final charge voltage is reached, the charging process must be stopped.

  • Do not undercharge: When the minimum voltage is reached, no more current may be drawn.

  • The drawn current must not be too high (rarely a problem in the velomobile).

  • The charging current must not be too high.

  • The temperature must not be too high.

These things are checked by any standard charger when charging, the charging current is kept constant and, with increasing battery voltage, the charging voltage is increased until the final charging voltage is reached. Small BMS boards are usually integrated in the removable battery (Battery Mnagement System), which switches off the battery if the current is exceeded or the voltage limits are exceeded or undershot.

In addition, batteries with cells connected in series also need balancing, in which all cells are fully charged and evenly if they have previously discharged unevenly. However, this is more relevant for high currents, for example for electric drive motors, with usual use as light batteries, one can usually do without balancing.

To be sure that the battery lasts as long as possible, firstly the charging current should be kept low, and secondly the battery should not be completely drained or fully charged, ideally it is not discharged below 30% and not above 80% to 90% (depending on the source).

Where do I attach the speedometer / GPS / Navigation?

Where you can see it well. If the display is large, this can be on the wheel arch or directly in front of the eyes on the front edge of the cockpit. With a tiller handlebar, you can also attach the navigation system there.

GPS reception under carbon?

It is well known, carbon fibers are electrically conductive – and will shield from electromagnetic waves accordingly. Experience has shown that the GPS reception inside a velomobile works quite well, at least with modern GPS receivers. But the signal strength decreases noticeably when the hood is put on, and if there is also rainy weather with thick layers of cloud, the devices find it difficult to find the exact position. However, signal is usually sufficient for navigation – at least when looking on the display and with a good map (see Fig. 16). However when using voice navigation, the exact position is more important; if the announcement is delayed due to poor reception, one can easily miss a turn.

To improve GPS reception, some drivers use GPS repeaters. Alternatively, on some velomobile models, you can purchase a service hatch cover or hood made of fiberglass or basalt fibres, these are not electrically conductive and therefore permeable to GPS signals.

GPS tracks with and without hood

Fig. 16 GPS recordings made from inside the wheel arch (device: Wahoo Elemnt). On both maps you can see 10 rides each, which were made directly one after the other – on the left without a hood, on the right with a hood.

Not only the GPS position, but also the measurement of altitude can be a problem in the velomobile. GPS altitude measurement is notoriously inaccurate because the visible satellites are all up, i.e. in the vertical direction the signal propagation times hardly differ – accordingly, with poor GPS reception, GPS altitude measurement becomes even worse. However, most GPS receivers use a barometric altitude measurement, i.e. the altitude is determined from the air pressure and the GPS altitude is only used to detect air pressure variations (see Fig. 17). So altitude measurement should not be a problem even inside a carbon body. But as you can see in Fig. 17, the GPS altitude deviates from the true altitude at higher speed – there is obviously a negative pressure building up there which makes the altitude look too high.

GPS elevation and true elevation at different speeds

Fig. 17 Altitude profiles from 42 rides, recorded with GPS (device: Wahoo Elemnt) on the wheel arch. While the shapes of the altitude profiles are very similar there are big differences in the absolute heights.

GPS altitude and true altitude at different speeds

Fig. 18 Comparison of the altitude measurement of a GPS with barometric measurement (Wahoo Elemnt) at different speeds (velomobile: DF, without bonnet, GPS inside on the wheel well). Uphill the values agree, downhill at over 30 km/h there are significant deviations. The higher the speed, the greater the measured height, there is therefore a vacuum in the vehicle.

Which navigation system?

In a velomobile, a navigation device or GPS with a map display is very useful, because it allows you to drive with much more foresight and thus save energy in unknown areas, after getting lost it takes a lot of time and effort to regain speed after braking, especially in a velomobile.

However, there are hardly any independent navigation devices that are suitable for bicycles and especially for velomobiles. Car sat navs are completely unsuitable because they prefer highways and main roads – where bicycles are either prohibited or one feels very uncomfortable due to large speed differences. Bicycle sat navs often prefer cycle paths and dirt roads, on which efficient travel is impossible and where confusion or getting lost at high speeds is dangerous.

Real-time navigation therefore only works effectively on a smartphone, where the route calculation is carried out by BRouter and any supported app can be used to display it (e.g. OsmAnd, Locus Map).

Otherwise you can create a track before the trip and either follow the line on the map, or, if the track is provided with navigation instructions, follow these via audio announcements.

In addition to this functionality, there are the following things to consider with the hardware:

  • Power consumption (higher with a large display)

  • Protection (against splashing water): a velomobile is like a stalactite cave, electronics have to withstand humidity

  • Usability, even when damp, beware of bad touchscreens

Motor yes or no?

A velomobile drives fairly quickly on the flat, but is slow uphill and also accelerates slowly. The average speed is the average over time, i.e. slow sections are more important – it is more useful to be faster in slow sections than to increase top speed. Therefore, an engine is useful for travel time if there are many slow sections that can be shortened with it. Because you only need the engine on these few sections and not otherwise, the battery can be comparatively small.

However, an engine does not help much in stop-and-go traffic, there are many exhausting acceleration processes, but because of the braking in between, you will not be able to reach the speed of the open road, even with an engine. Here, an engine relieves the driver but does not make her/him much faster.

Also the amount of elevation gain is not as meaningful. On short climbs you can get some momentum beforehand and still have enough remaining speed at the top to be faster overall with a much lighter bike, but this only works up to about 10 meters of altitude at a time (see Fig. 32). You only go slower on long climbs (see: Can I ride in the mountains?), and only here would an engine reduce the travel time noticeably.

And then, of course, there are the legal regulations. In Germany (and the rest of the EU), a pedelec motor is allowed to assist up to a maximum of 25 km/h, on a straight stretch, it would only be in use for a few seconds in a velomobile, and is therefore simply not worthwhile. Uphill, of course, it’s different. Faster motors can not be retrofitted, see: Is it possible to fit a 45 km/h motor?

Which engine?

There are basically two types here bottom bracket motor and hub motor, direct drive or geared.

A direct drive hub motor has the advantage that it can also brake, you can feed the braking power back into the battery (recuperation), and thus recover up to 2/3 of the braking energy. This is particularly interesting in very hilly terrain, where you have to brake a lot, so you can protect the brakes and then use the energy for short accelerations. The problem is that these direct drive wheel hub motors have no gears, So they can either drive fast or be powerful, but not both. And the steeper the slope, the more power the engine would have to deliver, but its power increases with engine speed. So the slower it goes uphill, the lower the efficiency – the engine overheats quickly. Direct drive wheel hub motors are therefore unsuitable for long gradients. They are also quite heavy.

Geared hub motors are lighter (2.3 to 3 kg) and more adapted to a velonobile, being geared they develop more torque and climb quite well with relatively little power. In the correct gear – the driver still has to participate – long gradients up to 13-14% are climbed with ease. Most popular geared hub motors fit standard dropout width and accept 10 speed cassettes. Some common motors (Mxus) are freewheel motors and limited to 8 speed freewheels to stay within 138 mm dropout width.

Bottom bracket motors, on the other hand, are mounted beyond freewheels and gears. Therefore, they can be small and light, because they can work at high speed and use the bicycles gears, i.e. always work in the optimal speed range regardless of the gradient without overheating. No recuperation is possible. If you have long or very steep inclines, a bottom bracket motor is the better choice.

How should the gearing of a bottom bracket motor be chosen?

Electric motors are most efficient at a certain speed. The gearing should be chosen so that it is at comfortable pedaling cadence in the assistance range (i.e. pedelec up to 25 km / h). If the motor then also ensures that the driver accelerates as quickly as possible beyond the support area, the power consumption is the lowest – either the driver (mostly) pedals himself or the engine works in its most efficient speed range.

Is it possible to fit a 45 km/h motor?

In Germany and the rest of the EU this is not allowed – because according to StVZO § 63a Abs. 2 a bicycle may have an auxiliary drive with assistance up to a maximum of 25 km/h. There are indeed 2 and 3 wheelers (S-pedelecs) and even velomobiles that have a motor that assists up to 45 km/h, however, these are legally no longer bicycles, but are considered light electric vehicles – three-wheeled velomobiles, for example, belong to EU vehicle class L2e. This has the following consequences:

  • Registration: the vehicle must have undergone a registration procedure and carry a number plate.

  • Driving licence obligatory: A driving licence of class AM is required.

  • Compulsory insurance: You need motor vehicle insurance.

  • Compulsory motorcycle helmet

  • Use on cycle paths is prohibited

  • Restricted to approved components: One may not install arbitrary bicycle components, but must stick to type approved equipment.

  • Prohibition of certain types of construction, e.g. a glass dome roof.

  • One does not necessarily have to pedal, but can also ride purely electrically.

This means: You cannot simply equip an existing velomobile with a 45 km/h motor, but you must buy a specially approved 45 km/h velomobile. Since the approval process is expensive and time-consuming, there are only a few 45 km/h velomobiles available. And they are very different from sporty velomobiles – the weight is much higher, the tyres have higher rolling resistance, and riding without a motor is possible, but much slower than with non-motorised velomobiles. 45 km/h velomobiles are therefore a completely different type of vehicle with a somewhat different type of use.

Aerodynamics

Why are velomobiles fast?

Because of their low drag. As can be seen in Fig. 19, bicycles usually have rather poor aerodynamics, their drag coefficient (cW) is much higher than that of a car. The only reason why drag is not higher is because the cross-sectional area (A) is much smaller than that of a car. In contrast, the cross-sectional area of a velomobile is not necessarily smaller than that of a racing or recumbent bike, but the aerodynamics are much better than even an average car, so that the effective cross-sectional area (\(c_\text{W} \times A\)) is much smaller – for a very good velomobile only about 0.03 m², which corresponds to the area of a DIN-A5 sheet of paper.

Vergleich von cW-Wert, Querschnittsfläche und effektiver Querschnittsfläche zwischen verschiedenen Fahrrädern, Velomobil und Auto

Fig. 19 Comparison of the drag coefficient (\(c_\text{W}\)), the cross-sectional area (\(A\)) and the effective cross-sectional area (\(c_\text{W} \times A\)) between different bicycles, velomobiles and cars As you can see, racing bikes and recumbents are faster than Dutch bikes mainly because their cross-sectional area is smaller, the aerodynamics are not very good either. On the other hand, the cross-sectional area of a velomobile is similar, but the aerodynamics are much better.

What is drag?

Air resistance is basically very complicated, the following effects occur there:

  • Pressure resistance: The air is slowed down (from the vehicle’s point of view) and accumulates, i.e. kinetic energy is converted into pressure energy (and to a small extent into heat), but this cannot be completely converted back into kinetic energy from the outflowing air because of frictional resistance and is therefore lost.

  • Frictional resistance: Friction occurs between the stationary surface and the flowing air, the rougher the surface, the more a boundary layer forms. The air directly contacting the surface adheres to it, its velocity is zero. With increasing distance from the surface, the speed of the air increases until it reaches a maximum outside the boundary layer, in the free air. Here the boundary layer thickness varies from 0 at the front to several cm at the rear of the velomobile.

  • Interference resistance: Interaction between the flow around neighbouring bodies, e.g. wheel and body.

  • Induced drag: Different pressures lead to an equalising flow due to their differential force, e.g. in an aeroplane the lift of the wings results from the differential force between a positive pressure on the underside and a negative pressure on the upper side, but the resulting equalising flow from the underside to the upper side causes turbulence at the wing ends.

Depending on the shape of the body, these effects vary in strength. With a blunt body, the pressure resistance dominates, but the more streamlined a body is, the lower the pressure resistance becomes, i.e. the proportion of frictional resistance becomes greater and thus dominant. A velomobile has a streamlined basic shape, but air accumulates at the wheel wells, mirrors and lid, so that both pressure and frictional resistance play a role. In a highly optimised record vehicle it is different, there the frictional resistance dominates, for example, in the record vehicle Aerovelo Eta the sponsor stickers were gathered at the very back of the fuselage in order to make the surface in the area in front of it as smooth as possible and thus low-friction (largely laminar).

In any case, drag leads to a gradient in the flow velocity air, and this can lead to turbulence.

What is the cW value?

Air resistance is caused by a combination of different effects (see: What is drag?), in which not only the shape of the body plays a role, but also, among other things, its surface properties and size as well as the density and viscosity of the medium. A turbulent flow is too complex to be calculated analytically, but for similar conditions one can describe the influences approximately by parameters.

One of these parameters is the \(c_\text{W}\) value (drag coefficient), which describes the amount of drag independent of velocity, size of the body and density of the medium. It includes both the shape of the body and (to a lesser extent) its surface characteristics and describes the magnitude of all effects that enter into the drag in a single number. Thus, it is a completely non-physical value to which no single law or measurement corresponds. But it can be used to calculate air resistance with sufficient accuracy by multiplying it by the density of the medium, the reference area and the square of the velocity.

Strictly speaking, the \(c_\text{W}\) value is not even a constant, because it changes with the flow behaviour of the medium. This is described by another key figure, the Reynolds number. Roughly speaking, it expresses the relationship between inertia and internal friction: If the number is low, the medium moves as a unit, a high viscosity causes a strong internal friction, which hinders relative movements within the medium – it is a laminar flow. With a high Reynolds number, on the other hand, the viscosity or internal friction is proportionally low and the inertia is high, so that parts of the medium can move more independently of each other and, if they are deflected at an edge, for example, they do so – turbulence occurs. The Reynolds number is not only material-specific, but also increases with velocity – while a slow medium is still laminar, the inertial forces increase at higher velocities, so that turbulence occurs. In addition, turbulence also needs a certain distance to develop – which is why the same flow can still be laminar with a small body, but already turbulent with a large one, and why you need a correspondingly higher flow velocity in a wind tunnel with scaled-down models in order to produce the same Reynolds number and thus the same flow behaviour.

For the \(c_\text{W}\) value, this means that it decreases relatively strongly with velocity in a laminar flow, whereas it decreases only slightly with velocity in a turbulent flow. In the case of vehicles, at least at higher speeds where air resistance is no longer negligible, one is in the turbulent range and can therefore assume a constant \(c_\text{W}\) value around the reference speed. However, this means that one has to know the reference speed at which the \(c_\text{W}\) value is measured – for cars this is 140 km/h, for velomobiles one would have to choose a practical reference speed of e.g. 50 km/h accordingly in order to be able to compare the aerodynamics.

How to achieve low Air resistance (drag)?

Ultimately, by releasing as little kinetic energy into the air as possible. I.e. the air should be slowly pushed aside and slowly return to its original position. Any acceleration energy that is released into the air is lost for the drivetrain.

A single air molecule cannot be accelerated linearly very much because the neighboring air molecules are in the way, they should also be accelerated. You can squeeze them a bit, but that creates back pressure. Linear acceleration therefore quickly reaches its limits. The situation is completely different with rotations: an air molecule can rotate almost anywhere on the spot without being obstructed by the neighboring air. You can lose a lot of energy by rotating air – so it is essential to avoid turbulence.

Why is a teardrop shape usually quite efficient?

In short: because it is easier to push air part without a vortex than to let it flow back together again without a vortex. The former can therefore happen quickly, the latter takes more time and therefore a longer journey.

But of course there are exceptions. For example, the Kamm tail, which corresponds to a sharply cut teardrop shape. This will interupt the flow suddenly, instead of it being sucked back in. Aerodynamically, this is not quite as good, but it saves you a long stern, with its extra weight and its larger wind attack surface.

Why no wing profile?

The profile of an airfoil is considered the perfect example of an aerodynamically good shape. But why does it have this shape? First, a wing must accelerate air downwards, and for this it must have an angle of attack relative to the flow angle. Second, on the other hand, where there is negative pressure, they do not break the current, that is why this side is not straight, but curved so that the air is slowly drawn around the curve instead of abruptly tearing off at the front edge. A velomobile, on the other hand, has no angle of attack (if the wind results from the front), it shouldn’t accelerate the air if possible. The same thing, however, is that a stall should be prevented – hence the even, flowing shape without sharp kinks.

Why is the front blunt instead of being pointed?

Usually one associates little flow resistance with slim, pointed shapes. But the front of most velomobiles is round. It is true that the incoming air has to be slowly moved to the side instead of suddenly hitting a blunt surface. But this is not the case: the air builds up in front of the velomobile during the journey, and this “mound of air” acts like an elongated snout that deflects the incoming air to the side beforehand. Such a mound of air would not build up on a point, i.e. the air would suddenly hit the body, nothing would be gained.

