cartoon with rear steered bike

Rear Wheel Steered Bike (RWSB) page

dutch version

russian version

by Erik Wannee.

This site pretends to be a platform for everyone who is interested in rear wheel steered human powered vehicles.
This is explicitly NO one way traffic: Everyone who can tell something interesting about this matter, is requested to tell me, so that I can place it on this site.

Several times, people have tried to design and build a well functioning Rear Wheel Steered Bicycle (RWSB).
Some people start from the principle of trial and error; others work very systematically towards their goal.

Why should anyone want to have rear wheel steering, while there is a good way of front wheel steering?
There can be many arguments.
One motive to build an RWSB could be: It is exciting to build something that is until now so very unexplored. It is a challenge to explore the unknown.

Other - more technical motivated - reasons could be:

That rear wheel steering indeed makes it possible to build very fast vehicles, is proven by the first four wheeled land vehicle that successfully broke the sound barrier: The Thrust SSC (Super Sonic Car).
This vehicle is propelled by two jet engines, whose fairings enclose the non-steering front wheels. The two steering rear wheels are placed in the rear of the fuselage. (I must remark that the maximum steering angle of this car is extremely small.)

the Thrust SSC scheme of the Thrust SSC

But let's no more talk about motorized vehicles, and first restrict myself to two wheeled human powered vehicles.

The first one.
The rear wheel steered bike of M.C. Donald

I have no idea what time the first serious attempt was made to build a working RWSB. In one of my bicycling books I found a sketch of some M.C. Donald dating from the year 1869. I don't know if this bike has really been built, and if yes, if it could be ridden.

I expect it could not.

The centre steered bicycle of W.H. Laubach

In the same year, a patent was registered by a man called W.H. Laubach, who had invented a centre steered bicycle. Because the saddle is connected to the front part, you might have called it a rear wheel steered bicycle.
Detail: Note the saddle pin suspension!

To this bike I have serious concerns about the practical usableness, too.

The first RWSB of which I know that it really can be ridden, is the ABT (dutch for RWSB) of Fred de Weert. He writes about it in the dutch magazine HPV-nieuws 7-2 (july 1991).
His ABT had a negative steering-head angle and a long trail. The steering mechanism had a bult-in reversal mechanism, so that moving the handlebars to the right meant that the bike indeed turned to the right.:

Fred's bike, called ABT

Fred's publication was an inducement for me to contact him. A short time later, I was allowed to borrow his RWSB, and to modificate it as much as I would like.
It appears to be possible to ride such a thing, but it is very unstable. Steering requires much brute force, and while making a turn, it wants to steer stronger than you want it to do. So strong counter-steering is necessary.

Knowing that, I started thinking about the theory of steering two wheeled vehicles.

When you ride a two wheeled vehicle, you are in fact continuously trying to avoid falling. The handlebars are non stop making small corrections.
In other words: you can also say that you alternately make turns to the left and to the right. You never ride exactly straight on. You can also see it as a bike rides over a sandy path or over snow: The rear wheel will never exactly follow the line of the front wheel.
Conclusion: We don't have to study riding straight on. We only have to look at cornering.
During the steering, there are a lot of different forces acting on the steering wheel.
It is important to know that.

Let's first look at the forces acting upon the steering system of a Front wheel steered bicycle (FWSB). Take such an old fashioned city bike in mind, and imagine it's steering to the right: city bike

  • The frame of the bike inclines slightly to the right. That gives a tendency to fall to the right.
  • In the corner, the part of the front wheel that is in front of the steering-head line, has the tendency to pull the handlebars to the right by it's weight. Because of the fork offset, that part is bigger than the other part of the wheel.
  • The smaller part of the front wheel that is behind the steering-head line, will try to pull the handlebars back to the left.
  • The bicycle has an inertia, which gives a centrifugal effect. Mass always wants to go straight on: the 1st law of Newton.
  • A turning wheel has this effect even stronger, because the top of the wheel moves twice the speed of the bike itself. We call this the gyroskopical effect.
  • Because of the fork offset, the turned front wheel will lift the bike. That limits the excursions of the steering handle. You can see that effect as you take a city bike by the saddle and move it sideways. The handlebars will turn somewhat, but then it will stop automatically.
  • Because the bike has a trail, the center of gravity of bike and rider will be placed somewhat to the right as the handlebars are turned to the right. That enhances the tendency to continue the steering movement.
  • The trail of the bike stabilises the excursions of the steering handle. Compare it with a caster wheel under an office chair. Move the chair in a certain direction, and the wheel will automatically turn into the right direction. (A negative trail is inherently unstable: You can try to push the caster wheel the other way: It will continuously try to turn around.)

