Mohsen Saleh’s RWS recumbent

Mohsen Saleh, a design engineer living in Milan, Italy, has been working on a rear wheel steering recumbent bike design (in his spare time) for about a year. In a recent post on his blog, Mohsen explained his three main objectives for the project:

To eliminate the long chains used on current recumbent bikes
To eliminate interference of legs with the steering front wheel
To design a recumbent bike with more potential for use in urban environments (a bike that is easier to stop and start, considering constant traffic interruption in cities)

Just recently, he built his first prototype of his design, which in addition to rear wheel steering, features a planetary hub gear in the front wheel. Check out his post for more information about the bike and its transmission, and see this video showing the prototype construction and the steering and gearing systems at work. You can also find additional photos of his prototype here. Prior to his RWS bike design, Mohsen worked on a lean steer trike with a similar planetary gear system, so take a look at his post about that project as well.

As you probably know, designing a stable rear wheel steering bike is a difficult task, and Mohsen covers a few of his design challenges in the post. He points out that his prototype, while rideable, is a bit wobbly, and he lists a few known issues that he wants to fix with his next design. He is also looking for feedback from anyone with experience designing a RWS bike. If you wish to offer any input that might help him with his second prototype, leave a comment here, at his post, or contact him through his blog.

Posted

February 3, 2012

Concept

James

Tags:

planetary gear, reader submitted design, recumbent

Comments

20 responses to “Mohsen Saleh’s RWS recumbent”

Wuss912

February 3, 2012

i guess you dont hang with any unicyclists 🙂 but you should check out the http://www.krisholm.com/khu/kh-schlumpf drive. it’s probably a lot more effeicent than your brake clutch 🙂

Reply
1. James Thomas
  
  February 3, 2012
  
  I don’t know any serious unicyclists, but I am familiar with Schlumpf drive. I have mentioned it on this blog a couple of times before…and I agree that it could work well for an application like this
  
  Reply
Peter Eland

February 3, 2012

Interesting design. It’s been done more often with front drive, though the problem of how to achieve a multi-speed transmission is a thorny one (unless you take a chain out to the wheel rim and back again…). A Schlumpf does get you two gears…

More at e.g.

http://www.hupi.org/HPeJ/0017/GarnetDirectDriveRecumbents.pdf

http://www.velovision.com/mag/issue8/Velo-Vision-08-Stegmann.pdf

Reply
James Thomas

February 3, 2012

Interesting links! Thanks, Peter.

…I’ll just add that VeloVison Magazine is always a great source for information about recumbents, folding bikes, work bikes, etc. Check it out if you are interested in such.

Reply
1. Peter Eland
  
  February 3, 2012
  
  Thanks James.
  
  I meant front steer not front drive in my first comment of course!
  
  Reply
Andy

February 3, 2012

RWS is the new hubless wheel

Reply
Mike

February 3, 2012

I wouldn’t necessarily agree with that. First of all, he actually built this thing, which is more than can be said of any of the “designers” who have drawn hubless wheels we’ve seen on this blog (other than the university team who just wanted to see if they could do it). Second, I think he’s identifying known issues with current recumbent designs (long chain runs, complex steering linkage, unsuitability for urban use due to size, personal sizing requiring either complex fittings for contact points or lots of design work for each size), not fixing what ain’t broke. I think this is an awesome project, and even if working out stable steering geometry for this design turns out to be impossible he will have achieved a huge amount by proving that.

I do think that the upper limit for how steep a grade you can climb on a front hub drive recumbent when the road is wet or unpaved is going to end up being impractically low. However, people have already worked on chain front drives with front steering that move the front wheel farther aft, so that nut can be cracked later when the rear steering is worked out (though of course a shorter wheelbase will change the steering).

Reply
Impossibly Stupid

February 3, 2012

I’d rather have the trike version. I long for an adult Green Machine . . .

Reply
Victor Ragusila

February 3, 2012

Unfortunately, stability is one of those things that you either do well, or dont bother building a bike. It seems so far that RWS is possible for slow speeds, but fails at higher speeds. I know track racers can ride their bike backwards, but i am not sure how fast. I look forward to see if anything comes out of this project.

Reply
Victor Ragusila

February 3, 2012

Also as a follow-up, a modified Rolhoff might be really neat as a transmission system. Right now the 1:1 gear is the 11th, and most are slower. If there is a way to modify the hub such that it starts from 1:1 and goes to 1:5.1 it would make this transmission very nice and efficient.

Reply
Peter Eland

February 4, 2012

Victor, the Rohloff is a lovely hub but how would it fit into a front drive like this? The pedals need to be attached to the same axis as the hub axle, and to rotate together. The axle joining them needs to be pretty hefty to resist pedalling forces, and the frame attachment also needs to be completely different (similar to a unicycle, hence the first comment). You would have to fundamentally redesign it, in fact the R&D effort would be no less than for making an entire new hub gear.

ISTR that Kretchmer (referenced in the HUPI link above) did actually propose a suitable hub gear design with dimensioned drawings back in the late ’90s – it’ll be in his patents somewhere. But getting a working hub gear made is beyond the resources of most experimenters.

BTW the Velayo is a commercially available rear steer trike which seems stable up to pretty good speeds, using a more elaborate linkage: scroll down to the last picture in the first link below for a look – not shown well on their own website.

http://www.velovision.com/showStory.php?storynum=1011

http://www.fortschritt-fahrzeugbau.de/produktkonzept.htm

Reply
Mohsen

February 4, 2012

Thank you Wuss and Petter for the links 🙂

Andy!!! I do have sympathy with you about hub-less designs, and if i dont manage to stable this bike i would agree with you also for RWS one as well. take it as an experiment…

Thank you Mike,

@Victor…you are right, stability is the main issue with RWS designs, and actually the main reason that i began building this prototype was to investigate stability of my 4 bar linked steering mechanism. the first try was not as i expected but I am hopeful that with this configuration i would be able to build a bike that is stable in low speed as well as higher one.