Why are open wheel wells sharp at the front and round at the back?

At the front of the wheel well, the edge is as sharp as possible so that the air flow does not follow the body, but tears as much as possible and flows back past the wheel. Of course, this is only possible to a certain extent. At the rear, on the other hand, the part of the flow that has penetrated further into the wheel well should not hit and swirl on the rear of the wheel well, but should be gently pressed outwards so that the air flow is again parallel to the body.

What are pants?

These are fairings that are used on velomobiles with open wheel wells and are attached outside over the wheels to improve aerodynamics (see Fig. 20). As a result, you have the good aerodynamics of a velomobile with closed wheel wells – but a slightly wider track (i.e. better cornering stability) and the flexibility to convert to open wheel wells again. Pants can be removed if something has to be done on the tyres or wheels, but they also make overall width increase significantly. The latter can be mitigated with a special wheelset with the rim spoked further inwards.

Pants were first made for the DF design, but fit most velomobiles with open wheel wells. They are quite light and flexible and are simply stuck on with adhesive tape.

Alpha 7 with pants

Fig. 20 Two velomobiles with open wheel wells, in front an Alpha 7 with pants on the front wheels, behind it a DF without pants.

Why is the underbody next to the foot bumps curved, not flat?

Foot bumps protrude outwards and displace the incoming air. However, this cannot escape arbitrarily, because the road is a few centimeters below. So it can only escape to the side. If the vehicle floor next to the foot bumps were flat, the flow cross-section for the air would be reduced at the level of the foot bumps, the air would be braked and pushed out to the side where it would disrupt flow. If the floor next to the foot bumps is arched inwards, the cross-sectional area remains the same and the current can flow around the foot bumps more easily.

Why are no wind tunnel tests done?

Because they are complex and expensive.

Also, it’s not that easy to generate realistic flow conditions. While an aeroplane is smooth on the outside except for the engines and is only surrounded by air, a velomobile has rotating wheels in wheel wells as well as the road, which is a few centimeters under the velomobile. For correct simulation, it is therefore not sufficient to place the velomobile in the wind tunnel or to model it on the computer (so-called CFD simulations), the wheels have to turn and the road has to move underneath it.

And last but not least: to get new insights, it is not enough to simply place a vehicle in the wind tunnel and observe it. You have to set up and test hypotheses in a targeted manner, i.e. have a clear idea of what you could change and how. And precisely because velomobiles are already pretty good aerodynamically, further improvements can only be achieved if you know exactly what detail you are looking for.

Why is the large air intake at the front of the body?

Because he stagnation point is there.

If the air intake were further back on the fuselage, then you would have to somehow channel the air inwards without the outside flow detaching from the fuselage – that would happen if an air inlet was in the wind. On the other hand, as the air builds up in the front, you don’t need a device that directs the air inside, just a simple hole is enough. In addition, the flow is always there due to the dynamic pressure, and an air inlet there does not damage aerodynamics.

What is a NACA duct?

This is a triangular air inlet (see Fig. 21), which is located on the side of bodywork. The inventor and namesake is NACA, a US aeronautical research institute and forerunner of NASA.

If you want to bring air from the side into the interior of a vehicle, a simple hole is not enough – the air flows across the hole and therefore not through the opening. One would have to redirect the air, however, a corresponding “scoop” would increase the cross-sectional area and thus the air resistance, and possibly create braking vortices behind it. The NACA duct takes a different approach: the opening is in the direction of travel (i.e. in the direction of flow), but offset inwards – this does not increase the cross-sectional area. In order for the air to come in, there are two sharp edges in the direction of travel. It tears at this edge, and vortices are formed that rotate inwards and come into the inlet opening. The NACA-Duct brings the air inwards by creating inward vortices and then sucking these vortices away – so that the outside flow is nice and smooth.

NACA-Duct on the bonnet of a Milan SL

Fig. 21 NACA-Duct on the bonnet of a Milan SL

Does ventilation slow you down?

Basically yes. The air hits the velomobile at driving speed and is braked inside – at the latest when it hits the driver. The energy for the deceleration of the air (from the perspective of the velomobile) or the acceleration (from the perspective of the environment) must indeed come from somewhere, namely from the kinetic energy of the vehicle. On the outside, on the other hand, the air would be deflected but hardly slowed down, meaning that less energy would be lost there.

Ventilation, on the other hand, has a positive effect if it succeeds in improving the external flow. For example, the steep front of the bonnet creates a stagnation point, i.e. a “mountain of air” accumulates there, which on the one hand makes the front of the bonnet less steep for the flow (see: Why is the front blunt instead of being pointed?), but on the other hand increases the effective size of the bonnet – the air flow has to flow around all the stagnated air and is therefore more difficult to get behind the bonnet/scoop. If some of the air is now diverted inwards here, it ends up exactly where it is needed (by the driver or on the inside of the windscreen, where it prevents the windscreen from fogging up), and not only is there no turbulence on the outside, but the “mountain of air” accumulated there with its damaging effects on the flow becomes smaller. It would also be positive if it were possible to divert an already swirled boundary layer towards the inside, so that the flow outside is again nice and smooth. However, this happens mainly in the rear part, behind the driver.

Are freestanding wheels a lot worse than wheel wells?

At first glance, a velomobile with free-standing wheels like the Le Mans by Cycles JV & Fenioux looks very attractive: the body is narrower yet the track is still wide, the steering linkage can be covered aerodynamically, larger wheels are possible, and steering lock is not limited by wheel wells. And if the wheels are covered with wheel disks, you would think that the aerodynamics should not be that bad either.

The problem, however, is that the top of the wheels are in the wind. Even a smooth tyre draws a lot of air with it (with tread it is much worse), and with free-standing wheels, the top moves against the air at twice the speed of the rotation. When the wheel wells are closed, the airstream is not attacked, the air in the wheel wells is stationary relative to the vehicle, so that the wheels only work against the air at that speed. Aerodynamic drag increases quadratically with speed, performance even cubic – i.e. a free-standing wheel needs eight times the power at the top because of its double speed. Worse: the air transported there by the tyres not only moves parallel to the tyres, but also to the side, where it hits the wind – i.e. the aerodynamically effective width of such a wonderfully narrow tyre is much larger. And even worse: since the wheels are not extremely far from the bodywork, these air turbulences also destroy the flow along the bodywork (interference resistance; see: What is drag?). Wheels are relatively far forward, thus most of the velomobiles laminar flow is destroyed. This is the reason why free-standing wheels are extremely poor aerodynamically.

It would be different if you provided the wheels with wheel covers – one could make them very narrow (because they steer in curves), and it is only important that the upper half or the upper 2/3 of the wheel are covered.

How does wind make itself felt?

Headwind or tail wind is felt much less than on normal bicycles. You can see on the speedometer that the speed differs slightly in headwinds or tailwinds, you don’t have the feeling of “pushing against a wall” in headwinds.

Cross winds are noticeable mainly due to the influence on the steering, if you have a suitably sensitive velomobile, you should drive carefully in strong winds. The absolute force is not the problem, but its sudden change – if you come out of the slipstream at the end of a hedge, for example, when driving downhill quickly and a gust of wind suddenly pushes on the velomobile, you need a space of one meter to the side to be able to compensate. So this is of course extremely unfavorable if it is a narrow street with oncoming traffic. The actual offset by the wind is much smaller, but without warning you need very good reflexes to keep the steering reasonably calm.

If the cross wind is very strong, it can also knock over a velomobile. This is not surprising, since the side of a velomobile is not much smaller than a car (approx. One third), but weighs much less (approx. one tenth with the driver), the ratio of the height (or height of the wind attack point) to the track width is also less favorable than in most cars. And so it can happen in a strong storm (approx. Wind force 9/10) that the wind tips you in a straight line.

How can wind sensitivity be reduced?

Wind sensitivity can have various causes:

  • Caster: If the struts are not perpendicular (to the longitudinal axis of the vehicle), the wheel contact point is not at the tracking point, but usually behind it. This tracking is desirable on a single-track because it stabilises the ride – the wheel follows the direction of the bike and counteracts tipping, so the steering straightens itself. With a multi-track this is not only less important, but even counterproductive in windy conditions: if the wind pushes the vehicle to the side, the steering follows and steers the velomobile away from the wind. Without caster, there is no lever through which the wind force could exert torque on the steering. Caster can be reduced by adjusting the trailing arm.

  • Flow around: A body that is flowing around experiences a force, if this is undesirable, it can help to deliberately break off the flow, for example with Stormstrips – these are thin, angular profiles that are glued to the body in the longitudinal direction. Thus, they do not create any additional drag in the direction of travel, but wind flow across the vehicle tears off and swirls there instead of creating lateral lift.

  • Shape of the body: If the wind-attack surfaces of the body are asymmetrically arranged, the pressure point is not in the centre of the vehicle, and thus a torque is generated which depends on the direction of the apparent wind. Again, Stormstrips could help balance the wind forces better.

  • Centre of gravity and height of the point of wind attack: the higher the point of wind attack, the greater the leverage over which it causes the vehicle to tip. And the higher the centre of gravity, the less it needs to be lifted when tipping. So to make a vehicle less susceptible to tipping, you can set the centre of gravity as low as possible (and as close as possible to the widely spaced front wheels), and use tricks such as stormstrips to reduce the wind force on the top.

  • Swirling: At high speeds, it can be a problem for vortices to shed unevenly, causing the tail to become unsteady. While this behaviour is basically dictated by the shape of the body, it is possible to provide for small-scale turbulence, e.g. by a rough surface at the stern (golf ball effect), which prevents the formation and detachment of large vortices.

What is the sailing effect?

Cross winds can provide propulsion, i.e. the velomobile “sails” forward. This is actually not surprising, because it has a relatively large side surface that the wind can blow along, it rolls easily, and it is very stable. So like a sailboat – with a very small sail, but very little drag and an excellent keel without any drift. The only problem is that a velomobile is usually much faster than the wind. That means: no matter from which direction the wind blows, the apparent wind always comes more or less from the front – and for it to come from the side at full speed, it has to blow quite hard.

A strong sailing effect does not necessarily imply a strong wind sensitivity (see: How can wind sensitivity be reduced?) – if the body is shaped in such a way that a strong wind force is generated and this has a large component in the direction of travel, this does not have to affect the steering, nor does it have to exert a torque on the body, nor does it have to lead to unevenly detaching vortices.

The sailing effect varies in intensity depending on the velomobile model. A more detailed investigation has so far only been made for the Milan GT 2011 in the Volkswagen wind tunnel, this has shown that the sail effect exists and is strongest at a direction of the apparent wind of 75°.

Under what conditions is drag the least?

The shape and cross-sectional area of a vehicle are unchangeable, and you don’t want to reduce the speed, you want to increase it – this leaves only the air density with which the air resistance can be changed. There is the lowest possible air density:

  • At high altitude. However, physical performance also decreases with altitude, since oxygen supply to the body becomes more difficult. Since oxygen is relatively heavy compared to the other air components, the amount of oxygen available decreases disproportionately with altitude. Therefore, attempting records at 1000 m is a good idea, but not at 3000 m.

  • At high temperature. Then the rubber of the tyres is not as rigid and the rolling resistance is lower. However, the body must not overheat, because otherwise the performance drops sharply.

  • In humid air. However, this makes cooling more difficult because sweat does not evaporate as well and the body overheats more easily. In addition, raindrops on the surface interfere with aerodynamics, and water on the road increases rolling resistance.

As you can see Battle Mountain in the Nevada desert at an altitude of 1375 m with a warm desert climate is quite well suited for record attempts. But there is another reason why the records are set there of all places: there are hardly any roads with perfect asphalt that are absolutely level over several kilometers. Accordingly, the IHPVA allows in their rules, a maximum gradient of 2/3%. The Battle Mountain course is very close, with a gradient of 0.64%. This is very little and is roughly in the range of the breakaway resistance, this means that a bicycle does not roll off on this slope. But if a record driver is traveling at 113 km / h, that’s a good 31 meters per second, with a gradient of 0.64%, that’s 20 cm difference in height per second, which means almost 200 W additional power for a weight of 100 kg – almost as much as a legal German Pedelec motor gives at continuous performance.

Basic Physics

Note: The example calculations here always refer to a 75 kg rider in a 25 kg velomobile, in reasonably good condition, but without record fitness.

What is rolling resitance?

Rolling resistance is fundamentally complicated because it is a collective term that includes many things:

  • Deformation of the tyre

  • Roughness of the road

  • Rolling and sliding friction in the bearings

In particular, the deformation of the tyre is not only geometrically complex, it’s hysteresis losses also depend on the material properties of the rubber. And this is itself a complex combination of many substances and a company secret of the manufacturer. What exactly happens there cannot be calculated – at least not as an outsider.

Nevertheless, it shows that the rolling resistance is primarily dependent on the load. Accordingly, one might summarize all influences on the rolling resistance value, which can be regarded as a constant, and thus very well describes the rolling resistance in the relevant speed range. This constant has little to do with the physical processes behind it, but it works well enough.

This is not enough for bean counters like velomobile drivers, here it has been shown that there is still a small speed-dependent share in rolling resistance – and this is the so-called dynamic rolling resistance. There are hardly any studies of the causes, only speculations. It is also an empirical factor that works well enough, and with the help of which the rolling resistance can be adequately described within the scope of measurement accuracy.

How do you reduce rolling resistance?

  • Tyre pressure: High air pressure = lower friction. But this only works to a certain degree, if the tyre is already hard, it hardly deforms, and therefore, no matter how high the pressure, it can hardly be made to deform even less. This increases the risk of tyre damage. And on rough ground, the tyre does not spring, but jumps, which also costs performance (see: What tyre pressure? and Fig. 12). Therefore, a good recommendation is to use the maximum pressure specified by the manufacturer and somewhat less on rough roads.

  • Fast tyres: It largely depends on the rubber compound, which cannot be assessed from the outside. Here you have to rely on the information provided by the manufacturers, together with practical reports.

  • Thin-walled tyres: All fast tyres have in common that they are relatively thin-walled – tyre flanks especially often have very little rubber. The less rubber there is, the less it has to be deformed, the losses are correspondingly lower and the comfort is also higher because the tyre adapts better to the road.

  • Latex tubes or tubeless: The tube also contributes to the thickness and therefore rigidity of the tyre. That is why a thin tube is better than a thick one, so use latex tubes or go tubeless (where the tyres are a bit thicker). As one can see in Fig. 22 these tyres do not only roll better, but are also less sensitive to reduced pressure – which better absorbs the braking vibrations.

  • Wide tyres: Geometrically, wide tyres are faster because the tyre contact patches are shorter and are not deformed as much in the direction of travel. However, a wide tyre can withstand less pressure, and is usually strongly built. Therefore, one cannot simply conclude the rolling resistance from the tyre width, but must look at the entire tyre.

Increase in rolling resistance when pressure drops from 8.3 bar to 4.1 bar

Fig. 22 Rolling resistance of road bike tyres at 8.3 bar as well as its increase at only 4.1 bar tyre pressure. As you can see, tubeless tyres not only have a low rolling resistance, but this also tends to increase less at low air pressure.

Why are good tyres so important on a velomobile?

On a straight line at constant speed, the drive energy is primarily spent in air resistance and rolling resistance. While the rolling resistance force is constant, the air resistance force increases quadratically with speed, that’s why air resistance dominates on normal bicycles as soon as you roll more than just comfortably (about 25 km/h).

It’s different with velomobiles. There the air resistance is much lower, so even at about 50 km/h air and rolling resistance are approximately the same value (see Fig. 23). But hardly any velomobile driver constantly drives more than 50 km/h where air resistance dominates. Because the air resistance is so small, the share of rolling resistance is much larger, so that it dominates during most of the journey (see Fig. 24). (On fast downhill sections, the air resistance clearly dominates, but that only makes up a small part of the driving time.) That is why good tyres are important for the velomobile, because it can reduce the dominant rolling resistance. This goes so far that even the material of the tubes (Butyl or Latex) and the ambient temperature noticeably affect the attainable speed.

Comparison of drive losses, rolling resistance and air resistance

Fig. 23 Comparison of propulsion losses (\(P_\text{chain}\)), rolling resistance (\(P_\text{roll}\)) and air resistance (\(P_\text{air}\)) at different speeds. As you can see, air resistance is completely negligible below about 15 km/h, and even at 50 km/h it is not even half. This means that rolling resistance mostly dominates during the ride – except for a few high-speed sections. On other bicycles, rolling resistance and drive loss are roughly similar, but air resistance is significantly higher, by a factor of 5 to 10.