I hope this wasn't TOO complicated for you. The following will be a little more complicated:

Let's have a look at the steering mechanism of an imaginary bicycle with REAR WHEEL steering. Because there is no standard design of such thing, I'll take the ABT of Fred de Weert. On this bike, I'll discuss a corner to the right, as well.
I discuss the differences with a FWSB. Differences are printed in red: Fred de Weert's ABT

  • The frame inclines somewhat to the right. That gives it the tendency to fall to the right.
  • While steering to the right, the biggest part of the rear wheel (i.e. the part of the wheel that is behind the line between steering head and the point where the wheel touches the ground) has the tendency to fall to the right, thus enhancing the tendency to increase the steering movement.
  • The centrifugal effect on frame and rider gives a strong tendency to increase a steering movement.
  • The gyroskopic effect on the rear wheel gives a tendency to increase a steering movement.
  • Because of the fork offset, the steering rear wheel will lift the bike. That will decrease the steering excursions.
  • Because the bike has a long trail, the center of gravity of bike and rider will be strongly displaced to the left as the handlebar is turned to the right. That increases the tendency to enhance the steering movement, and it enhances the tendency to tumble to the left.
  • The long trail of the rear wheel stabilises the steering behaviour.

As a conclusion it can be said that there must be big differences in steering behaviour. In advance it is not easy to predict how all those factors will work together in practice. But practical results show that all the factors that enhance a once started steering movement, are responsible for a very unstable steering behavior. Even the 25 cm. long trail cannot compensate for that.

Analysing how this RWSB could be improved, I thought as follows:
The biggest disadvantage of the ABT lies in the fact that the long trail pulls the center of gravity so strongly to the wrong direction, while making a steering movement. In order to avoid that effect, the steering head could better be placed behind the center of the rear wheel. But then you create a negative trail, making the rear wheel unstable.
So the question is: Is it possible to move the center of gravity of the bike to the right side, without creating a negative trail?
That won't be possible with a single turning axis. But I considered that such could be possible with a complex movement.

With that in mind, I designed a completely different prototype. I called it ABT-2:


This prototype had no steering head tube, but the front fork was attached to a ball-and-socket joint (left in top of the picture), while the movements of the fork were determined by two bars with screw-threads (bottom of the picture), connecting frame to both ends of the rear axle. The attachment-points of the bars could easily be displaced (because of all those Mecano-like holes). By replacing the bars with longer or shorter ones, their length could also be easily changed. The steering mechanism works through two separate short handle-bars on both sides of the seat (comparable to the system of the Jouta-tricycles) and a bar on both sides, connected to both sides of the top of the fork. This system makes possible a typical lemniscate ('8'-shaped) movement that cannot be made with a normal head tube.

I would have thought that it should be a Columbus' egg, but after four months of intense trying and changing all parameters of the bike, I couldn't ride a single meter on it!
Unfortunately, the movements of all parts are so complex that it is impossible to understand why this bike was unsuccessful.
I couldn't solve a couple of fundamental problems with my 'solution'. Disappointing but instructive.

ABT modificated by Jurriaan Bol

Some time after I borrowed Fred de Weert's ABT, Jurriaan Bol borrowed it, and he modificated it in many ways. He reported about it in HPV-nieuws 11-3 (May 1995). It is an instructive article, because he analysed the steering system very systematically. But, whatever he changed and altered on it, the results always proved disappointing.
(Click here for a summary of this article.)

