The Recumbent Bicycle and Human Powered Vehicle Information Center

www.recumbents.com - Linkage Steering

www.recumbents.com

Username:	Password:
Save Password

All Forums

www.recumbents.com

Technical / Bike Building

Linkage Steering

New Topic

Reply to Topic

Printer Friendly

Author

Topic

Page: of 2

Victor Ragusila
recumbent enthusiast

Canada
337 Posts

Posted - 05/03/2012 : 14:51:24

Hello everyone

For our 2012 ASME/BM bike, Bluenose, I designed a linkage steering that was meant to change the steering axis from vertical to horizontal. The purpose is to allow the rider to crank on the steering, similar to what an upright rider does. Some pics below:
<img src="http://jnyyz.files.wordpress.com/2012/04/p4271734.jpg" border="0">
<img src="http://jnyyz.files.wordpress.com/2012/04/p4261723.jpg" border="0">
<img src="http://jnyyz.files.wordpress.com/2012/04/p4261724.jpg" border="0">

It did not work, the rider seemed to have too much delay in the steering, from either slop (which by the end of the day, we eliminated) or from flex.
I have seen linkage steering used before, so i know it is possible. AA has a remote steering NoCom, and EIVIE3 has cable and linkage steering.

I would like to ask some help from people who have build and ridden linkage or cable steering bikes or trikes: How did it work? What parts did you use? What lessons did you learn? Can you think of any reason why my solution doesnt work?

cheers

Victor

alevand
human power supergeek

USA
1645 Posts

Posted - 05/03/2012 : 15:24:10

Try this syntax:

C:
Tony Levand

Edited by - alevand on 05/03/2012 15:25:02

Speedbiker
human power supergeek

USA
1992 Posts

Posted - 05/03/2012 : 15:57:05

So much for the K.I.S.S. principal. It looks like it should work. I suspect that accumulated flex, drag, and slop is delaying reaction times. You fork geometry mystifies me. Or very possibly the control unit's central processor(rider's brain) isn't calibrated for the task. Many riders who switch to narrow, parellel grip steering crash because of CPU malfunction.

Speedbiker
human power supergeek

USA
1992 Posts

Posted - 05/03/2012 : 16:01:37

Also, your chain must run parallel to the steer tube, or at least as much as possible. In low, starting gear you have the worst case scenario for torque steer.

Jeff Wills
human power supergeek

USA
1123 Posts

Posted - 05/03/2012 : 16:04:42

quote:
Originally posted by Speedbiker

So much for the K.I.S.S. principal. It looks like it should work. I suspect that accumulated flex, drag, and slop is delaying reaction times. Your fork geometry mystifies me. Or very possibly the control unit's central processor(rider's brain) isn't calibrated for the task. Many riders who switch to narrow, parellel grip steering crash because of CPU malfunction.

It looks like it's got a steep steering axis and "reversed" rake fork to give it a bunch of trail. If so, it's very similar to the Cal Poly Velox Solium: http://www.ohpv.org/events/albums/bm2004/velox/pages/velox5.htm .
I found the Velox Solium to be very easy to ride at parking-lot speeds, despite being about 3 sizes too large for the bike. It also went 60mph that year, so it's a a proven combination of slow-speed and high-speed stability.

I would think that some sort of push-pull flexible cable (called Morse cable, I think) might help. I can see 4 ball joints and 3 pivots that would need to have no slop for it to work correctly. My Cuda-W-2 streamliner has a bunch of slop in the steering (due to the suspension fork) and getting it started and up to speed is nerve-wracking.
__________________
Jeff Wills
All my bikes:
http://home.comcast.net/~jeff_wills/Gallery/index.html

Edited by - Jeff Wills on 05/03/2012 16:09:47

Speedbiker
human power supergeek

USA
1992 Posts

Posted - 05/03/2012 : 16:05:44

Also, without adding an idler to pull the final drive chain closer to the steering axis, you force the return side chain to bend more than twist. While minor, this adds to the steering effort.

Speedbiker
human power supergeek

USA
1992 Posts

Posted - 05/03/2012 : 16:35:45

All reasons why my new liner is RWD. All fwd affects steering. The more complicated it is, the more it affects the steering. Get it perfect and it isn't too bad, but it is still there. Add that to a heavy(ish) liner and you have a handful. After many changes the Coslinger is pretty good, but not without extra effort.

AA
human power expert

USA
2373 Posts

Posted - 05/03/2012 : 16:52:08

Victor,
Make an adjustable setup that will allow you to change handlebar width and linkage rod ratio.