Speed histogram of a long-distance journey

Fig. 24 Histogram of the speeds of a long-distance ride (to SPEZI 2019, first approx. 200 km). Assuming the same resistance coefficients and the same system weight as in the other diagrams (and neglecting braking), we see that despite a brisk speed (median: 36.7 km/h) about twice as much energy went into rolling resistance (62%) as into air resistance (31%).

Why are velomobile riders so obsessed with low weight?

Many people say that low weight is not so important to them:

  • You won’t be racing, so gaining a second or two is not important, reliability is more important.

  • The weight saving is negligible in relation to the total weight (driver, vehicle and luggage).

Both are basically correct, but apart from the fact that reliability has little to do with weight (a light vehicle is not necessarily more fragile, but only more elaborately manufactured), you have to look at the energy balance. And that’s where air resistance comes in again – or its absence. While with a normal bike at high speed a large part of the energy is used to beat air resistance, it is usually less than half of that in the velomobile, despite significantly higher speeds. And the mass does not go into overcoming air resistance, but into the following resistances:

  • Rolling resistance: This is usually the dominant resistance, and it increases linearly with weight.

  • kinetic energy: Since a velomobile can reach much higher speeds, the acceleration phases are also much longer – the kinetic energy increases quadratically with speed, i.e. for twice the speed, four times the energy has to be put in. Accordingly, acceleration phases account for a much larger proportion in the energy balance (see Fig. 28 and How to achieve the highest possible average speed?).

  • Uphill: Here you also notice the weight clearly, and because uphill the velomobile has no aerodynamic advantage but a disadvantage due to its higher weight (see Fig. 31), the difference feels even stronger.

So with a velomobile, weight plays a bigger role, but at the same time it is significantly heavier – that is why a weight reduction pays off more (see Fig. 25). And not only for sporty riders: the weaker a rider is (i.e. the lower the speed), the lower the air resistance – and the rolling resistance has a greater share. In addition, strong riders can often take hills with momentum (dolphining) on undulating terrain, while weak riders have to lug their weight up without momentum.

And quite pragmatically: It is very pleasant when a velomobile is so light that you can carry it alone, e.g. over a few steps or lift it onto a train, instead of always needing a second person to do it.

More power with luggage, for road bike and velomobile

Fig. 25 Additional power requirement due to luggage, for a road bike (dashed) and for a velomobile (solid). At low speeds, the luggage plays a greater role on a road bike because the bike is significantly lighter – the luggage therefore accounts for a greater proportion of the total weight. However, this is also the case with light riders. At higher speeds, air resistance is added, so weight is less noticeable, but because a velomobile has less air resistance, the influence of luggage weight remains much greater. Accordingly, especially light, low-powered riders should pay attention to keeping the total weight as low as possible.

Why is it said that the wheels count twice in weight?

The following energy is required to bring a wheel up to a certain speed:

  • Kinetic energy: \(E_\text{kin} = 1/2 \times m \times v^2\)

  • Rotational energy`: \(E_\text{rot} = 1/2 \times J \times \omega^2\)

As you can see, both formula are quite similar:

  • \(J\) is the Moment of inertia`, it depends on the mass and its distance from the fulcrum. If you assume that the mass is on the outside of the wheel, the following applies: \(J = m \times r^2\).

  • \(\omega\) is the Angular velocity`, i.e. the number of revolutions over time (\(2 \pi / T\)). Since the tread of the wheel must move at driving speed, i.e. the circumference over time must be equal to the speed ( \(2 \times r \times \pi / T = v\)), we get: \(\omega =v / r\)

If you use this formula, the following results as rotational energy: \(E_\text{rot} = 1/2 \times m \times r^2 \times (v / r )^2 = 1/2 \times m x v^2\). This means that if the entire mass of the wheel sits on the outside of the tread, the rotational energy is equal to the kinetic energy – hence the statement that the weight of the wheels would count twice, because rotational energy and kinetic energy taken together are twice as large as the kinetic energy of one rotating component.

In reality, the rim and tyre make up the largest part of the wheel, but the mass is not completely on the outside, so the contribution of the rotational energy is smaller – so it is not quite twice as much. And as you can see, the size of the wheel is irrelevant: a large wheel has a higher moment of inertia (into which the radius is squared), but rotates correspondingly slower (the angular velocity is also squared) so that the rotational energy is the same.

But how much does this effect matter in practice? Suppose you have a velomobile with a system weight of 100 kg, its wheels weigh 4 kg. Their rotational energy is correspondingly about 3% of the kinetic energy. At 50 km/h, the kinetic energy is 9645 J, together with the rotational energy of the wheels, this is about 10 kJ. If you manage to make the wheels 500 g lighter on the outside, this corresponds to a weight reduction of 1 kg, and for acceleration to 50 km/h 100 J less have to be applied. With a pedalling power of 300 W, it takes almost a minute to reach 50 km/h – but the 100 J saved correspond to only a third of a second of pedalling power. Moreover, the rotational energy is not lost if you don’t have to brake it away. Therefore, it does not play a major role in normal cases.

Why is it so difficult to to drive faster?

Top speed is mainly determined by air resistance – because this increases faster than other driving resistances. The air resistance force increases quadratically with speed, so to travel a distance twice as fast requires four times as much energy for air resistance. But since you then cover the distance in half the time, you have to provide the aforementioned four times the energy in half the time – work per time is power, so you have to provide eight times the power for twice the speed. (This is somewhat simplified and only approximately valid for very high speeds, since air resistance is only a part of the total driving resistance, it increases more slowly, see: Why don’t you accelerate in a velomobile “like against a wall”?)

A second limiting factor is the large amount of kinetic energy that has to be built up (see Fig. 28 and Why are the acceleration phases so long in a velomobile?). This is because it takes a lot of time and distance – especially at high speeds, a large part of the pedalling power is already consumed by the driving resistances, so that only little power is left for further acceleration. This means that you only accelerate very slowly, but because of the already high speed you need a lot of free distance to do so.

And a third problem is time: If you need 20 minutes for a certain distance at a certain speed, you can save 10 minutes by doubling the speed – but (if you only consider the air resistance) you have to produce eight times the power. A further doubling of speed again results in eight times the power requirement (i.e. 64 times the power in total), but only saves a further 5 minutes (i.e. only saves a further quarter of the original journey time).

How fast can a velomobile go?

With a velomobile air resistance is significantly lower and the theoretically achievable top speed is significantly higher than that of other bicycles. However, kinetic energy is not less – due to the higher mass of a velomobile, it is even slightly higher. The speed is quadratic to the kinetic energy, in order to drive twice as fast, you not only have to achieve eight times the power to overcome the air resistance, but also four times the kinetic energy during acceleration.

If you can achieve 400 watts, for example, it takes almost one and a half minutes and 1.1 km to reach 65 km/h (system weight: 100 kg). If you drive downhill, you get quite a few watts from the slope, however, as the speed increases, the necessary downhill section becomes longer and longer: if you go downhill at 3%, for example, you can reach around 104 km/h with 400 watts of pedaling power in two minutes – but the downhill section must also be almost 1.7 km long and have 160 m difference in altitude – and be significantly longer for the braking distance without curves or obstacles. And it would take even three times the time and distance to reach the terminal velocity of 118 km/h (see Fig. 26). Therefore, a trained driver rarely ever reaches 100 km / h in ideal conditions, although much more should be possible on an ideal gradient.

acceleration with muscle power and downhill

Fig. 26 Comparison of acceleration: If you pedal at a constant 400 W, you accelerate faster at the beginning than if you roll downhill with a 3% gradient. But on the downhill you accelerate more strongly at high speeds and reach a significantly higher top speed, because with increasing speed the altitude energy released per time also increases, so that after 180 metres of descent more than twice the driving power (without drive losses) is effective.

Are Velomobiles faster because of their high top speed?

Not really. Since people are quite underperforming, the maximum speed that can be achieved depends more on the gradient than on the pedaling power (see Fig. 26), and since high top speeds require very long acceleration distances (see Fig. 34), in practice the maximum speeds of velomobiles and other fast bicycles are not all that different.

But what is significantly different with a velomobile is that you can keep a high continuous speed much longer than with another fast bike – if you do not have to brake.

Why don’t you accelerate in a velomobile “like against a wall”?

With aerodynamically poor bicycles, you feel a relatively clear maximum speed, even with significantly more effort, you no longer get appreciably faster. This feels different on a velomobile: you don’t “pedal until you’re hitting a wall”, but can accelerate further more easily (which only takes a correspondingly long time and requires a correspondingly long distance, see Why are the acceleration phases so long in a velomobile?). The reason is the high proportion of rolling resistance (see Fig. 23), as this increases more slowly than air resistance, the total resistance also increases with a lower power (see Fig. 27) – so at the same speed, not only is the total resistance lower on a velomobile than on a conventional bicycle, but the total resistance also increases more slowly there, and so it is easier to go even faster.

Increase in driving resistance

Fig. 27 Effective exponent by which the driving resistances increase with speed, at different gradients, the solid line is a velomobile, the dashed line is a racing bicycle. The various driving resistances increase with speed at different rates, while rolling resistance (which is not speed-dependent) increases linearly with speed, air resistance increases cubically. But because rolling resistance is a large component of total resistance on a velomobile, it effectively grows only approximately quadratically even at high speeds, whereas on a road bike it is much closer to cubic growth. Uphill, the effective exponent is smaller because the climbing work, which only grows linearly with speed, is added – the total resistance is of course much higher uphill than on the flat, it just grows more slowly.

Why are the acceleration phases so long in a velomobile?

In any case, it is not due to inefficient power transmission or a higher total weight – as you can see in Fig. 34, a velomobile accelerates practically as well as any other bicycle – at low speeds, the higher mass is noticeable, but from about 30 km/h onwards, this is overcompensated by the low air resistance.

Nevertheless, the acceleration phases are very long – so long that you can’t sprint back to cruising speed after every braking, but have to divide your power. With other bicycles this works well, and is even recommendable from an efficiency point of view (see: How to achieve the highest possible average speed?), with the velomobile, however, you would totally exhaust yourself if you wanted to ride the long acceleration phases with a high sprinting power.

The reason for this is: With the velomobile, the air resistance is small, but not the kinetic energy. So you can ride at high speeds, but you have to build up the necessary kinetic energy first. As you can see in Fig. 28, at 50 km/h the kinetic energy is more than 50 times the energy needed for air resistance, rolling resistance and drive losses per second at that speed. Or to put it another way: someone who kicks enough continuous power to drive 50 km/h (in our example, that’s about 179 W) first has to do twice as much for 40 seconds to accelerate the vehicle upwards – and needs 400 metres of free distance to do it. And then when a red light comes, it starts all over again.

With any other bicycle, by the way, it would take just as much effort and time to build up the kinetic energy needed for high speed – you are just stopped much earlier by air resistance.

That, by the way, is the reason why in Battle Mountain a run-up of 8 km is taken even for the 200 m sprints – there is no other way to build up the kinetic energy for 130 km/h, the measuring distance itself is then passed in a few seconds. And it might also be one of the reasons why long-distance endurance records are popular in the velomobile scene – because there you only have to accelerate once, and then “only” maintain the speed.

kinetic energy and driving resistances

Fig. 28 Comparison of the kinetic energy with the driving resistances, the reference value is the energy consumed by the driving resistances in one second. Since in the velomobile the driving resistances are lower at high speed, the build-up of kinetic energy is much more significant.

Why is performance data alone meaningless?

Since power meters are now widely available, many velomobile riders indicate how much power they needed to go a certain speed. However, power requirements depend on many things:

  • Weight of rider and vehicle

  • condition of the road (rough, smooth)

  • which tyres, at what air pressure

  • temperature, wetness of the road

Without the specification of these boundary conditions, however, a performance specification is not worth much. Firstly, it determines the rolling resistance (see: Why are good tyres so important on a velomobile?), and secondly, the acceleration work – because of the long acceleration phases (see: Why are the acceleration phases so long in a velomobile?) you rarely ride at a constant high speed. And so it can happen that a only moderately efficient velomobile achieves fantastically low wattage values if the rider is a lightweight and only rides in good conditions and on good roads.

Why do cars accelerate better even though they are heavier?

Because they have much higher power reserves.

An average-sized athletic cyclist can indeed produce many hundreds of watts – but only over a few seconds. Over a longer period of time, it is more in the order of 200 watts. As you can see in Fig. 23, you can do a good 50 km/h with it in a velomobile. However, the power requirement uphill or when accelerating is a multiple of this (see How to achieve the highest possible average speed? and Why are the acceleration phases so long in a velomobile?). For example, the kinetic energy at 50 km/h is already just under 10 kJ, for a car, accelerating from 0 to 50 km/h in 10 seconds is pretty lame, but a cyclist would have to muster an average of 1000 watts for this in addition to the other driving resistances. Only a professional athlete can do that.

A car needs 13 kW for 100 km/h (with the data from Fig. 19 and 1000 kg) – converted, that is just under 18 hp. But hardly any car has less than 100 hp. A leisurely acceleration from 0 to 50 km/h in 10 seconds only takes an average of 13 hp – so together with the driving resistance, the car needs less than a quarter of its maximum power (which a car cannot produce for just a few seconds) instead of having to expend all its energy on it like a cyclist.

Uphill, by the way, it is similar: to ride up a 5% incline at 50 km/h requires almost 1000 watts with the velomobile – hardly feasible even for a professional athlete. A 1000 kg car needs a good ten times that power (over 11 kW), but these 16 or so hp are ridiculous for a car. This only slowly reaches its limits when driving on the motorway through the Kassel hills (up to 8%) at well over 100 km/h.

How to achieve the highest possible average speed?

Actually, it is quite simple: only the energy invested in rolling and air resistance is lost (and of course the braking energy), in contrast, acceleration and climbing work is converted into another form of energy and can be recovered again. And of the losses, it is only air resistance that increases strongly with speed. Therefore, it is energetically most favourable to travel at as constant a speed as possible.

On the other hand, the body wants power output to be as constant as possible – high power peaks require anaerobic metabolism and fatigue the muscles.

What does this look like in numbers?

  • As you can see in Fig. 31, for the calculated example velomobile, the power demand at 1% incline is about twice as high as on the flat (slightly less at high speeds). Because of the small-angle approximation, one can accordingly assume that one needs three times the power for 2%, four times the power for 3%, and so on. Uphill, therefore, you can spend an extreme amount of power if you want to maintain your speed. Or vice versa: If you assume that on uphill gradients at correspondingly low speeds the power increases approximately linearly (this is quite true, see Fig. 27), then at the same pedalling power the speed at 1% gradient is only half as high, at 2% gradient only a third, etc.

  • Acceleration is a little more difficult to describe, because you don’t consciously ride with a certain acceleration power – you accelerate with the remaining power, after subtracting all the other riding resistances. But in the end, acceleration can be described in the same way as an incline – because with the latter, the acceleration due to gravity is effective, which is 9.81 m/s². So a 1% gradient that doubles the driving resistance corresponds to an acceleration of 1% of the acceleration due to gravity, that’s about 1/3 km/h per second. This is not particularly fast – at this acceleration, it would take just under five minutes to go from 0 to 100. In practice, however, you accelerate much faster because it takes less time, as you can see in Fig. 32, 50 km/h corresponds to a hill of only about 9 m height, i.e. it is easier to sprint away than a high hill, and you are faster for it. (As you can see in Fig. 34, the acceleration distances are still long enough even with strenuous pedalling). So you often invest a multiple of the driving resistance, but only for a few seconds.

  • It is quite different when riding at constant speed on the flat. Of course, the drag power increases cubically with speed – but firstly, this is no longer true for the total drag (see Fig. 27), and secondly, the power does not vary so much here – between comfortable cruising and sporty riding there is perhaps 30% more speed, which then corresponds to just twice as much power (see Fig. 23). On the other hand, this is not a sprint of a few seconds, but the power is delivered over minutes to hours.

  • The human body is not so easy to calculate (and is also individually different). But at least the concept of Normalized Power could give a hint. This describes how strenuous a certain average performance is perceived to be. Here, a moving average of pedalling power over 30 seconds is calculated and taken to the fourth power. Low and high power do not cancel each other out, but strong deviations from the average power in both directions are clearly noticeable.

In summary, this means that although losses increase disproportionately with speed, both on inclines and accelerations the power required can multiply at the same speed. However, the body reacts very sensitively to power fluctuations – at least over minutes, not seconds. Therefore, you have to find a compromise: On the one hand, don’t let the speed drop too much, on the other hand, keep the averaged power constant.