A picture of one version of this altered ABT is shown here. This is the version that I mentioned above: the one with the steering head behind the rear wheel.

There is someone else who has studied the rear wheel steered two wheeled bike extensively: dr. Craig J. Cornelius, living in Woodinville, WA, USA.
In a 10 year's span, he has built 7 different prototypes, one of them has more or less the same mechanism as my ABT-2.
He reports comprehensively about his experiences in the magazine 'Human Power' issue# 8/2, spring 1990 (page 6-7,17-20).

Of all these things it may be apparent that until now, it has not been succesful to build a well-usable RWSB.
Also from the theoretical view-point, there are some reasons why this principle will inevitably keep having big disadvantages:
  • If you'd ride a RWSB and would have to avoid a suddenly opened car door, you can only steer with the rear wheel in the direction of that car. It would be very difficult to avoid a collision.
  • If you would for any reason ride just along the pavement, it would be difficult to get away from it.
  • Moreover, a bicycle is principally a vehicle that is permanently losing its balance. So it must be corrected continuously by steering.
    When losing your balance on a frontwheel-steered bicycle, you can turn the handlebars in a way that the front wheel moves directly under the centre of gravity of the bike.
    On a RWSB, it's just going the wrong way: As you want to pull the rear wheel under the centre of gravity, the bike turns the other way, thus increasing the dis-balance.

Then, it was a big surprise to me when I found a photo on a german HPV home page of an RWSB, named 'Kalle', on which the driver seems to ride quite relaxed on a cycle-track. I could find this bike on two photographs in the (no more existing) magazine Bike Culture Quarterly (resp. no. 11, page 52 and no. 18, page 49), from which I took the following pictures:
Kalle-bike 3e Kalle-bike 3f: improved version of the 3e
Supplementary information from this magazine and readers of my page taught me that this bike is made by Hans-Ulrich Reimers from Reinickendorf (near Berlin, Germany). He builds them on custom order.
It seems to be well rideable. The manufacturer claims speeds up to 40 km/h. The design resembles to a FlevoBike, but with the seat attached to the front part. Just like the Flevo, both wheels are sprung. Moreover, I remark that the rider is in a pretty vertical position, there is a positive steering head angle, and the steering head is far to the front. A visitor of this site attended me to another special property of this bike: The steering head line points towards the ground in front of the front wheel.
With some fantasy, you could say that both wheels have a positive trail.

By the way, you can wonder if this bike can be really called a RWSB. You can also consider it as a FWSB on which the rider is actually sitting on the handlebars.
In contrast to the Flevo, I expect this bike will not be rideable without using your hands. I hope to find more information about this remarkable bike, soon.

Rear Wheel Steered 'unicycle'

On the 2004 DaVinci art festival the Kinetic Sculpture artist Duane Flatmo was tooling around on this RWSB quite well. Steering is done via steel cables.
Actually the position on this bike is so vertical that it doesn't take much to go forward over the front wheel, so it can be basically seen as an RWS unicycle. This brings me to the thesis that rear wheel steering is easier if the riding position is more upright, and/or more towards the front wheel.
Thanks to Paul Hanrahan who sent me the picture and the story.

A number of other RWSB's has been made, and even more people take the challenge.
In order to limit the size of this page, I put these projects on a separate page. Click HERE to visit them.

On several places on the globe, rear wheel steered trikes are built. Because stability problems play a less dominant role in trikes, some of those trikes have even been in serial production.
I dedicated a separate page to them. Click HERE to visit them.

This is what I can say about RWSB's. Not very encouraging. In the meantime, I stopped my plans about RWSB's, and at this moment I'm designing and building other HPV's. (as you can see on the other pages of this site.)
I want to advise you: If you want to start a rearwheel-steering adventure: Think good about what you're beginning. And contact people who have built RWSB's before.

If you want to contact me, look at my personal home-page. If you want to give comments or if you want to add something (pictures or experiences) to this page, I like to know!

Erik Wannee / last update: 2005-05-07 Homepage recumbent bike site