I've been envisioning a vehicle suspension layout that would have certain properties that could theoretically benefit from an alternative steering setup. Let's say this vehicle would have packaging that would make installing a steering shaft to a rack very difficult. The proposed vehicle would also have a solid front axle and it would be desirable to have the steering actuator mounted to the axle without any kind of rigid link so there would be absolutely no bump steer issues with suspension travel. And finally it is desired that the overall steering ratio be pretty fast, not kart 1:1 fast but like 4:1 with one turn lock-to-lock and 90* nominal total front wheel steering. My idea is to mount a hydraulic double ended piston, basically exactly like what is used in the off road world but sized appropriately, and have that mounted to the front axle. the hoses from that would run to a master cylinder(s?) that would connect to the steering column shaft with some sort of arm or linkage in a more convenient location. No hydraulic pump should be required as the master cylinder would create this pressure.
The potential advantages are:
- no bump steer
- the ability to choose steering ratio by altering cylinder sizes
- easier packaging
- smooth, nearly friction free steering
The things that worry me are:
- designing a linkage that keeps the master cylinder travel vs steering input roughly linear over 180* of rotation
- the possibility of the steering travel slipping and becoming misaligned with respect to the slave cylinder
- it might require custom cylinders
- steering feedback
So is this a crazy idea? Full hydro steering is pretty common and reliable in the off road world but they have pumps and things that this wouldn't. I don't want to reinvent the wheel here but if something could be proved to better for a situation then better is better right? I think this would work in theory but my main concerns are the reliability of the system and driver feedback.
NOHOME
UltimaDork
2/26/18 6:11 a.m.
Hydraulic is great for where steering feedback is not a thing. Boats, tractors and rock crawlers have been using it for a long time.
Wonder what kind of power you would need to operate two independent but digitally linked linear actuators? Course, you would probably die four or five times before you got this dialed in as a DIY effort.
Pete
I can't see how you'll be able to move enough fluid fast enough. You're basically transferring the fluid from a master cylinder to to a steering cylinder while forcing the fluid from the other steering cylinder back to it's master cylinder, all without the added flow rate of a pump. I doubt it would function any better than a manual steering rack, the force needed should be about the same and possibly higher as you try and force the fluid to return. Have at it though. I'm curious to see if the idea would work.
How about using a pair of big cables instead? One at the front of the axle, one at the rear, in order to only transmit force by pulling. Wind them around the steering shaft.
The DOT requires a mechanical connection. If this is to be street used, it will require a steering shaft and gear.
To meet current regulations you need a mechanical link between wheel and steering wheel. Without that it’s not street legal.
Would steer-by-cable be legal? I'm thinking it would use 3x 7x19 aircraft cables, 3/16" diameter.
If you're not set on any particular way of achieving this goal, you might want to do something like Infiniti's DAS steer-by-wire system. It's street legal although it does have redundant computers and a full mechanical backup system with a clutch - this would conflict with your desire for a bump-steer-free system.
BMW also has a system that can alter the steering ratio or allow a computer to alter the steering angle by using something like a differential gear on the steering column connected to a motor. I thought that system was particularly scary because it had a system to lock the motor in case of failure, returning the system to plain mechanical steering. If that locking system failed you'd just lose steering 
That BMW system is pretty cool. It's a planetary gear IIRC.
I teched a FV at a regional that had a double acting/ double rod hydraulic cylinder acting as a steering rack (like OP said). It was driven by another DA cylinder mounted w/ linkage at the steering wheel, no pump or MC. Quite innovative.
I shot it down.
Chief tech passed it.
I spose you could change ratio w/ cylinder diameters and linkages.
It worked, I wasn't crazy about it.
There is no need for this application to apply to any particular DOT or other legal specs so that's no concern.
DeadSkunk: The flow rate is something I had not considered so thanks for bringing that up. I suppose if the lines were sized large enough and the ports in both cylinders were not a restriction then it should be pretty free flowing. A conventional steering rack (albeit miniaturized for the application) is a non starter because of the unreconcilable bump steer.
fasted58: That is very interesting! I'd like to hear more about that. I guess it's not an original idea after all! I didn't see any "prior art" after a five minute google search. The tech inspection is a very good point (oh yeah it's getting autocrossed) with something out of the ordinary like hydro steering or cable steering. I mean when you think about it conventional steering has no redundancy in case of failure either, but we have a lot more trust of it because we're so used to it. There is a large amount of unknown with alternative systems and I guess the burden falls heavily on the builder to make others trust it. I can't recall seeing any rules in the SCCA solo rulebook about unconventional steering systems on a mod level build.
I have thought about cable steering as an alternative. It has some definite advantages if done well. Super smooth, super light, easy to set up geometry.
Another option is a bevel gear box as in the lower left of this picture (credit to Erik Zapletal). This would package very well and would be easy to tune bump steer out of. However, it has a major disadvantage of likely needing a custom gearbox or gears to make it work.
All that electric stuff is way too complicated for me. The concept for the car is simple (and also using unconventional methods).
Anybody else have any insight on cable operated steering? I'm actually kind of leaning towards that now. Just for reference almost every small aircraft ever made uses cables and pulleys for all the controls, many of them without redundancy.
Only thing that scares me about it is that the higher end of the force on small aircraft control surfaces is probably similar to the lower end of the force on a small car's steering system...if I were to try it, I'd use those aircraft cables though.
I think cables would get pretty complicated, and space consuming, pretty quick. Cable actuated aircraft control surfaces typically operate in a simple single axis pivoting motion. So no problem if you are running a rigid suspension. But once you add the more complex motions of the axle, it seems to me that it would be rather difficult to maintain constant and uniform cable tension through the full range of motion...And even if you figured out a way to do so, it might become a bit of a packaging nightmare.
Would a 'traditional' beam axle steering mechanism, which moves the linkage further outboard than a conventional steering shaft, circumvent your packaging constraints?
Driven5 said:
I think cables would get pretty complicated, and space consuming, pretty quick. Cable actuated aircraft control surfaces typically operate in a simple single axis pivoting motion. So no problem if you are running a rigid suspension. But once you add the more complex motions of the axle, it seems to me that it would be rather difficult to maintain constant and uniform cable tension through the full range of motion...And even if you figured out a way to do so, it might become a bit of a packaging nightmare.
Good point. Could be fine for a track car, but not for an offroader.