  • Uphill, this means sprinting away short ramps and recovering on the flat – but taking long climbs comfortably. That’s why dolphining is so efficient on small hills: you don’t exhaust yourself because the average power is not excessive, but your speed still doesn’t plummet.

  • When accelerating: Get up to cruising speed quickly, and then ride along comfortably. However, while you can easily sprint to top speed in a few seconds on a racing bike, the acceleration processes are much longer on a velomobile because of the higher top speed and thus much higher kinetic energy. Therefore, you should accelerate quickly as long as the driving resistance is still low – but not immediately up to the final speed, but divide the forces. Otherwise, routes with frequent braking and acceleration will be very strenuous, but hardly faster. This is not to say that a velomobile is fundamentally slow in frequent stops in city traffic – as you can see in Fig. 34, accelerating to the same speed takes about the same time as a road bike. But higher speeds are not achievable due to lack of free track.

  • And on the flat, it doesn’t do much good to be maximally fast if you lack the grains elsewhere. It’s better to rest here. Or slowly build up additional speed to be able to ride the next climb with more momentum.

  • Last but not least, the average speed is the time average, i.e. slow sections have more weight than fast sections.

So you don’t ride fast by having a high number on the speedometer as often as possible – but a very low number as seldom as possible.

Does tilting technology for velomobiles make sense?

With tilting technology you can drive faster through bends they say. But that’s only partially true, whether the vehicle is traveling up or down through a curve usually does not change the possible speed at all (if the position of the cente of gravity is the same).

The first limiting factor in cornering speed is losing grip. In addition to the surface quality of the road, this mainly depends on the tyre material – a good tyre should roll easily, but have a lot of grip in the corners. There there are actually models that use different rubber compound in the middle of the tread than on the edge, so that you drive straight on the smooth-running rubber, but when you corner, you are on the better-adhering rubber. Of course, this effect cannot be exploited in a multi-lane vehicle without tilting technology.

The second limiting factor is tipping over. And that mainly depends on the height of the cente of gravity (as low as possible) and its distance from the outside wheel (as large as possible). With tilting technology you can of course move the cente of gravity towards the inside of the curve, thus increasing its distance from the wheel on the outside of the curve. But you can also do this by simply leaning towards the inside of the curve – provided you have a seating position with sufficient lateral support where possible.

So tilting technology would actually have certain advantages, but also higher weight, greater technical complexity, and that also needs space. You can shift the cente of gravity laterally with tilting technology, but you will probably not get it as low as without tilting technology. Even if you don’t lean the whole vehicle, but only the heaviest part – the driver. And then you also need an energy source that performs the inclination in the opposite direction to the centrifugal force – if you do it by hand, you can immediately do without the mechanics and simply lean your body inwards. There is something like a passive tilting technique, but in principle the vehicle is suspended on the chassis like a swing – yes, it tilts in the curve of its own accord, but that only prevents you from slipping on the seat, you can’t drive faster through the curve, but on the contrary, only slower (because the cente of gravity swings outwards instead of inwards).

And last but not least there are very few tight bends in practice that could be driven at high speed, real curves are usually wide enough, and with narrow junctions you rarely have the overview to drive through without slowing down. For these few special cases and a few seconds saved, complex mechanics are not worthwhile.

Measurement and Optimization

How to measure rolling resistance?

  • In order to minimise air resistance, the test must be carried out at low speed.

  • To rule out uneven pedalling, a rolling test is recommended. (See also: How do you know if tracking is set correctly?)

  • To rule out environmental influences, it should be a windless track with perfectly even asphalt and no bends.

  • The height energy lost then corresponds to the energy spent on rolling resistance – the lower this is, the greater the distance rolled.

How can I measure air resistance?

  • In order for air resistance to dominate, the test must be carried out at as high a speed as possible, for example, 70 km/h.

  • To exclude the uneven pedalling power, a rolling test is recommended.

  • Since at high speed the acceleration phase costs a lot of energy and therefore takes a long time (see: Why are the acceleration phases so long in a velomobile?) and thus rolling resistance plays a major role during this time, the beginning of the rolling distance should be as steep as possible in order to reach the measured speed as quickly as possible. For 70 km/h, 20 metres of altitude would be necessary.

  • The following measuring section should then be a constant downhill, where the velomobile rolls at a constant speed, for 70 km/h this would be just under 2% downhill (see: Fig. 15).

  • At the end, there is then ideally a steep counter-climb where you come to a stop quickly, the difference in height between the start and end position then corresponds for the most part to the energy that was expended for air resistance.

  • The cross-sectional area of a velomobile can be drawn by projecting it with a light source that is as parallel as possible (= as far away as possible, e.g. sunlight), and then e.g. cutting out the drawing paper and weighing it to determine the area. Thus one can finally determine the \(c_\text{W}\) value from the air resistance.

How to measure drivetrain losses?

Since it is about the difference between pedalling power and riding power, a rolling test is unsuitable, but here the pedalling power must be recorded with a power meter.

  • In order to minimise the influence of air and rolling resistance, one rides up a steep hill, then the necessary climbing power is much higher than these losses.

  • In addition, one precisely determines the difference in altitude between start and finish as well as the total weight with rider.

  • The pedalling power varies a lot, but added up over the whole ride it should correspond quite well to the climbing work expended plus the riding resistances. Therefore, one can simply multiply the average power by the riding time.

  • While air resistance is virtually completely negligible at these low speeds, rolling resistance is not, it is only a few per cent of the total resistance, but since drive losses are of a similar order of magnitude, one must know the rolling resistance coefficient at least reasonably well. Since the rolling resistance force is largely constant (except for dynamic rolling resistance), the speed does not necessarily have to be constant in order to calculate the rolling resistance losses simply using the average speed.

  • The difference between the measured muscle work and the sum of rolling resistance and the climbing work calculated from weight and height difference are then the propulsion losses.

Can air and rolling resistance be measured at the same time?

There are methods that use regression analysis to search for parameters that describe the measured data well. A particularly interesting method is Robert Chung’s <https://wallace78tria.files.wordpress.com/2013/02/indirect-cda.pdf>`__ method, in which a hilly circuit is run several times at different speeds. It must be as windless as possible (however, on a circuit the wind eventually comes from all sides), and one must not brake. Since it is a circuit, an exact measurement of the absolute altitude is not so important either – even if the altimeter drifts slowly, you know that the lowest and the highest points of the track must always be at the same altitude, and accordingly you can adjust the parameters for rolling resitance and air resistance so that the calculated net altitude gain becomes zero. (Speed-dependent errors in the altitude measurement, as seen in Fig. 18, on the other hand, are very much a problem, in this case one should resort to external altitude data). The evaluation is supported, for example, by the software GoldenCheetah.

Is there a calculator for air and rolling resistance?

Yes, even several:

  • Kreuzotter calculator: Calculates the final speed or pedalling power from numerous parameters, and includes presets for various upright bikes, recumbents and velomobiles. This variant of the calculator even allows free input of rolling and drag coefficients.

  • HPV Speed Simulator: This calculator is similar to Kreuzotter, but also graphically displays the increase in speed, so you can also see how long it takes to reach approximate final speed.

  • Power Analysis Graph: Here you can upload a file with recorded power data, and get a graph showing the different driving resistances at any given time (see Fig. 29).

  • Velomobile simulator: Here you can enter the curves of the altitude profile and pedalling power in a diagram and calculate the speed of a velomobile from this.

On the respective pages there are also sample values for air and rolling resistance coefficients. These were mostly obtained from performance measurement data, so the calculations from them are reasonably realistic.

Power analysis diagram

Fig. 29 The power analysis diagram that calculates and displays the rolling resistance, air resistance and other power data for a recorded ride with altitude profile and power data.

Ergonomics

What are the limiting factors in size? What do you have to watch out for?

  • Leg length: Since the velomobile tapers at the front, the feet rub against the bodywork if the legs are too long. The knees can also be a problem.

  • Upper body length: small people can not look out over the nose, and tall people do not fit under the hood.

  • Shoulder width: in rare cases, a velomobile is too narrow at the shoulders.

  • Thigh width: with large thighs you no longer fit between the wheel wells. Since you have to move your legs, absolutely nothing should rub on the bodywork here.

  • Foot length: people with large feet usually have long legs and may also want to use longer cranks, overall, this ensures that the feet rub against the side of the bodywork and/or below and above.

  • Preferences: e.g. flat or upright sitting position

  • Technical features: For example, for riders with short legs, where the bottom bracket is far back, the chain often runs very steeply down to the idler pulley. This can lead to problems with the front derailleur, which – depending on the size of the chainrings – cannot be adjusted far enough.

What is the ideal sitting position?

Basically, it has to feel pleasant – some people prefer a steep sitting position, others prefer a flat one. Upper body opening angle (i.e. the angle between the upper body and legs) must be correct, many people can exert more pedaling power when sitting upright.

Compared to an unclad recumbent bike the bottom bracket height in a velomobile is rather low. While with the recumbent bike the legs are in front of the body, reducing cross-sectional area would not work in a velomobile – you still have to look over the hood, so the legs are lower. In addition, the low aerodynamic drag of a velomobile does not primarily result from a small cross-sectional area, but from a low drag coefficient.

What is the impact of a short crank?

A short crank means that the lever that turns the chainring is shorter – at the same pedal frequency thus requires more power and less distance (= less circumference of the pedal circuit), the pedal speed drops. Conversely: To pedal with the same force and pedal speed, you have to drive in a lower gear than with a long crank. Since the pedal circumference is smaller, the same pedal speed means that the cadence is higher.

You also have to push the bottom bracket forward so that the leg is stretched equally when the crank is in the front. The leg is then angled significantly less in the opposite crank position.

Why do recumbent riders often use short Cranks?

Of course, the crank length depends strongly on personal preference and habits, and of course also on anatomy – longer legs need longer cranks. Apart from that, shorter cranks have proven themselves because the seating position is more fixed than on an upright bike (i.e. there is no out of the saddle) and at the same time significantly higher pedalling forces are possible (i.e. you can not only push on the pedal with your own weight but also push against the seat). This leads to high forces with strongly bent legs, which can cause knee problems. On upright bikes, on the other hand, it is rather unusual to stay completely in the saddle when pedalling powerfully – but when you pedal standing up, at top dead centre your leg is hardly bent at all. With short cranks on the recumbent, the knee is less bent, i.e. experiences the highest load in a position where there are less shear forces.

In velomobiles, however, there is also a purely geometric reason for short cranks: with long legs, the bottom bracket is far forward – i.e. where the body is narrower and lower than further back. At the same time, tall riders usually have large feet. In order to still be able to pedal without bumping your feet on the top and bottom, the cranks have to be shorter.

On an open recumbent, on the other hand, there are aerodynamic reasons: if the legs extend less thanks to short cranks, then less air is swirled.

How do I shorten cranks?

Most pedal cranks have a length of approx. 170 mm, shorter than 165 mm is usually difficult to obtain. That’s why you have to shorten yourself or have the cranks shortened.

In itself, it’s pretty easy if you have the right tool: drill a hole and cut a thread. Since pedal thread has a diameter of a good 14 mm, you cannot simply place a second hole next to the existing one at a closer distance – because the new hole must be surrounded by sufficient material, it is advisable to use a crank that is 20 mm longer than the desired length.

Solid aluminum cranks can of course be shortened without any problems. But also hollow-forged cranks are usually not a problem – you only drill the edge of the cavity, where there is still enough wall thickness for the thread. An exception are rotor cranks, for example, because they contain several cavities next to each other – among other things, where the thread would be on the side. Also carbon cranks cannot be shortened eaasily, since the thread cannot be cut directly into the carbon (which would mean point loads much too high), but is done via metal inlays.

What is the Q factor and why is it best to make it small on a velomobile?

The Q factor is the width between the cranks, i.e. the lateral distance between the pedals. It is determined by the width of the bottom bracket, the width of the crank arms, and their offset.

In the case of an upright bicycle with wide tyres, the chainstays must offer enough space, therefore the Q factor must not be too small. Accordingly, the cranks are offset to get past the wide chain stays at the rear. But the chainrings can also come into contact with the chainstay if they sit too far inside, so the bottom bracket should not be too narrow there either.

With a velomobile, on the other hand, there are no chainstays at the bottom bracket, even the widest bottom bracket mast is even narrower than a bottom bracket. There is therefore no reason for a large Q factor. On the contrary, with a tall rider (= long legs, big feet) the feet can easily touch the bodywork if they are too far apart. For this reason, attempts are being made to keep the Q factor small – for example, by shortening the bottom bracket axis a little and using cranks that don’t have much offset.

And last but not least, it is also more pleasant for many drivers to be able to put their feet closer and keep their legs parallel.

High or low cadence?

Depending on personal preference and anatomy, but also habit. Both have advantages and disadvantages, since low cadence means high strength and vice versa.

It is often said that too high a pedaling force would break the knee joint. This is only partially true, because a trained knee can easily withstand great forces. But you shouldn’t go untrained on long tours using high strength, but with a slower approach. The direction of movement also plays a role: a joint can withstand very high forces if it is only subjected to pressure. If it still has to move, complications are much more likely – therefore a pushing movement with short-term high strength should be much less critical than a grinding movement with constantly high strength.

But even a high cadence is not necessarily a problem. In order for a knee to be free of symptoms, well-trained muscles that hold the joint in position – and muscles only develop under load, and not if you only ever crank there without force. It also needs a certain alternating load because articular cartilage has fluid and therefore nutrients that pump into the interior.

The cadence also has an effect on technology: a bike with a very responsive suspension bounces more when the cadence is low and uneven – because there the mass of the legs is accelerated and braked more. Accordingly, more energy is lost into the suspension damping. But you can change the pedaling force faster than the cadence, those who pedal at a high frequency need a better-tuned gearshift because they cannot jump gears when accelerating as easily as a powerful driver.

Uphill the cadence gets lower for most people. This also makes sense because the pedal resistance is higher there, i.e. in the dead cente speed drops more – and you need more strength behind the dead cente. In contrast, the load is more even in the plain, it would be counterproductive to exert a high shear load on the knees with a very slow and grinding movement over the entire pedal rotation.

Everyday use

How does a velomobile drive?

Good question – in any case, clearly different from other bicycles:

  • The speed is higher, so you have to ride much more carefully.

  • Acceleration phases are much longer, so you have to manage your power well and ride with foresight (see Why are the acceleration phases so long in a velomobile? and Fig. 34).

  • Distances seem shorter because you usually ride faster.

  • Luggage is less critical – you don’t have to carry extra bags or straps and you don’t have to pack your luggage in a special way, you can just carry it clean and dry inside (see: How much luggage can I accommodate?).

  • You hardly notice the wind, because headwinds slow you down much less, you can see a slight difference on the speedometer, but you don’t have the feeling that you’re not making any progress at all. And if you ride with the bonnet on, you don’t feel the wind at all, but at most notice the influence of gusts on your driving behaviour (see: How does wind make itself felt?).

  • Cold is no problem, because even in the depths of winter you can still ride with a T-shirt and thin trousers (see: Which clothes?).

  • Precipitation is not a problem either, you don’t stay completely dry (on the contrary), but you don’t get soaked to the skin by a downpour, and you don’t freeze either. A ride in the rain is not particularly pleasant, but it is not particularly daunting either (see: Can I ride in the rain?).

  • Ventilation is very important, at low temperatures because the windscreen fogs up, and at higher temperatures because the cooling and drying effect of the airstream is almost completely absent. Even with good ventilation you are actually always sweaty (see: What about ventilation?).

  • Small hills seem flatter because you can ride them with momentum for the most part (see: can I ride in hilly countryside?).

  • Long hills seem steeper because, firstly, you are lugging more weight around, secondly, the power requirement is high but the cooling is poor, and thirdly, you are spoilt by the other high speed (see: can I ride in hilly countryside?).

  • Fast tyres are much more important because rolling resistance is a much bigger part (see: Why are good tyres so important on a velomobile?).

  • Good grip is less important because velomobiles are multi-trackers – they don’t tip over immediately when it’s slippery. You can risk slipping a bit instead of constantly riding like on raw eggs (see: Do you need winter tyres or studs?).

  • Weight is more noticeable because air resistance, where weight is not a factor, is lower (see: Why are velomobile riders so obsessed with low weight?).

  • You can deliver your power much more evenly because the losses are lower, i.e. you get back more of the power you put in, e.g. you can accelerate before a climb and then ride it with more momentum, or coast for an extremely long time after a descent. In the end, you have many more rolling phases than with a normal bike.