Set up your handlebar with an adjustable clamp for the rod end bearing bolt. This will allow you to change the linkage rod ratio to the amount of leverage you need. Start out with a wider than planned handlebar. Use bolt on handlebar grips and do some test rides to find what works best with hand position and angle. When your remote is dialed in remove the excess handlebar width.

warren
human power expert

4225 Posts

Posted - 05/03/2012 : 17:37:39

Victor, your steering looks like it rotated vertically. If you make it rotate horizontally it will work fine.

raymondg
recumbent guru

824 Posts

Posted - 05/03/2012 : 17:41:44

Victor:

Typical of such things are:
1. Lost motion, or backlash
2. Flexibility, or hysteresis
3. Steering ratio
4. Inherent steering stability.
5. Biomechanics stability

So...
1)Single linkages like you show will have backlash. It is common practice to use a push-pull linkage so that they can be tightened against each other to eliminate backlash. Given the short throw of your mech, even a few .001 inch(which you can't detect by hand) can give significant steering backlash.
2)Again, given the short throw, your mech acts as a force multiplier. It looks to me that the linkage forces are on the order of 10 times what the rider feels at the grips. To my eye, the mech looks a little flimsy for that much force, at least by precise control standards. Hysterisis is about the force to stiffness ratio of something, so you might want to look at that. Also, the push pull will help stiffen that up.
3)It�s hard to tell if your steering ratio is 1:1. If it is not, it will take time for the pilot to adapt, and the vehicle will feel unstable.
4)Can you verify that your steering stability is good with a standard steering setup to verify that is not the problem.
5)On most bikes (upright and recumbent), the pilot�s body has natural biomechanical steering stability. For example if you fully extend your arms horizontally in front of you and then try to steer about a mostly vertical axis, you will find you have great control with your shoulders, and you naturally want to center the wheel, especially with negative tiller, like on a road bike with a long stem. If you do the same but try to steer about a mostly horizontal axis, you find you don�t have the same degree of control, and there is no self-centering moment.

Just some thoughts after looking at your design. I do like it. I hope you find a solution.

Regards,
Raymond

Speedbiker
human power supergeek

USA
1992 Posts

Posted - 05/03/2012 : 18:51:56

Thanks to Raymond for quantifying my thoughts. Or nearly so.

AA
human power expert

USA
2373 Posts

Posted - 05/03/2012 : 19:26:42

Victor,
Maybe the pivot rod on top of the handlebar with bar ends that can rotate on the handlebar and lock into position.

Victor Ragusila
recumbent enthusiast

Canada
337 Posts

Posted - 05/03/2012 : 21:03:55

I want to keep the steering axis horizontal, unless i cannot find any solution. I saw some bikes with single sided steering linkage. Any comments how those work?

I will probably try to have a double linkage system, but the one on the right might hit the chainring.

The steering geometry is very similar to the CalPoly bikes, i talked with Will a lot about handling. With a tiller it feels ok, a bit twitchy but not bad. It needs a bit more trail i believe.

thanks for the ideas, will keep up to date with the modifications.

Victor

purplepeopledesign
recumbent guru

Canada
568 Posts

Posted - 05/03/2012 : 23:34:22

I'll guess that it's not a mechanical issue but one of control or specifically the amount of negative feedback. That is, when you removed the slop, you also removed any of the sensitivity the ride uses to sense feedback. That alone would make the system very difficult to control when rides are typically used to handlebars that have both high feedback and zero slop. Put that system and rider in an enclosure with limited vision and more feedback redundancy is lost. I'll bet only the most experienced riders were able to control the machine, but with a lot of trouble and descriptions of suddenly losing it in corners...

:)ensen

Those who claim to be making history are often the same ones repeating it.

Video of my trike
http://www.youtube.com/watch?v=MdSLRD_2vzc
Photos of my trike
http://www.flickr.com/photos/purplepeople/

LongJohn
recumbent guru

Netherlands
532 Posts

Posted - 05/04/2012 : 00:53:32

Victor,

Looking at your setup and reading Raymonds comments above my give on this is that your lever arms are too short.
This will increase the force needed to steer on your push/pull rods and rod pivots. If you increase the size of (at least) the arm at your steer and the one at your handlebars you should be fine.
About steering needing to be proportional; I disagree. I once rode a bike at our local funfair that had two cogs in the steering column, reversing steering action and giving it a 1:2 ratio. As long as the bike is dynamically stable, no problems riding it (just reverse riders brain input).

Larry Lem
human power supergeek

South Sandwich Islands
1971 Posts

Posted - 05/04/2012 : 07:55:18

Where is your test mule? I never saw photos of that. I only saw photos of the streamliner.
(I don't believe you'd design a new, complex steering system into a liner that takes over half a year to build without testing it first.)

Larry Lem

Speedbiker
human power supergeek

USA
1992 Posts

Posted - 05/04/2012 : 08:08:37

Oh Larry!!!!

Good point Thomas.

Victor Ragusila
recumbent enthusiast

Canada
337 Posts

Posted - 05/04/2012 : 09:26:47

The test mule does exist, we tested a different system on that. (2 independent levers moving back and forth. steering axis was paralel with the wheel axis. When one linkage moved forward, the other moved back).
The linkages worked well, but on that bike the rider could push on both sides and preload the linkages against eachother. I guess this hid the real issue of the slop and/or flex.