Reply
Cycling’s Week on the Web – Bike Rumor

February 4, 2012

[…] the pros and cons of his prototype and what he plans to change in future designs. Thanks to Bicycledesign.net for the coverage of […]

Reply
Mark Swartz

February 8, 2012

Rear steering designs keep on getting recycled ad infinitium. Unfortunately, they never go away as someone working in isolation will identify supposed ‘problems’, and then go and find supposed miraculous solutions in the form of ridiculous designs that require unique riding skills and present new dangers, i.e., case in point, rear steering. I should know, I’ve ridden and even owned one ( the Vacuum Velocipede). The design in this post looks very similar to a trike built and flogged by the Thebis company in Toronto during the 1980’s. Ask yourself why you don’t see rear steered bikes or trikes succeed in the marketplace. Darwin is right.

Reply
1. Andy
  
  February 8, 2012
  
  And there’s most of bicycle design in a nutshell. They assume that if something isn’t widespread today, then they should create it themselves, only to fall into the same pit that previous designers did. Only the truly useful concepts emerge, and those are very few.
  
  Reply
Hephaestus

March 11, 2012

Ah the lure of a rear-steered recumbent bicycle. I spent 10 years building prototypes and testing. After which I published what I hoped was a definitive article in “Human Power” so that others didn’t have to start from scratch on the problem. Alas that was back in 1990.
http://www.ihpva.org/HParchive/PDF/27-v8vn1-2-1990.pdf
Pages 6-7 &17-20. This should help you understand the design space a bit better.

When I finished I had a vehicle that could be ridden at moderate to high speeds but could not be balanced at very low speeds. I did a 50 mile ride on it but had trouble negotiating through walking traffic.

I do not believe either a single pivot or a four-bar linkage design can produce the auto-stability you need for easy starts and very low speed maneuvering. My first prototype had a four-bar like the one on your post with the instant center behind the contact patch of the wheel. It couldn’t be ridden. All the prototypes that worked had very long trails with the pivot in front of the wheel’s contact patch.

Regarding the hub drive, it is very clean and compact. Unfortunately, if the vehicle wheelbase is a reasonable length, the drive wheel is too lightly loaded. You don’t achieve the 55-60% loading you need for good traction.

http://www.lefthandedcyclist@blogspot.com

Reply
1. Mohsen
  
  March 13, 2012
  
  Hi Craig
  Thank you for responding to this post, i have read your story of rear wheel steering and I tried to find a contact from you, I have actually quoted from you in my blog,
  I saw a parallelogram wheel positioning in HPV magazine but as my concept was about a linkage with different configuration i assumed that i might have found the solution.
  The common ground in bicycle dynamic is that the bike should steer into fall direction in order to have auto stability, on paper it seemed that i would achieve auto stability however in practice it wasn’t. I had the concept for a year.and i felt that i have to do it… it took me a a month to design the details and 2 weeks to build. Was worth trying though.
  
  Reply
Jeremy Garnet

March 15, 2012

I think the reason that rear-steered parallelogram linkages work on paper, but less so in practice, is because auto-stability is not merely a matter of turning into the lean and centering the wheel by its rolling resistance. Stability is also concerned with the correct response to sideways forces from ridges and road imperfections that approach the wheel from an angle. When a steered wheel encounters a small ridge or imperfection in the road at a small approach angle, it will deflect the wheel sideways, away from the ridge. If the steering geometry does not turn the wheel into the ridge, so as to ride up and over it, there will be instability. On a front-steerer, the geometry that will do this is the same as that required for a correct lean-steer response. In other words the “lean steer effect” is synchronized with the “ride up on a side ridge” effect. This is a requirement for real-world stability. I don’t think this can be done with a rear-steered bicycle.

The main advantage of rear wheel steering for a direct drive recumbent is the absence of any pedal force feedback to the handlebars. However with careful design the feedback to the handlebars on a front-steerer can be reduced to as little as 12 percent of the applied pedal load (see my HP article link below). This level of feedback is completely manageable. In fact, I have ridden a non-optimized direct-drive recumbent many kilometres without experiencing much of a problem managing the pedal loads. This is because the rider’s feet naturally produce an laterally outward pedal force which tends to negate the force feedback to the handlebars.

http://www.hupi.org/HPeJ/0017/GarnetDirectDriveRecumbents.pdf

Reply
Mohsen

March 23, 2012

Thank you Jeremy for sharing your opinion.
I like your front steerer…
Peter Eland already shared this project with us in previous comments and I liked what you did in your design….would love to ride it to learn more about its handling.

Reply
Randall

August 29, 2023

I love the minimalist design. I too have investigated the planetary gear hub as a front wheel drive solution (chainless). More knowledgeable and experienced people have commented, so take anything I say with a grain of salt.
Since this was posted a decade ago, I’m sure you have moved on in design, but will offer my comments humbly.
1) Have you considered lowering the center of gravity by bending the connecting bar. The rider would sit lower and the pedaling angle might be more natural and easier to bring force to bear. That change in COG might reduce the wobbliness once you are up to speed.
2) A somewhat exotic idea, and definitely more complex, would be to allow the front and rear wheel to be linked to turn in unison. This might help the bike adjust to imperfections in pavement etc. Conversely, the wheels could be linked and steered inversely, but in either case would probably benefit by some limits (stops) to prevent oversteering etc. The advantage of linked steering would be that smaller increments yield bigger results, and thus require smaller changes in COG for balance purposes. Calibration of that type of steering would require a lot of trial and error.

Reply