How much luggage can I accommodate?

That depends very much on the respective velomobile. Basically bulky items are often problematic. But otherwise the luggage volume is surprisingly large, even in the smallest velomobiles you can accommodate at least as much as a touring bike can fit in two large saddlebags plus on the luggage rack. So touring is not a problem at all. Larger velomobiles even have the same luggage volume as a smaller bicycle trailer, and thus space for large purchases.

Can a trailer be used?

Basically yes, if it has a low drawbar. If you look at the back of the bodywork and it can be strengthened, you can bolt on a standard trailer hitch. But this only makes sense in exceptional cases, because in the velomobile you can accommodate quite a lot of luggage, and a trailer is heavy, bulky, and significantly increases air resistance. So you can’t drive fast, and uphill you not only have to pull the weight of the trailer and the load, but also the heavy velomobile. Therefore, it usually makes more sense to pull the trailer with a normal bike.

Which clothes?

Similar to upright bikes but less. Even in deepest winter you can still drive with a T-shirt, then you have to have warm clothes with you, otherwise a breakdown will quickly make you very cold. Overall, a thin shirt, short or long trousers, and possibly a scarf or a neck warmer are usually sufficient. You can definitely leave windbreakers, sleeves and gaiters at home.

Do you need click shoes?

Actually yes. Firstly, because on a recumbent bike the foot pushes on the pedal from behind instead of from above i.e. it is not already pressed on the pedal by gravity. Second, because otherwise you can slip off the pedal. While that on an upright bike often has only minor consequences, in the velomobile there is almost no space next to the feet – if the foot is not sitting correctly on the pedal, you bump into something, and if you slip off while pedaling hard, you can even damage the bodywork on thin-walled velomobiles. And thirdly, efficiency is also somewhat higher if you simply don’t have to worry about slipping, but can always rely on a firm hold on the pedal.

How big is the turning circle?

About 10 meters, about the same as a car.

That sounds like a lot for a bike (it is too), but since you mainly drive on roads for cars, you almost never have a problem in practice. On known routes you can often adjust to problem areas and make sufficient swings in time, only on unknown routes and then typically on bike paths, it can happen that you occasionally have to maneuver to get out of a trouble spot. But that happens too rarely to be a real problem.

The exact turning circle depends on several factors:

  • Velomobil type: the narrower the wheel well, the larger the turning circle.

  • Velomobiles with closed wheel wells usually have a slightly larger turning circle.

  • Tyre width: the wider the tyre, the smaller the possible steering angle. First, the tyre hits the side of the wheel well faster, second, the tyre is also taller and protrudes further forward / backward.

  • Wheel lacing: in velomobiles with closed wheel wells, the turning circle is smallest when the wheel is in the cente of the wheel well. You can also dress up open wheel wells with so-called pants, this increases the turning circle significantly and you have to countersteer on many curves beforehand. However, if you re-spoke the wheels so that they no longer sit flush with the open wheel well, but in the middle of the well, the turning circle is similar to that with open wheel wells.

How do you go backwards?

Bicycles usually do not have a reverse gear. You don’t normally need it either, because the turning circle is small and otherwise you get off and push very quickly. Neither is the case with the velomobile, getting out is more cumbersome, especially if you have to remove the hood or foam cover, and the turning circle is similar to that of a car.

In a car, you need to go into reverse to park or to turn it several times. You don’t need the former with the Velomobile: you just push it out of the parking space. And you rarely have to turn, especially not on familiar routes. It may only be necessary on cycle paths with hairpin bends on a regular basis. But even then you can often do without a reverse gear:

  • Hardly any street is really flat, and certainly not adjacent entrances, you can often drive uphill there, let yourself roll back, and then get around the curve. A barely perceptible slope is sufficient for this.

  • With velomobiles with open wheel wells, you can at least do without getting out and turn the front wheels with your hands like a wheelchair. But you get your hands dirty, and it only works on a really flat roads, not uphill.

  • And then of course there are velomobiles with two foot holes where you can push back with the Fred Flintstone technique.

What about ventilation?

In the velomobile you are shielded from the wind, i.e. if not driving during winter, the body needs cooling – especially on the thorax and head. Another point is humidity, on the body this does not necessarily matter, but if it is too moist, you can irritate yourself in the crotch, for example. With the hood on it mists up the windows at cool temperatures. With a fresh air supply you can blow away the humid air.

Most velomobiles have an air inlet at the front – this is the stagnation point where there is increased pressure, so you can achieve a large air throughput with a small opening. In addition, a hole there does not disturb the air flow along the body.

On some models there is also the headlight cut-out – this has the advantage that no further holes have to be drilled in the bodywork, and the LEDs are well cooled by the wind, which increases efficiency. In return, the headlamp sits quite low, and lighting of the road is correspondingly poor (only a small proportion is scattered back).

In some Velomobile models, the bottom bracket mast is attached to the front of the air intake. This has the advantage that the bottom bracket mast can also be supported from the front, and the air flows through the mast backwards and the feet are not cooled by the cold fresh air.

If this cooling is not sufficient, you can also install a NACA duct duct usually in the front and in the top of the bodyork (see: What is a NACA duct?).

Anyone who drives with a hood can also drive with a raised visor, a narrow opening is usually sufficient there for sufficient air flow.

In addition, some drivers have drilled holes in the seat so that the back is better ventilated.

With all these measures, however, it is important that not only additional air is fed into the velomobile, but that it can also flow out well (e.g. via vent holes at the rear) – otherwise an additional vent does not bring additional air, but lowers the air flow through the other air intakes.

Can I ride in the rain?

Rain is not as annoying in a velomobile as on an open bike, but you don’t stay dry either. You are protected from street dirt and splash water thanks to wheel wells and you are not soaked to the bone by a cloudburst, but due to the high humidity inside, no sweat is removed, and a partly open visor or without a hood the head and upper body get wet.

In addition, raindrops obstruct the view – regardless of whether you are driving with a hood or in convertible mode with glasses. In the former case, it helps to have windshield wipers with which at least the large drops can be distributed so that the view is tolerable.

While there are no real windshield wipers, at least the following solutions have proven their worth:

  • Magnetic wiper: Two elongated neodymium magnets, wrapped with absorbent material, stuck together on both sides of the pane. A viewing window can be wiped freely from the inside – that’s enough because you sit close to the window and therefore don’t have to keep large areas clear. In addition, the magnets provide contact pressure so that even large snowflakes can be wiped away.

  • Wiper: There is also a manual windscreen wiper which is screwed into the visor; the elastic wiper arm presses the wiper blade reliably against the double-curved visor.

  • Thread wiper: an elastic cord is stretched lengthways or across the visor and can be pulled back and forth perpendicular to it with a second cord. This means that larger areas can be wiped than with a magnetic wiper, but the contact pressure is lower.

Driving at night and in rain is still not fun, because first of all the oncoming drivers headlights reflect in the raindrops on the windshield or your glasses, and secondly, the often low to the ground attached velomobile headlights only dimly illuminate a wet road.

What to do against fogged up windows?

What is the cause of fogged windows? The driver sweats and the warm, moist air condenses on cold surfaces, such as the windscreen.

The remedy is accordingly:

  • Insulate the pane so that it is not cold on the inside. That works for example with the Pinlock visor, with which you can almost achieve double glazing.

  • Keep the moisture away from the window. You can do this by ventilating the velomobile well, if fresh air flows in continuously from the front, the moist air is pushed out to the rear.

  • You can also open the visor a little, and place a small barrier in front of the lower edge of the visor, which deflects the incoming air upwards, over the inside of the visor. This air film prevents the moist indoor air from coming into contact with the visor.

  • And if none of this helps – for example because you are not moving and therefore have no air flow – you have to wipe the inside of the window.

Then why is that not a problem in cars?

  • You sit passively in the car, so you don’t sweat. The humidity is correspondingly lower.

  • In the car you sit much further away from the windscreen. So you don’t breathe directly against the window.

  • The interior of the car has a much larger volume. Even if you were sweating, it would take much longer for all of the air to become damp.

  • The car has a heater. An internal combustion engine is inefficient, only about 20% of the energy goes into the drive, the rest is heat – and with these thousands of Watts, the interior can be heated without any problems, and fresh air can also be heated and directed to the window.

Is a velomobile suitable as a car replacement?

Depends. It is and remains a bike that only offers a slightly higher speed and range and a little better weather protection. A car, on the other hand, is a “one size fits all”, you can take several people or large amounts of luggage with you, drive hundreds of kilometers through very mountainous terrain, but also only drive a short distance alone into town for shopping. A bicycle is of course significantly more restricted. But that also means that a car is completely oversized for most everyday tasks, this can be done excellently with the Velomobile. If this is the case on most days, you can do without a car – and still rent a car that fits your needs for the few remaining transport tasks.

You can drive without being sweaty on arrival?

Difficult. Because inside the airflow is missing and you have to drive very slowly to avoid sweating down your back. If you are not going uphill, a pedelec engine is not much help, because you practically always drive faster than the maximum speed of the engine.

How does a velomobile drive in winter?

Not much different from summer, you may need long trousers instead of shorts and a scarf against the cold wind, and with a hood a Pinlock visor against fogging – but otherwise everything is like summer, you can usually drive in a T-shirt (but you should have a warm jacket with you if you have to get out in the event of a breakdown at freezing temperatures and tinker for longer).

In addition, good lighting is even more important in winter than in summer – because drivers expect fewer cyclists in winter. Semi-puncture-proof tyres are also recommended, first, it’s not fun to have to patch a flat in the dark and cold, and secondly, the cold also makes the rubber stiffer, so just changing the tyres becomes very tedious. However, the rolling resistance also increases noticeably in the cold – that’s why easy-running tyres are not unimportant, especially in winter.

Do you need winter tyres or studs?

Not really, at least not on a flat route. Spikes on the front wheels prevent from slipping and shorten the braking distance, and tread on the rear wheel ensures that the rear wheel is less likely to spin in snow. The latter is actually only a problem when starting off and with steep climbs, so only in a few places. And since a multi-track does not fall over on black ice, it is not worth installing spikes due to a few smooth bends – spikes, like tread tyres, significantly increase rolling resistance so that high speeds can no longer be reached even on a straight road. It is better to drive carefully in curves and make slow steering movements, and calculate that the velomobile can slip a bit – which is normally completely harmless. Only steep, winding descents are critical on black ice, because then you may not be able to steer or brake. But as long as you travel on cleared and clean streets, there are generally few problems – and if it is very slippery, then the car traffic also slides.

How do people behave?

Basically, velomobiles are received very positively, there is almost exclusively positive interest or comments – but you always draw everyone’s attention, more than with the most expensive automobile. It’s rarely a problem on the go, children point at the velomobile, and drivers often drive slowly alongside, marvel at the vehicle, crank down the window, make a comment and/or film or take pictures with their smartphone.

Interest, however, can become a real problem when parking. People don’t just have to look at everything up close or touch it carefully, damage is not uncommon. Children seem magical mirrors and other exposed parts, it is not uncommon for the mirror to be twisted, maybe even broken off. If the velomobile was open, footprints can be found inside because, for example, parents have put their children in without having asked.

But it gets annoying when something breaks. Even if the Velomobile foam lid or the hood is closed, it can be partially torn open violently, so that a Velcro strip on the foam lid was torn off or a crack appeared in the carbon on the hood. It is simply incomprehensible how naturally some people tear off the foam lid as they walk past so that they can take a look inside. But also the bodywork is at risk. Again and again there is a crack in the carbon at the front because someone has obviously sat on it. While none of this affects functionality and security, it is very annoying and also very expensive. So it’s best not to park the velomobile where there are too many people with too much time and very bad manners.

How to lock and secure?

Like every bike: connect to a fixed object. Due to the lack of triangles, this is only possible on the rear wheel – or on the stern handle or carrying eye, if present. While that’s not as good as a normal bike where you can put a lock around the stable main frame – however, a velomobile is neither handy to steal nor inconspicuous, i.e. not for casual thieves. In addition, the removal of the rear wheel on some models requires special knowledge, and sometimes also special tools.

There is a bigger problem – vandalism. The best remedy here is a tarpaulin in the most inconspicuous color possible – while an open velomobile attracts everyone’s attention, a covered vehicle is largely ignored.

How big does a storage room have to be?

See Fig. 30, here the dimensions of many velomobiles are recorded. The height is usually just under a metre, and in terms of width, most velomobiles fit through normal doors.

Scatter diagram VM dimensions

Fig. 30 Length and width of many velomobiles, the colour indicates the weight – green is 20 kg, red is 40 kg. Note that the weight can vary greatly depending on the equipment, and e.g. with the WAW, different front and rear sections are available, so the length can be 269-301 cm. Leitra, Velayo and DuoQuest are outside the diagram.

Can I take a velomobile on the train?

In principle it is possible, but it is not without problems – velomobiles are large and bulky, unsuitable for normal bicycle racks, and cannot be hung on hooks. So they can block half the bicycle compartment or the boarding area, and the train employees then rightly refuse to take them along. In principle, you don’t have a right to this, often even special bikes (tandems, cargo bikes) are not allowed in the conditions of carriage – in the end, it’s always up to the discretion of the staff.

But with a few measures you can significantly increase the likelihood of being taken along:

  • Choose a route where you rarely have to change trains, and if you do, choose the transfer times very generously.

  • Find out what type of carriage is used on the route. A bicycle compartment must be available, well-suited are, for example, double-decker carriages with a deep entrance or driving trailers with bicycle compartments, as these also have wide doors. Old carriages with narrow doors and steep stairs, on the other hand, are unsuitable.

  • Do not board or disembark en route, but at the starting or terminus station. Then there is more time to get on or off, and the train is not yet so full or is already a little emptier.

  • Arrive at the station early and be in the right place on the right platform.

  • If necessary, organise help in good time so that you do not take longer than other passengers to get on or off the train.

  • Travel when there is little traffic. This means not during rush hour, but e.g. very early or late on weekends or public holidays, or on routes with a lot of leisure cycling traffic, preferably in the off-season or during the week.

  • Possibly buy two cycle tickets at the same time. This is not explicitly required, but avoids discussions if you also need the space of two bicycles.

  • Take into account that the train may be surprisingly full and you have to take the next one.

Ultimately, there is no guarantee that you will find space on the train or that the conductor will take you. But if you don’t cause any problems (don’t block a path, don’t cause delays, don’t disturb fellow passengers, don’t make extra work for the staff and don’t violate any safety regulations) and are friendly, it usually works out quite smoothly – then the railway employees have no reason to complain. And if you are so quick that you have created a fait accompli before the staff can react, they usually don’t feel like starting another unnecessary discussion afterwards.

Route selection

Can I ride in the mountains?

If you have a suitable gear ratio, this is no problem. But you need a different technique than on the upright bike: do not pedal hard and out of the saddle, but crank uphill in a low gear – slowly but steadily.

Nevertheless, a velomobile shows its strengths especially in the lowlands:

  • Uphill aerodynamics have no advantages.

  • Downhill, the air resistance brakes much less, so the brakes overheat much faster (see Fig. 15).

  • The higher weight costs more power uphill (see Fig. 31)and puts more strain on the brakes downhill.

  • Uphill there is no cooling without a headwind, that makes mountain trips in the summer heat very tedious.

Power at different uphill speeds

Fig. 31 Power demand at different gradients, the solid line is a velomobile, the dashed line a road bike. As you can see, velomobiles are increasingly at a disadvantage at low speeds as the gradient increases – or rather, they have to go faster and faster as the gradient increases in order to have an advantage over a racing bike. While at a 1% gradient a velomobile is already at an advantage from about 17 km/h and 100 W pedalling power, this point shifts to over 30 km/h and over 500 W pedalling power at a 5% gradient. With 200 W pedalling power, on the other hand, a velomobile is only as fast at a 4% gradient as a racing bike would be at a 5% gradient.

Can I ride in hilly countryside?

In contrast to high hills, short hills work much more smoothly because, firstly, you can pedal with a lot of power and/or a high output for a short time – it doesn’t matter so much if the gears aren’t suitable.

Secondly, the high speed of a velomobile means that you have a lot of kinetic energy (see Fig. 32) – so you can take small hills in your stride without getting bogged down. This is called dolphining.