The arms on the fork and steering axis cannot be bigger, the issue is that of clearance to the knees.

Victor

jjackstone
recumbent enthusiast

USA
119 Posts

Posted - 05/04/2012 : 09:38:19

Victor,
Here's a photo of the steering from the bike that Kevin Berls built several years ago.

Having never ridden a recumbent at the time I bought this I found it made for quite sloppy steering feel for me and eliminated it from the bike. However, Mr. Berls rode it like this for several years and apparently did quite well with it. A few of the more experienced recumbent riders at BM tried it and said they didn't think it was too bad with the linkage. Just a simple single rod linkage much like the one in the AA photo. I'm not an ME like a lot of you guys but I would thing that one would want to keep the linkage and the steering axis as parallel as possible to keep forces in only one direction. Hope this helps.
JJ

purplepeopledesign
recumbent guru

Canada
568 Posts

Posted - 05/04/2012 : 10:02:08

IMO, like so many engineering projects, it's as if the complexity is part of the impressiveness. I say go for the elegant. If you want the rider to have fixed reaction handles, make them fixed. Then put a small joystick on the end for control... like the Varna trikes. Of if you need two controls, like a jet fighter, put two and run the cabling opposite. Cable tension can be varied to minimize slop and/or increase sensitivity.

:)ensen.

Those who claim to be making history are often the same ones repeating it.

Video of my trike
http://www.youtube.com/watch?v=MdSLRD_2vzc
Photos of my trike
http://www.flickr.com/photos/purplepeople/

raymondg
recumbent guru

824 Posts

Posted - 05/04/2012 : 10:07:03

Victor,
One way to handle the biomechanical stability might be how you support the rider's arms. Some Varnas have arm rests, and on the Orion I have an extension on the frame that I rest my hands on while holding the steering column, so all I have to do is rock my wrists to steer. On the Orion this doubles as a damper to deal with the trike shimmy. On my drive into work this morning, I tried steering the car by just holding my arms out, and then put the arm rests down and rested my elbows on the rests. I felt much more in control of the steering resting my elbows on the rests.
For push-pull linkage, I don't believe they need to be exactly symmetric. If there is room to move the interfering side inboard, you will still get the play eliminating effect. Of course the linkage forces get higher for that side, but you get anti-backlash, and some stiffening none-the-less.

-Raymond

Speedy
recumbent guru

USA
656 Posts

Posted - 05/04/2012 : 13:25:02

the vertical motion steering is not highly adaptable. few people get the hang of it on the first go.
linkage type needs quality rod ends with no slop, think aircraft or aerospace quality not hardware grade
cables and pulleys actually are smoother and will yield higher angles of deflection but again top quality is required
.09" cable does work .125 for an extra margin of safety
rule of thumb for pulley size is 50 times cable diameter
make your own aluminum pulleys ... the standard aircraft grade phenolic breaks

warren
human power expert

4225 Posts

Posted - 05/04/2012 : 16:16:15

Victor, please verify that you understand what we are saying here. The steering steering should pivot horizontally, ideally the pivot should be at the same angle as the head tube. A vertical pivot in not intuitive, especially in a 'liner. Vertical pivot has been used before, look at the Velocar, but there is a reason nobody else uses it.

-Warren.

Victor Ragusila
recumbent enthusiast

Canada
337 Posts

Posted - 05/04/2012 : 23:45:26

And that reason is because it is not intuitive?

I think I understand what you are trying to say. I will try to make the vertical pivot work, because i think it can lead to a power gain in the sprint (by torquing on the handlebar with the upper body). If we cannot figure out any combination of linkage and damper/steering spring to make it work, i can modify it to a horizontal steering pivot (steering axis parallel to the fork axis).

What other liners have tried a vertical steering pivot?

cheers

Victor

Edited by - Victor Ragusila on 05/04/2012 23:46:53

PUGZCAT
recumbent enthusiast

Canada
240 Posts

Posted - 05/05/2012 : 03:48:34

I've never been a big fan spherical rod ends, they always introduce instant slop into any linkage, if the movement of a joint is on a single plane I've always favored the drill slightly undersized holes slide in a bolt with plenty of grease and washers method.
More riding time will make any setup more intuitive, in my experience, admittedly it can take a lot of practice before it suddenly clicks into place.
Another option would be adding over the shoulder yokes to the seat back so you get the same effect as pulling on the bars while sticking with a "normal" steering linkage and not pulling on the bars.

warren
human power expert

4225 Posts

Posted - 05/05/2012 : 05:40:09

Pugzat, you have to use the aircraft grade rod end bearings.

Victor, yes, it's not intuitive. You can pull on the remote bars when they are mounted horizontally too. My old Barracuda was a monster sprinter because of the stiff remote mount and (relatively upright seat. Even though it was a tank...

Page: of 2

Topic

New Topic

Reply to Topic

Printer Friendly

Jump To:

www.recumbents.com

Snitz Forums 2000