However, all this only applies to main roads that are designed for speedy riding. Pass roads in the high mountains, for example, are surprisingly easy to cycle on because the gradient is very constant and the roads are very well adapted to the landscape profile. In contrast, in hilly country, especially on secondary roads, there are often also very steep (albeit short) sections that are poorly adapted to the landscape and that you therefore cannot ride with constant power. A series of such climbs is often surprisingly exhausting – with any bike, but especially with the velomobile it makes itself felt negatively if you can’t take advantage of any momentum.

kinetic energy and vertical metres

Fig. 32 Conversion of kinetic energy into altitude metres. So at speed 50 you have enough energy to roll up a hill about 9 m high. On the one hand, you have to subtract the driving resistance from this, on the other hand you can pedal along. So the exact balance then depends on the gradient, for steep hills the diagram is quite correct, while on a flat hill you can pedal along for longer and thus get higher.

Velomobile users and topography

Fig. 33 Map of maximum gradients (= total differential of topography) and velomobile riders (from data provided by Velomobiel and InterCityBike and members of the Velomobile Forum). It can be seen that only in the Alps and parts of the low mountain ranges there are steep gradients, e.g. the foothills of the Alps and the southwestern German strata are rather flat, except for the northwestern sides (e.g. Albtrauf) and deeply cut valleys (e.g. Jagst, Kocher, Altmühl). Accordingly, velomobile riders can also be found in the hilly country – the distribution corresponds more to population density than topography.

Can I ride on bike paths?

Good question – the quality of bike paths is far too different to make a general statement about it, but the following generally applies:

  • Cycle paths are not designed for velomobile speeds, even with a racing bike you reach your limits.

  • Cycle paths often have tight curves and are very confusing, even if you could sometimes drive fast, you would often have to brake.

  • Junctions are often confusing and blocked by stopped cars and cyclists are not often given the right of way.

  • Cycle paths are often narrow, so that other cyclists can hardly be overtaken.

  • Cycle paths often have obstacles such as jostle bars or hairpin bends that cannot be passed with a velomobile.

  • Roots and other bumps are particularly uncomfortable with a multi-track with a tight chassis.

That is why it is usually not only more comfortable, but above all much safer to drive on the road. But there are also good, wide and clear inter-country cycle paths that can be used perfectly by a velomobile. However, you cannot rely on a minimum standard.

Can I drive on unpaved roads?

Walking speed does it, but it’s not fun. As mentioned, the rolling resistance plays a major role in the velomobile, this is much higher on unpaved roads, and accordingly you are much slower. In addition, a velomobile typically has narrow, hard-inflated tyres and a tight chassis, which makes driving less comfortable. A multi-track not only hops up and down, but also tilts around the longitudinal axis. And since a velomobile is a wonderful sound box, it rattles and booms quite well.

How about driving in city traffic?

In principle, you can also ride well in the city, however, a velomobile has hardly any advantages there.

The main advantage is speed, but since the acceleration distances are long (see Fig. 34) and you have to stop frequently in the city, you hardly reach a high speed. However, this does not mean that velomobiles are slower than other bicycles there – the acceleration is very similar, but the high final speed is only achievable if you can let it run for kilometres. In the city, a velomobile is therefore permanently in the acceleration or coasting phase.

As velomobiles are not that manoeuvrable, many cycle paths are unsuitable, either impassable because of obstacles, or too narrow to overtake other cyclists, or too confusing for higher speeds (see: Can I ride on bike paths?). The recumbent position is practically irrelevant: because the head is further back, you can see junctions later – but it is only a matter of fractions of a second, at higher speeds this time becomes shorter and shorter, but the braking distance longer and longer. Therefore, unclear cycle paths brake all bicycles to a similar speed, with the velomobile, only the difference to the normal cruising speed is greater. On the road, on the other hand, heavy traffic is often a problem, there you are stuck in traffic jams with the cars. On major thoroughfares, however, you can definitely benefit from a “green wave” for which you would be too slow with a normal bicycle.

Time and distance to reach the final speed

Fig. 34 Time and distance to reach 90% and 99% of the final speed with 300 W pedalling power. Both increase strongly with speed, that is why the acceleration phases are so long in the velomobile. In the lower speed range, all bicycles accelerate similarly well, a velomobile is minimally slower due to its higher mass and drive losses (upper diagram), but has made up for this shortfall at a good 30 km/h. The velomobile’s acceleration is also slower in the lower speed range. This is hardly noticeable in the acceleration section (lower diagram). After about half the time and a third of the distance, you are already close to the final speed, further acceleration is very slow.

How does it work on short distances?

A velomobile is rather unsuitable for short distances. The higher speed does not save any significant time, the weather protection is less important (you can hardly cool down in a few minutes, you can better close a gap between rain fronts), getting in and out takes longer, and you cannot carry a backpack, you have to Stow luggage and take it out again.

How does it go on long distances?

Outstanding. This is where the strengths of a velomobile come into play: great time savings thanks to high average speeds, good weather protection, large luggage capacity, no seat problems, no numb hands, no neck pain.

Are velomobiles only suitable for rural areas?

Rural areas with lots of open space and little traffic should be ideal for velomobiles, at least if the topography plays along (see Fig. 33). However, in Fig. 35 you can see that most velomobile owners do live in densely populated areas, maybe not necessarily in the middle of big cities, but at least on their outskirts or in the suburbs.

Velomobile owners and population density

Fig. 35 Velomobile users and population density (Eurostat, population density 2019 by NUTS 3 regions).

How do you drive as energy-efficiently as possible?

By not braking.

You can’t do anything about drive losses, rolling resistance and air resistance while driving, because they are always there. Or, by keeping your speed as constant as possible, you can reduce air resistance, which increases disproportionately with speed (see: How to achieve the highest possible average speed?). You get the energy from climbing back on a descent, the acceleration power when coasting (or by rolling uphill with momentum), but everything that is braked away is lost and has to be laboriously built up again.

The first thing that helps against this is a good route planning, so that you don’t have to brake if possible, and if you do, then let it roll as early as possible – you get surprisingly far, see Fig. 36.

kinetic energy and rolling distance

Fig. 36 Conversion of kinetic energy to rolling distance for different types of bicycles

How to plan a route for the velomobile?

If you want to travel as quickly and relaxed as possible, you have to be able to drive at a constant speed. That means:

  • clear thoroughfares instead of winding alleys or residential streets

  • Priority roads or roundabouts instead of stop signs

  • as few gradients as possible

  • Descents should be as flat as possible so that you don’t have to brake, but can always pedal along

You can tend to look at online route planners for cars instead of bicycles, the latter mostly lead over low-traffic routes, but are often very winding. In contrast you don’t need a high maximum speed, but as few braking operations as possible – each acceleration not only costs a lot of power, but also several 100 meters of distance, so that with frequent traffic lights you never get to high speed, but still exhaust yourself. It is better to choose a route where you can not drive quite as fast, but instead smoothly – instead of a single braking process that ruins the speed over several hundred meters.

BRouter is recommended as a route planner, this has, among other things, a routing profile for velomobiles and is completely configurable. The data comes from Openstreetmap, incorrect routing is therefore usually caused by incorrectly entered or incomplete Open Street map data. Fortunately, everyone can correct this themselves, until the changed data arrives in BRouter, however, it will take some time. The assumptions made by BRouter have generally proven themselves, however, they can be inappropriate in certain areas or in other countries.

If you want to plan the route by hand, you can still use the BRouter velomobile profile as a guide. This converts all obstacles into detours, For example, the specification uphillcost = 80 from the profile file means that one vertical meter uphill corresponds to a detour of 80 meters, and turncost = 150 means that a right-angled branch corresponds to a 150m detour.

Commuting

Is a velomobile suitable for commuting?

Depends on, for example, if you want to travel to work in less than an hour, you can travel up to 30 km in the lowlands – correspondingly less with mountains. Depending on the route, the velomobile is a few minutes faster than an upright bike, but above all much more comfortable (since it is less worth driving at full throttle everywhere), you can take more luggage with you and is also less dependent on the weather – the clothes are almost the same summer or winter, and even a cloudburst doesn’t get you completely soaked, you can continue driving.

You should always have a change clothes with you – unless the route is very short. Ideally, you can then take a shower at the workplace, otherwise you can get by with a reduced “cat wash”.

Which distance can be covered daily?

This depends a lot on the particular route, but the lists at Velomobiel.nl and Intercitybike can give a rough idea. Entering the kilometers is voluntary and not always up-to-date, but one can assume at least for larger mileages that it was not a short vacation trip, but that the kilometers have accumulated over several months. This gives fig_rijderslijst_dist, here one can see that faster velomobiles tend to be used for slightly larger distances.

Monthly travel distances of velomobiles from  Velomobiel.nl and Intercitybike

Fig. 37 Monthly travel distances of velomobiles from the manufacturers Velomobiel.nl and Intercitybike. Data entry is voluntary and thus not representative. Only velomobiles with a total mileage of at least 7500 km have been evaluated (as of: 02/2020).

Do you need a motor for a long commute?

To make the driving time on a long route bearable, it is often desired to install a motor. Whether this makes sense, however, depends very much on the route and the motor (see Motor yes or no?). Although a 45-km/h motor reduces the riding time in almost all cases, it is not so easy legally, at least in Germany – only a few models are even permitted and cycle paths may not be used (see Legal situation in Germany). A 25 km/h motor, on the other hand, is legally quite unproblematic. However, you usually ride a velomobile above 25 km/h, where the motor is of no use. Therefore, such a motor is only useful on long climbs, where you can save a lot of time. On frequent accelerations, a motor also helps, but only saves power – you don’t accelerate significantly faster, and accordingly you save almost no time.

Do you need a fast velomobile for a long commute?

Beginners looking for an all-weather bike for a longer commute are often advised to buy a fast velomobile – but these are very expensive new and hard to get second-hand. Beginners, on the other hand, rarely want to spend that kind of money and don’t intend to set speed records. Who is right?

The top speed is actually not that relevant in everyday life, and compared to normal bicycles, the aerodynamics of “slow” velomobiles are still excellent. In contrast, velomobiles trimmed for maximum speed often have other disadvantages, such as a very large turning circle, small suspension travel and low ground clearance, which makes them impractical to unsuitable for some routes. Moreover, even at high speeds, air resistance only accounts for a good half of the drag, and even less at the median speed in everyday life (see Fig. 23) – so it is hardly worthwhile to put a lot of money and effort into the final optimisation here.

But fast velomobiles not only have very good aerodynamics, they are also particularly efficient – i.e. they are relatively light and have a stiff drivetrain. You also benefit from both at low speeds: a low weight is not only beneficial uphill (more so than on normal bikes, see: Why are velomobile riders so obsessed with low weight?), but also reduces rolling resistance (see: Why are good tyres so important on a velomobile?) and kinetic energy (see: Why are the acceleration phases so long in a velomobile?). And a stiff drivetrain makes powerful acceleration more efficient, as needed when accelerating or going uphill (see: Why a stiff drivetrain?).

This means that on long and mountainous routes it is worthwhile to have a velomobile that is light and has a stiff drivetrain.

Repair and maintenance

Do you need a special workshop?

Not really. Velomobile technology is not rocket science, basically most things are easy to master with a little basic knowledge of bicycle technology. And since the vast majority of components come from the normal bicycle sector, a bicycle workshop should have no problem with most things. You don’t have to be a hobby mechanic to drive a velomobile. The fact that there are many velomobile drivers who are is probably due to their interest in bicycles and bicycle technology, and secondly, they often do high mileages and therefore save a lot of time/money if they can do routine work themselves.

Manufacturer support is only required for comparatively few defects, but even special parts (e.g. struts) are usually sent quickly and are easy to install. And if you have bought a somewhat older velomobile model whose teething troubles have now been eliminated, you should be able to do many tens of thousands of kilometers without any velomobile-specific spare parts.

However, the Velomobile is definitely more difficult for troubleshooting because many parts are difficult to see and you cannot turn the crank with one hand and watch the rear derailleur. It is therefore a thankless task to find the cause of some symptoms. A velomobile driver should therefore keep an eye and both ears open when something feels or sounds different and get to the bottom of things as soon as possible.

Wear parts, in contrast to normal bikes?

Basically, there is less wear and tear on the velomobile, the drive train in particular is very durable because it is protected from dirt. The chain lasts a very long time because the wear is also distributed over a much greater length. Also drum brake pads are very durable – not because they are not used very much, but because they are quite oversized. The short-lived parts are usually the tyres.

What are common maintenance tasks?

Most likely: check air pressure, inflate tyres, and examine the tyres (especially on the rear wheel) for damage (stones, cuts, damage to the carcass). And of course charge the light battery.

Occasionally the chain must be wiped and oiled and the brakes adjusted. If the latter no longer resets properly, you can place shims on the brakes (see What is it about placing shims on drum brakes?), Or replace the brake pads completely. And if necessary, the braking and Shift cables can be changed. And if the chain tubes are dirty, you put a thin piece of fabric around the chain and pull it gently through the tubes.

How long does carbon fiber last? How can I protect it?

Carbon fibers do not age, just the epoxy resin under the effects of UV light. But that’s just a cosmetic problem, if the top layer of resin changes color or decomposes, you can grind it down and apply a new thin layer of resin. A UV resistant lacquer, Gelcoat or Film can protect the epoxy resin.

How to repair carbon?

By gluing a patch over the breakage with epoxy. Important here is the :term:` Scarfing <scarfing>`, around the broken point the material is beveled in a V-shape and filled with several patches of increasing length. Firstly, this increases the adhesive surface and secondly, the overall thickness remains constant, which means that there are no stress peaks at the transitions; the respective edges of patches and original material are thin and therefore flexible and adapt to deformations of the other side instead of seperating. Since carbon fibers can only absorb tensile forces (see: `Why are velomobiles made of carbon or glass fiber?`_), the fibers of the patches must be aligned in the direction of the tensile forces and lie as flat as possible. A well-made carbon repair is neither heavier nor less strong than the original material.

What is twill? What is a roving?

A carbon fibre is only a few micrometres thin, that is why fibre bundles of several thousand fibres are usually used. This is called roving.

In order to be able to produce flat components from the fibres, they are usually woven into a fabric. Various weave types are possible here. In plain weave, a weft thread runs alternately over and under a warp thread, shifted by one warp thread for each weft thread. This results in a quite strong fabric, but it is difficult to drape over curved shapes. For this reason, the twill weave is usually used for carbon, here the threads cross over less frequently than in the plain weave, which makes for better drapeability, and also for the typical carbon look.

In a woven fabric, the rovings support each other through the crossing points, this reduces the sensitivity to puncture stresses. This is why carbon fabric (with its typical carbon look) is usually used on the surface of components. In contrast, uni- or bi-directional fibres are often used in deeper layers of a heavily loaded component. Since these are not interwoven, the fibres lie straight instead of wavy, which provides more stiffness – under tensile load, the fibres are not first pulled straight, but already are.

Why shouldn’t aluminum parts be glued directly on carbon?

Because of the electrochemical voltage effect. Both carbon and aluminum conduct electricity and there is also sufficient moisture in the velomobile, Carbon has an electrochemical potential of +0.75 V (compared to the normal hydrogen electrode), aluminum of -1.66 V – making aluminum the less noble material. If a current can flow between the two materials, it beccomes contact corrosion, Carbon serves as the cathode and aluminum as the anode, it is oxidized to aluminum oxide – it slowly dissolves.

Epoxy resin has an insulating effect, but, for example, at holes there is a direct connection between carbon and aluminum. It is therefore advisable to place a thin layer of glass fiber fabric around the metal that isolates it from the carbon.

How can the Velomobile body be decorated?

Three methods have been established:

  • Gelcoat: This is a colored synthetic resin, which is incorporated into the mold during construction. Only a few colors are possible here, and only single-color components.

  • Painting: the finished velomobile is painted, this means more colors are possible and, by masking, the combination of several colors. In addition, the layer tends to be thinner (approx. 0.08 mm) and therefore lighter (approx. 200 g / m²).

  • Filming: vynil film offers the most design options and requires the least equipment to achieve a good result. The thickness (approx. 0.09 mm) is comparable to lacquer, but tends to be lighter. Compared to a car, however, velomobile filming is relatively expensive because there are many curved surfaces.

Safety

How is a velomobile perceived by drivers?

Basically, you are treated more like a car than a cyclist. This also means that you usually have more side passing distance, and are rarely referred to take the bike path.

However, speed is often underestimated, it often happens that you are slowed down by drivers who want to pass the bike too quickly, or that overtaking drivers need a lot more distance and are close to oncoming traffic or another constrictions.

In addition, overtaking is often prohibited in oncoming traffic, Here you have to react appropriately and just close the lane in dangerous places.

Visibility?

A velomobile in itself is definitely not inconspicuous or invisible, and if a still don’t see it, you should consider whether he is fit for public road traffic.

Nevertheless, there are weak points, for example, the narrow and low silhouette is difficult to see, especially in difficult lighting conditions. Good lighting helps here especially daytime running lights. The second main problem is likely to be selective perception – cyclists are often not expected, least of all to be fast, and also not in bad weather. To change that, more people would have to cycle.

How to behave in traffic?

As usual: with foresigh, confidence, calm.

However, a particular danger is waiting behind stopped cars without being visible in their mirrors. Many SUV-like cars only have very high and small rear windows, so you should keep a good distance to be visible from the inside mirror. In addition, a velomobile is too narrow to be seen from the side mirrors if you are not exactly offset to the side of the car. Many cars have reversing sensors, but these are often mounted quite high, so that a velomobile may not be reliably recognized.

What is the risk of injury when an accident?

Solo accidents are generally less common in multi-lane vehicles because it is not so easy to tip over, when it’s slippery you slide a bit, but usually nothing else happens.

In a collision a velomobile has neither the stability nor the mass of a car in order to counter it with something worth mentioning. Nevertheless, many accidents have gone surprisingly lightly so far for the following reasons:

  • Protection by the bodywork: the velomobile slides on its body over the asphalt instead of the driver with his bare skin – even if the bodywork is damaged by a collision.

  • Lying position: in the event of a frontal impact, the legs are affected first, however, these can withstand high forces quite well, in contrast to the head or arms that often contact first on an upright bike.

  • Shape: the convex shape of the body prevents the bike from sliding under the car and being overrun.

This is why injuries often arise primarily from sharp objects in the interior, such as screws, handlebars, brake and shift levers, or from a collision of the head and upper body with the coaming or the inside of the hood.

However, there have also been some serious accidents, in these, the driver has either been thrown out of the velomobile or has collided with an object (e.g. stone, post). For this reason, some velomobiles have a cockpit hatch that is as small as possible, which is not only aerodynamically favorable, but (especially if it tapers to the rear) fixes the body at the shoulders. Since the rear hood is usually higher than the head, it acts as a roll bar – if the driver sits far enough back. A shoulder strap would also help here.

In contrast, a hood is usually not so stiff and firm that it can withstand high forces. But above all, it is only fastened with rubber tensioners, so that it could fly away under greater force. At least it should direct objects past the head and prevent skin contact with the asphalt.

Fear is relatively widespread that a sharp-edged carbon splinter will result in a crash. But this is largely unfounded, the carbon shell is so thin that it tears quickly and hardly produces splinters. In addition, modern velomobiles have a nylon liner that largely prevents the cover from tearing, some carbon fibers break, but remain in place. A danger would only arise if very massive and stiff carbon parts were pulverized by the collision, but in such a violent accident, you have completely different problems than a few splinters of carbon in your skin.

Because of the roll bar function of the scoop and the lower risk of self inflicted accident, there are significantly fewer reasons for one to wear a helmet, in addition, many do not fit under the hood or are too long at the back. In addition, the face would still be unprotected.

Contributors and Sources

Version 3.0 (Git commit: 7ca4fd1, date: 2022-12-20)

Contributors:

Sources:

Calculations and parameters

All diagrams in this book are calculated with the following parameters and formulas:

  • Driver mass: 75 kg

  • Mass of the velomobile: 25 kg

  • Effective cross-sectional area: 0.05 m2

  • Rolling resistance coefficient: 0.005 (and \(c_\text{dyn}\) = 0.1)

  • Drive losses: 8%

The exact calculations can be found in the source code of the scripts in the Git repository of the book.

..index:: ! rolling resistance, dynamic rolling resistance, adder

Rolling resistance

Rolling resistance is the force required to roll a wheel over a rough surface. It mainly depends on the load; all other influences are summarized in a constant, the rolling resistance coefficient \(c_\text{R}\) (see also: `What is rolling resistance?`_). So the formula is:

\[F_\text{roll} = c_\text{R} \cdot F_\text{N}\]

The normal force is \(F_\text{N} = m \cdot g\), i.e. the proportion of the weight (= mass times gravitational acceleration) that presses vertically on the surface. So the mass goes in linearly, the speed doesn’t matter.

Power is work per time, i.e. force times distance divided by time, or summed up force times speed:

\[P_\text{roll} = F_\text{roll} \cdot v = c_\text{R} \cdot m \cdot g \cdot v\]

So the rolling resistance power increases linearly with the speed \(v\).

However, measurements show that the rolling resistance force is not constant, but depends on the speed (see: `What is rolling resistance?`_). According to the calculation convention of Kreuzotter the following formulas result:

\[ \begin{align}\begin{aligned}F_\text{roll} &= c_\text{R} \cdot F_\text{N} + c_\text{dyn} \cdot v\\P_\text{roll} &= F_\text{roll} \cdot v = c_\text{R} \cdot m \cdot g \cdot v + c_\text{dyn} \cdot v^2\end{aligned}\end{align} \]

rolling resistance coefficient

The website Bicycle Rolling Resistance offers what is probably the most comprehensive and consistent test of bicycle tires. There the tires are tested on a 77 cm drum, with a load of 42.5 kg, a constant speed and air temperature and after a 30-minute warm-up phase.

As you can see in Fig. 12, the rolling resistance coefficient of road bike tires is mostly in the range between 0.003 and 0.006.

However, fast tires are only tested there in racing bike sizes, while the front wheels of the velomobile are significantly smaller. It can therefore be assumed that the rolling resistance of small tires is slightly higher - if the material and production is identical to that of the large tires. In addition, the front wheels of most velomobiles have a negative camber; that should also increase the rolling resistance a bit (see: `Why are the front wheels tilted? Doesn't that brake?`_).

Another source for rolling resistance measurements is Wim Schermer’s blog; in his tests, he focuses on tire sizes for velomobiles and recumbents. However, he does not use a roller dynamometer, but his pendulum method - two wheels connected to an axle carry an eccentrically attached weight, so that the construct rolls back and forth. However, some of his measurement results are not at all consistent with the measurement results from the roller test stand and also the subjective experiences of many drivers. It is probably due to the slow rolling speed of the pendulum method and thus to the dynamic rolling resistance, which causes the total rolling resistance to increase sharply at high speeds.

Dynamic rolling resistance

It is widely agreed that power metrics are best explained in terms of a rolling resistance force, which is not constant but increases with speed. However, there are hardly any measurements of the dynamic rolling resistance coefficient, and no conclusive explanation of the cause.

In the Kreuzotter calculator \(c_\text{dyn} = 0.1\) is assumed, as well as in the diagrams here. This means, for example, that a rolling resistance coefficient of 0.005 measured by Bicycle Rolling Resistance at 28.8 km/h corresponds to a base rolling resistance coefficient of 0.0042.

Another measurement comes from Charles Henry, determined on a roller test stand; he comes to an average aily increase in the rolling resistance coefficient by 0.0001 per m/s increase in speed. In the previous example, the base rolling resistance coefficient would also be 0.0042.

However, his graph also shows that the increase in rolling resistance coefficient is non-linear – it is more pronounced at low speeds, and that the increase is different for different tires.

Bicycle Rolling Resistance also has a small measurement, but at relatively low tire pressure; here the result is significantly lower, namely 0.000067 per m/s.

temperature and humidity

When it comes to drag, the effects of temperature changes are pretty clear; the temperature \(T\) influences the air density \(\rho\), which decreases with increasing temperature:

\[\rho = \frac{p \cdot M}{R \cdot T}\]

The humidity also influences the air density (via the gas constant \(R\)); it decreases with increasing humidity. However, the influence of temperature and humidity on the air resistance is rather small at normal outside temperatures.

The rolling resistance is completely different. This also increases with decreasing temperature, but non-linearly - in some cases it increases drastically near freezing point, so that a good summer tire becomes a brake anchor in winter. Depending on the tire, however, the temperature dependency differs greatly; thick-walled tires tend to be more temperature-dependent. The material of the bicycle tube also plays a role; At low temperatures, butyl hoses become more sluggish than latex hoses. The reason for these temperature dependencies is unclear, so they cannot be calculated, and measurements are complex and therefore hardly available.

Moisture also affects rolling resistance; the rolling resistance is higher on wet roads. Adhesive forces are likely to be the cause; these are probably not easily calculable.

air resistance

Drag is fundamentally complicated (see: What is drag?); but for calculation it is usually reduced to the pressure resistance and packs all other effects into the cW value (see: What is the cW value?). Therefore, the drag force is simplified the dynamic pressure (\(p_\text{dyn} = 1/2 \cdot \rho \cdot v^2\)), which acts on the effective cross-sectional area:

\[ \begin{align}\begin{aligned}F_\text{air} &= p_\text{dyn} \cdot A_\text{eff} = \frac{1}{2} \cdot c_\text{W} \cdot A \cdot \rho \cdot v^ 2\\P_\text{air} &= F_\text{air} \cdot v = \frac{1}{2} \cdot c_\text{W} \cdot A \cdot \rho \cdot v^3\end{aligned}\end{align} \]

Thus, the cross-sectional area \(A\) and the air density \(\rho\) have a linear effect on the air resistance, but the speed \(v\) has a square effect on the drag force and even on the third power on the power.

However, since velomobiles have a fairly aerodynamic shape, not only the pressure resistance plays a role, but also the frictional resistance. This means that it is not just the shape and cross-sectional area that matters, but also the total surface area in the flow - a large surface area provides a lot of surface friction and is reflected in an increased cW value.

drag coefficient

Air resistance measurements are expensive; to measure it independently of rolling resistance, you need a wind tunnel, and there are correspondingly few measurements. Probably the only measurement of several different velomobile types was carried out at the Ostfalia University in Wolfenbüttel; the results were published in 2022 in the publication “`Air resistance measurement of various velomobiles in the H. U. Meier wind tunnel <https://www.ostfalia.de/cms/de/m/fakultaet-maschinenbau/News/Velomobil_Vergleichsmessung_HUM.pdf>”. __” published by Velten et al..

The cross-sectional areas of the measured velomobiles are between 0.393 m2 and 0.472 m2, with the velomobiles known to be fast all being around or below 0.41 m2.

The cW value for the hooded velomobile at 53 km/h was in the range between 0.11 and 0.17; the (more relevant) effective cross-sectional area for these vehicles was between 0.044 m2 and 0.071 m2. There was only a slight difference between velomobiles with open and closed wheelhouses en (or pants). However, the measurement was not performed with the wheels rotating, so in reality velomobiles with closed wheel arches should have a greater advantage.

Also, the ground was stationary, so it didn’t move relative to the velomobile like a road. This means that the flow on the underbody is slightly different in reality than in the wind tunnel. (However, at least the distance to the ground was increased in order to reduce the influence of its boundary layer.) And since the velomobiles were stationary, there were no vibrations in the bodywork, as would be the case on the road. The measurement results are therefore not directly transferrable to reality.

An interesting result, however, is that for the fast velomobiles the cW value decreased with speed in all measurements. That’s not entirely surprising; this value is only an approximate constant for a turbulent flow (see: What is the cW value?) - for these velomobiles the Reynolds number should therefore still be close to the laminar range.

drive losses

In a bicycle drive there are both load-dependent and load-independent losses. Accordingly, the efficiency depends on the pedaling power - the higher, the better the efficiency, because the load-independent losses are less important.

With a clean and well lubricated chain drive, drive losses can be 2% or less. With derailleur gears, the jockey wheels are added; if these are smooth-running, the drive losses can be similarly low - however, they can be 5% or more if the chain is skewed and the chain is dirty (each based on a high pedaling power). The same applies to hub gears: While the losses in direct gear can also be 2% or less, they increase to 5% or more in other gears (Speedhub, measured by Rohloff). A measurement by Andreas Oehler comes up with consistently higher losses: Derailleur gears and good hub gears are around 5% loss, bad hub gears reach 10 % loss or more.

There is no systematic study for velomobiles, so the losses can only be estimated. Velomobiles usually have derailleur gears (with hardly any skewing), but there are also two deflection pulleys in the drive train (= load-dependent losses) and in the slack side a chain rubbing in the chain tube (= load-independent losses). So you have to assume losses of 5%-10%. Kreuzotter assumes a power loss of 8.3% in his performance calculator.

Glossary

aerobelly

A larger amount of visceral adipose tissue (bioprene) jokingly said to improve aerodynamics – but irrelevant in velomobile. Also called fine food vault, big muscle or dumpling graveyard.

air resistance

Force that has to be used to displace the air. The formula is \(F_\text{air} = 1/2 \times c_W \times A \times \rho \times v^2\) that is, the cross-sectional area A is linear and the speed v is squared. The factor \(1/2 \times \rho \times v^2\) corresponds to the dynamic pressure, which depends on the speed \(v\) (square) and air density \(\rho\) (linear).

apparent wind

Vector sum of the wind (“true wind”) and the head wind – if, for example, the true wind and head wind are of the same size and the true wind comes from the side at right angles, then the apparent wind comes at an angle of 45 ° from the front and is according to the theorem of Pythagoras 41% stronger than true wind and head wind.

bigfoot

component of the DF velomobile, where the bottom bracket mast supports the rear end.

bioprene

Joking name for body fat which, like neoprene for example, provides thermal insulation and padding, but also serves as an energy store and does not add any weight to the bike.

body opening

The angle between the torso and the legs. Many people can climb better when this angle is smaller, ie they sit more upright or the cant of the bottom bracket is larger.

bolt circle

Diameter where the chainring fixing bolts are located. Common standards are 130 mm and 110 mm ( compact crank ) and 104 mm (mountain bike).

Boruttisieren (need English equivalent)

theft protection by making the bike look ugly and neglected makes it look unattractive for Thieves.

bottom bracket mast

Mast that runs from the steering bridge to the front and is often attached to the front of the velomobile at the funnel , and carries the bottom bracket with the cranks and pedals. So that this can be adjusted, either the mast itself is adjustable (e.g. extendable and swiveling), or the bottom bracket is slidably clamped on the mast. When the mast is attached at the front, the fresh air is usually also directed through the mast to the rear.

bottom bracket elevation

The difference in height between the bottom bracket and the lowest point of the seat shell. For an optimal body opening angle and the smallest possible cross-sectional area, the elevation must be relatively large, on the other hand, the legs are then in the field of vision – especially when they are also surrounded by a body.

boundary layer

area of a flowing fluid near a surface at which its speed is relative to the surface growing from zero to the flow rate of the remaining fluid. This can be laminar (i.e. there is a speed within the current gradient) or turbulent (i.e. there are small vortices along the surface that are seperate from the laminar flow).

bovid mat

Joking name for a seat pad made of lambskin

brake anchor

A tire with very high rolling resistance. Particularly thick-walled tires with puncture protection inserts seem predestined for this name.

bump Steer

Suspension-induced steering influence, see What is “Bump Steer”?,

camber

Inclination of the wheel so that it is not perpendicular to the road, but tipped outwards (“positive camber”) or inwards (“negative camber”). See: Why are the front wheels at an angle? Doesn’t that slow you down?

caster

The distance between the tracking point and the contact point of a steered wheel. If this is positive, i.e. the wheel contact point is behind the tracking point, the wheel follows the direction of the frame – or from the point of view of the frame, the steered wheel automatically returns to the straight-ahead position. You can achieve a positive caster by placing the steering bearing in front of the wheel, or if this is not the case, tilt the steering axle so that its extension points in front of the wheel. See: What is caster and what is its function?

catenary

A bicycle chain has a certain degree of lateral mobility, but should not run at an angle permanently, as this increases pedaling resistance and wear. However, on velomobiles, the deflection rollers are far enough away from the chain ring or pinion that the skew is limited in all gears. In addition, the deflection rollers on some velomobiles can be moved sideways.

capacity

Length that a chain tensioner can shorten the chain. Chain tensioners with long cages, as are common in the mountain bike sector, have a capacity of around 40 chain links. In order to be able to shift all gears, the capacity must be equal to the difference between the sum of the number of teeth of the largest chainring and the largest sprocket and the sum of the number of teeth of the smallest chainring and the smallest sprocket. If the capacity is not sufficient, you should still be able to switch from large to large, as otherwise a chain that is too tight can cause damage. With small – small, the chain then sags, but this is less problematic.

carcass

A fabric that is embedded in the tyre and absorbs the tensile forces due to the tyre pressure and deformation. The rubber only ensures the airtightness and grip of the tyre on the road, therefore, damage to the carcass (e.g. due to a deep cut) often leads to a dent in the tyre or causes the tyre to burst immediately. To temporarily repair such a flat, patching only is not enough, but you need a piece of solid material that distributes the pressure of the tube to the surrounding enveloppe for example tape.

cente of gravity

Average location of the mass of a body. For many calculations, it can be assumed that the entire mass of a body is gathered in the cente of gravity and all forces attack there.

chainring

Chainrings are the gears driven by the crank. In order for them to fit a certain crankset, the bolt circle must match.

chain line

A bicycle chain has a certain amount of lateral mobility, but should preferably not run permanently at an angle, as this increases the pedal resistance and wear. However, the deflection rollers on velomobiles are far enough away from the chainring or sprocket that the skew is limited in all gears. In addition, the deflection rollers on some velomobiles can be moved laterally.

cleat

Metal plate on the sole of click shoes that locks them into clipless pedals.

clothes hanger

A component in Velomobiel.nl velomobiles, eg the Quest or Strada.

clipless pedal

A pedal on which special shoes (click shoes) snap into place with a metal plate (cleat) and thus attaches the foot solidly. This allows more efficient power transmission because you cannot slide off the pedal. There are several different systems, those in the racing bike area usually have large triangular cleats that are attached with three screws. In the mountain bike area, on the other hand, the SPD system dominates, the smaller cleats of which are fastened with two screws and sunk into the profile of the sole, so that you can walk quite well with them. In addition, Speedplay is also worth mentioning, for example, which offers a particularly large turning range.

click shoes

Shoes for click pedals, See also: Do you need click shoes?

compact crankset

A crankset with two chainrings. Since the bolt circle is only 110 mm, a fairly small chainring can be installed – you can achieve a similarly large transmission range with just two chainrings as with three chainrings.

CFD

Computational Fluid Dynamics i.e. simulation of the flow of fluids (liquids, but also gases such as air) by computer.

conch

Hang upside down in the velomobile (so that only the legs are sticking out) to adjust something at the front, for example on the gear shift. However, many velomobiles now have a maintenance hatch or a removable front, so that should be largely unnecessary.

control arm

A rod holding the lower end of the strut , typically fixed in the middle of the velomobile, below the steering bridge. The longer the wishbone, the less its angle changes when the shock absorber deflects, and the bump steer effect is correspondingly smaller. See: What are all the parts on the chassis called?

crack damper

A rear wheel damper for certain velomobiles. Offers better damping behavior, but has a reputation for being more sensitive.

delta

tricycle with one wheel in front and two in the back. Named after the Greek letter Delta, which like looks like a triangle.

development

Distance traveled with one crank revolution, and thus a measurement of the of the drivetrain. The development depends on the chainwheel, the selected pinion and the size of the drive wheel and tyre, in the lowest gear, velomobiles have a development of about 2 m, in the highest gear about 11 m.

dolphining

in hilly terrain you can go quickly and be energy efficient when the hills are not very steep or high – then you can rest on the flats between. Take up momentum to overcome the slope, and downhill gain speed for the next hill. Like dolphins swimming, jumping out of the water and not slowing down when they dive back in.

double manta

Driving style, in which both arms are hanging out of the cockpit. This only works well with tiller handlebars, because tank steering levers are too low to to operate them with arms hanging out. The name comes from the Opel Manta, whose drivers were known for liking to keep the left elbow out of the window. Since the velomobile is narrower, it can be used as the poor mans air conditioning – especially in summer.

effective cross-sectional area

The product of cross-sectional area and drag coefficient (\(c_\text{W} \cdot A\)). It is a reference value that can be used to compare the air resistance of bodies of different sizes and shapes - the air resistance corresponds to the dynamic pressure on the effective cross-sectional area.

empty strand

The part of the chain that runs from the sprocket back to the chainring and does not transmit any tractive force ( i.e. is not tensioned) – in contrast to the traction strand.

ETRTO

(European Tyre and Rim Technical Organisation), standard for designating wheel sizes. The diameter is measured up to the height of the rim flange, i.e. that is neither the inner diameter of the rim, nor the outside diameter of the rim. The specification indicates not the size of the rim, but the total size with tyre – therefore a mountain bike bike wheel with large tyres can actually be larger than a theoretically larger road bike wheel.

foam lid

With some velomobiles (especially from the Dutch manufacturers ), the entry opening can not only be closed with a hood, but alternatively with a flexible lid made of a foam. This is attached from the inside and offers an opening so that the head can protrude outside, but the rest of the body is largely protected from the wind and rain. The remaining opening can also be closed for parking.

nel Component at the front at the top into which the bottom bracket mast is glued. Since the funnel is also at the stagnation point , it directs the fresh air inside through the mast.

gelcoat

Colored synthetic resin, which, as an outer layer, contains fiber composite material, and among other things protects from UV radiation. Is mainly used in boat building, but also in some velomobiles. When laminating, the gel coat is first painted into the negative form, then the fiber fabric and epoxy resin is laid up over it. If no gel coat is used, the finished velomobile must be painted or filmed.

hood

The cover of the entrance hatch, usually with side windows and at the front a hinged motorcycle visor, is called hood.

HPV

Human Powered Vehicle, so accordingly muscle powered vehicle. Also the name of the German association, HPV Germany eV, the international umbrella organization is called IHPV.

hump

Rise in the rim bed of some tubeless rims, which ensures that the tyre remains close to the outside rim flange and does not slip inside. To do this, you usually have to over-inflate the tyre a bit until the tyre slides over the hump. You can retrofit a hump yourself by putting a cord on the rim bed parallel to the rim flange and covering it with tubeless rim tape.

IGUS

A company that produces plastic plain bearings. Since several ball heads are installed in the steering linkage of a velomobile, you can save a lot of weight with plastic versions.

ingmarize chain

Chain care by not caring for the chain. Since wear is mainly caused by dirt and this adheres particularly well to oil, an oil-free chain has a certain justification if the chain is permanently exposed to water and dirt, i.e. you would have to re-oil and the chain is effectively water-lubricated – and the bike too is used frequently so that the chain does not rust. This is not a good idea in a velomobile because the chain is exposed to almost no dirt and water, and a clean chain has the least friction if it is sufficiently oiled.

Kamm tail

A vehicle rear that is not completely tapered, but is cut off sharply. The sharp edge ensures that the flow is not diverted inwards at the back and swirled there, but abruptly breaks off. Although it is slightly more inefficient than a long tail, the losses due to the greater turbulence are partially compensated for by the smaller surface area and the lower weight. Named after Wunibald Kamm, hence the Kamm-tail.

kinetic energy

must be be used to set a mass in motion. The formula is \(E_\text{kin} = 1/2 \times m \times v^2\), i.e. the mass m is linear and the speed v is squared.

multi-track

A vehicle with multiple wheels that do not run in a row. Such a vehicle cannot lean into the curve (except for special leaning vehicles), nor does it fall over when it is stationary or a wheel slips.

Meufl

An abbreviation for Mitteleuropäisches Fahrradlabor (Central European Bicycle Laboratory), this is how Harald Winkler describes his bicycle projects. But he became known in the 1990s for his bicycle covers made of closed-cell foam, which were aerodynamic and very light. Correspondingly, devilish mostly means that a component is made of foam.

NACA Duct

See What is a NACA Duct?

Normalized Power

Rides with the same average performance can have very different levels of stress on the body – depending on how much the performance fluctuates. This is expressed in Normalized Power , an empirical formula that accounts for power fluctuations against average power. For this purpose, the pedaling power is calculated as a moving average over 30 seconds, the fourth power is formed, averaged and the fourth root is taken from it. The result is an average performance that would be just as taxing if the power output was constant as the fluctuating performance.

out of the saddle

Kicking style in which you stand up from the saddle while rocking your body relative to the bike. Has the advantage that the pedaling movement is partially decoupled from the crank rotation – you not only push the pedal down with your legs, but also your body up, and can thus partially compensate for an unsuitable gear and provide more variety when pedaling. Unfortunately, that’s not possible on recumbents, because you pedal perpendicular to gravity, and therefore you can’t use gravity to temporarily store pedaling energy.

overrun

The distance between the toe point and the contact point of a steered wheel. If this is positive, ie the wheel contact point is behind the toe point, then the wheel follows the direction of the frame – or from the point of view of the frame, the steered wheel returns to the straight-ahead position by itself. A positive caster can be achieved by arranging the steering bearing in front of the wheel, or if this is not the case, by tilting the steering axle so that its extension points in front of the wheel. See: What is caster and what is its function?

pinion

The sprockets are the gears driven by the chain, which sit on the freewheel of the drive wheel. It is usually a Shimano freewheel body with grooves on which the sprockets sit either individually or riveted together on a common sprocket carrier (“spider”). The entirety of the sprockets is called a cassette.

Pinlock

Pane that is attached to the inside of a motorcycle visor and ensures airtight double glazing. This serves as thermal insulation and prevents fogging. See: What do you do against fogged windows? and How does a velomobile drive in winter?

pants

An aerodynamic cover for open wheel wells, See What are pants?,

PCD Bolt circle

Diameter at which the fixing screws of the chainrings are located. Common standards are 130 mm, 110 mm (compact crank) and 104 mm (Mountain bike).

rear derailleur

The rear derailleur is the derailleur system for the pinions, typically on the rear wheel. It is usually shifted by moving the derailleur cage with the pulley wheels, this also serves as a chain tensioner – the longer the cage, the greater the capacity of the rear derailleur. In some velomobiles, however, these tasks are separate, there the rear derailleur pushes the chain to the side using two plates, and the chain tensioner is located separately at another point.

rolling resistance

force that must be applied to a Roll the wheel over a rough surface. (.. math:: F_text{roll} = c_text{R} cdot F_text{N})

shock absorber

Part of the strut

single-track

A single-tracker is a bike in which the wheels are one behind the other and which can therefore lean into the curve. This can be a recumbent bike, but also an upright bike. However, velomobiles are practically always multiple tracks.

sprocket

The sprocket is the chain-driven gears which are on the freewheel of the drive wheel. It is usually a Shimano freewheel body with grooves, on which the sprockets slide either individually or riveted together on a common sprocket carrier (“spider”). The group of sprockets is called a cassette.

snakebite

Loss of air in the tyre through two holes in the tube that looks like a snake’s bite. The cause is a puncture, The tyre is pressed in so much that the tube is crushed between the tyre and the rim flange, which leads to the two holes. With higher air pressure and / or wider tyres, the susceptibility to snakebites can be reduced.

stagnation point

The point in the front of a vehicle where the air does not come in at an angle and is deflected sideways, but hits the surface perpendicularly and builds up. The kinetic energy of the air results in an increase in pressure. Here, a ventilation opening is a good idea because, due to the increased pressure, a small opening is sufficient for a large air throughput. In addition, there is no laminar flow here that could be disturbed by the opening.

stormstrip

A tape in the longitudinal direction of the vehicle which hinders the overflow in the transverse direction and thus reduces the susceptibility to cross winds. See: How can wind sensitivity be reduced? ,

stall

Air flow is either laminar to a surface and follows its course, or detaches and swirls. The transition is relatively abrupt, hence the name.

scoop

The rear upper part of a velomobile that starts behind the head and runs in an arc to the tail. It serves as an aerodynamic continuation of the head and prevents turbulence there, similar to an oversized aero helmet. At the same time it is the highest point of the velomobile and because of the strong curvature it is very stiff, so it also serves as a roll bar in an accident.

scarfing

Repair of fiber composites by beveling the fractured surface and inserting matching patches of increasing size. Firstly, this means that there is a large adhesive surface, secondly, the weight hardly increases as a result of the repair, and thirdly, the adhesive area remains thin and elastic, so that there are no stress peaks.

saddle

Step by step, when you get out of the saddle and move your body back and forth relative to the bike. Has the advantage that the pedaling movement is partially decoupled from the crank rotation – with the legs you not only press the pedal down, but also the body upwards, and can thus partially compensate for an inappropriate gear and provide more variety when pedaling. Unfortunately, this is not possible on recumbent bikes because you step at right angles to gravity and therefore cannot use weight to temporarily store pedaling energy.

sealing milk

A latex-containing liquid with which a tubeless tyre is sealed. This is put in before assembly of the tyre, or can, with the MilkIt systems special valves, be put in after with a syringe. Because the sealing milk dries out, it needs to be refilled every few months.

steering bridge

Cover between the front wheel wells under which the steering linkage is located (i.e. tie rod and wishbone). Since the bridge is quite voluminous and therefore stiff, the bottom bracket mast or the front pulley is often attached there.

steering knuckle

A fastening at the lower end of the shock absorber, to which the trailing arm, wishbone arm, tie rod and, in the case of tank steering, the steering levers are attached. See: What are the names of all the parts on the chassis?

scrub radius

The radius by which the wheel turns around the pivot point of the steering when steering. See What’s the deal with the scrub radius? and where is the steering axis on a MacPherson strut?

straight running

Straight running is positively influenced by a large wheelbase, neutral steering, low sensitivity to wind, and (especially with a Einspurer) from a large wake.

strut

Part of an independent suspension, on which the wheel hub carrier with suspension and damping Is provided. Also called MacPherson strut.

tadpole

tricycle with one wheel at the back and two at the front. Named after the tadpole, since that Tricycle is large at the front and thin at the back.

tyre contact patch

The area of the tyre that is flattened by the load on the road. The size of the patch depends on the load and the air pressure, since pressure times area equals pressure force, the area is so large that the pressure force equals the load – with higher pressure the tyre patch can be reduced accordingly. The shape depends on the tyre width, with a wide tyre the tyre contact is short and wide, with a narrow tyre narrow and long.

The icing on the cake

is what a hood is called for the Quest velomobile.

tracking

The tracking is ideal when both front wheels point straight ahead with neither toe-in or toe-out.

tracking point

Intersection of the steering axis extended to the street

trailing arm

A rod that holds the lower end of the strut and typically extends from the front end of the wheel well to the steering knuckle. The farther inside the wheel well and the further to the front of the trailing arm is attached to the steering knuckle, the further out the directions of trailing arms and wishbones intersect, ie the more negative the steering scrub becomes. See: What are the names of all the parts on the chassis?

tie rod

Crossbar of the vehicle that connects the two front wheels and transmits the steering movements. With tiller steering, the steering movement is transmitted to the front wheels via the tie rod, while in the case of tank steering, the steering levers act directly on the steering knuckles and the tie rod merely couples the steering movements of both wheels. A tie rod can be continuous or split, in the latter case, it moves similar to the wishbone when deflected, which reduces the bump steer. See: What are the names of all the parts on the chassis?

tension strand

The part of the chain that runs from the chainring to the sprocket and is tensioned because it transmits the driving force – in contrast to the empty strand.

tiller steering

Steering on tricycles, in which a central steering rod moves the tie rod via a universal joint. The English name originally means boat tiller.

tank steering

Steering on tricycles, where two steering levers steer the wheels on the left and right. The steering levers are connected to the steering knuckles, the tie rod only serves to couple the two wheels and thus the steering levers.

tubeless

Normally, a tyre only ensures grip on the road and absorbs the forces created through its carcass, the airtightness is guaranteed by a tube inside. In the case of a tubeless wheel, the tube is dispensed with because the tyre is made airtight and also sits particularly firmly on the rim. This has a flat rim base and often also humps to fix the tyre better, In addition, a special rim tape seals the rim bed airtight. The valve is screwed directly into the rim. Usually, there is also a sealing milk placed in the tyre that seals small leaks. Eliminating the tube reduces rolling resistance and increases driving comfort, also lower pressures are possible because no Snakebite may occur and small holes are often sealed by the sealing milk. But repairing large holes is more complex because the tyre sits more firmly on the rim.

versatile roof

An airy roof, not as closed as a hood , but not as aerodynamic either.

water slide

component of the Alpha7 Velomobile that supports the bottom bracket at the rear end.

wheel well

Housing in which the wheels are located. These serve as aerodynamic fairing, but also to protect against dirt (like fenders), and since the wheel wells are often an integral part of the body, they also have a supporting function, For example, the struts are screwed to the wheel wells at the top of the front wheels.

wheel disks

Discs that are attached to spoke wheels to ensure a smooth surface and thus reduced air resistance. The discs are usually glued to the rims, if they are made of fabric, also clipped onto the spokes.

wishbone

A rod that holds the lower end of the strut and is typically attached in the middle of the velomobile, below the steering bridge. The longer the wishbone is, the less its angle changes when the shock absorber is deflected, and the effect of bump steer is correspondingly less. See: What are the names of all the parts on the chassis?

Index

A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

R

S

T

U

V

W