Steering with Pitman Arm.

[MCoach]

I briefly considered go-kart type linkage, but found that without attempting to compromise desired geometry, either the forces were too high or the linkages would bind around the steering shaft due to their close proximity. The steering system ended up weighing nearly nothing anyway, so I wasn't too bothered by over complicating the steering system with a rack & pinion system.

With regard to EPS, this is likely a more feasible option. It's difficult to find a sufficiently sized engine that has PTO (Power take off) shaft for accessories. Monash may be a bad example because fitting an alternator rather than upgrading the stock stator is more of an outlier.

"The key here is to get the "whole car set-up" right. Briefly, the car should be stable with "hands off the wheel" (in simple terms this means toe-in front and rear, stiffer cornering-compliance rear-tyres, positive static-margin (*), etc.). Then, as soon as the driver signals "go left", the car becomes "unstable" to the left (front-wheels turn left and develop big toe-out, etc.). When driver lets go of HW the car proceeds stably in its current direction (well, after a minimum amount of yaw overshoot)."


I would disagree that toe-in front and rear is a good idea on these cars. Quite the opposite really. I think the car you will find that outperforms this one is the statically unstable car that is ready to dance with the devil but still on a leash. Toe out front and rear, with a large, fast available control input from the tire construction itself. Rear steer could aid with this strategy. This could double down with "side force" generation from big MF endplates, restoring vehicle stability, even though you are "hero driving" into the corner nearly sideways and at full throttle.
Soft tires for life.

Kevin, I want to say I saw one of the Canadian Hybrid teams with a similar set up. They did so to clear their front electric motors.

[Claude Rouelle]

Kevin,

No worry about compliance? My experience is that the more you add parts the more the causes for flexibility...

The long link between rocker and the steering bracket (that is bolted on the upright) tension, compression, buckling.....the rocker itself, the brackets that hold the rockers..... Was a load case made?

Claude

[Claude Rouelle]

From the reading of all these posts (and especially the one of mdavis) it seems that steering torque calculation is not (or at least was not) part of the design process... I find that a bit worrisome.

[Z]

I would disagree that toe-in front and rear is a good idea on these cars. Quite the opposite really. I think the car you will find that outperforms this one is the statically unstable car that is ready to dance with the devil but still on a leash...

MCoach,

And what is that "leash"?

A steering linkage that has "static toe-in" plus some self-centring from positive Trail gives the car good stability when travelling straight-ahead. If this linkage quickly develops large "dynamic toe-out" as the wheels are steered away from straight-ahead, then the car has rapid turn-in from this destabilising effect. Win, win.

The "static toe-in" is the leash. The "dynamic toe-out" unhooks the leash. How else would you provide the leash?

The "computer controlled, rigid HW, active steering" system I described (ie. like a fighter-plane) would do the same as above, possibly also on the rear-wheels, but with minimal physical effort from the driver.

Z

[BillCobb]

Kevin,

No worry about compliance? My experience is that the more you add parts the more the causes for flexibility...

The long link between rocker and the steering bracket (that is bolted on the upright) tension, compression, buckling.....the rocker itself, the brackets that hold the rockers..... Was a load case made?

Claude

Now Clausde, look at the bright side. These flimsy mechanisms, loaded by discalculated forces and ignored moments will result in highly understeering cars that will perpetuate the textbook notion that oversteering cars (as determined from using ONLY tire force/weight data) are the best setups for any form of vehicle "dynamics" based on a "statics" representation. Then, student and armchair designers of cars having lots of 'yaw overshoot' and characterized as lack of 'yaw damping' can continue their religeous zeal towards the perfect roll center migration strategy. Now someone with a brain (at least the left side) tell me how an oversteering car having only first order (as in exponential) yaw velocity and sideslip responses can have any sort of overshoot going on. Oh, that's right... a first order response has it's peak frequency at 0.0 Hz. The same pre-'engineer' needs to tell me how a higher order roll responce with a fixed damped natural frequency will interact with this newly characterized yaw peak frequency as the yaw peak frequency changes from being below the roll frequencey and migrates to above the yaw frequency as the speed changes (actually speed squared). Why hasn't anyone connected the dots between roll natural frequency and the speed dependent^2 yaw natural frequency? Is the roll center religion so pius that it's effect' on roll dynamic interaction with the other state variables just so much unorthodox heresy? Then we add the soggy stiffness of a torque sensor in a hydraulic, electric or spaghetti power steering mechanism. As for steering effort, Just take a moment and add up the moments. Right now, some of the statements we are reading are being shot from the ankle.

Man, you folks need some Control System Engineering education instead of the agricultural zone you are in. If any of these cars ever get measured at an event, make sure there are enough ambulances to cover all the attempted suicides. Just sayin'...........

[BillCobb]

The beatings will continue until morale improves. Remember, I measured the understeer of my freakin outboard bass boat using only a VBOX. Different with 3, 4, or 5 bladed propellers. This is high school project work.

[MCoach]

Bill,

better?
I can't seem to load a bigger picture than posted...

[Kevin Hayward]

Kevin,

No worry about compliance? My experience is that the more you add parts the more the causes for flexibility...

The long link between rocker and the steering bracket (that is bolted on the upright) tension, compression, buckling.....the rocker itself, the brackets that hold the rockers..... Was a load case made?

Claude

Claude,

Obviously slop and compliance were both forefront in the designer's minds. Calcs were done, as well as careful design of all components in the path. There were many iterations of this and various other steering systems when the car was being designed. Despite a typical floor mounted rack and pinion (with spur gears) may have fewer components and simple installation it has big problems with slop, compliance, and packaging. The initial design was to run a Pitman arm straight off the planetary, but could not get bump steer working well without having a high nose (big COG penalty).

Compared to previous R&Ps with double UJ's the design has both less slop and less compliance. The planetary gearbox has less slop than a rack and pinion, and the rockers/rods are all sized similar to a push/pull rod suspension system. Rockers were designed with reasonably long actuation radii to minimise the effect of slop. All holes in sufficiently thick material reamed to size. The long rods are not as small as most steering tierods being used and are designed with buckling in mind. The bolted steering bracket at the upright is not a problem at all (nor should there be any reason that it is) and the upright structure is significantly reinforced around that area. Loads were measured on the car to validate load calculations.

I think that a lot of the teams are not aware of how much the universal joints are deflecting during actuation.

Note that the extra parts added (compared to a conventional setup) are the rockers and one more tierod, in the process no universal joints were needed.

The image posted was the 3rd version that was manufactured after a number of improvements to reduce compliance, the team has since further improved the system. The 2014 car with that system had the least slop an compliance of all the ECU vehicles. this includes 2 vehicles with a drop box to a R&P and 3 vehicles with 2 UJ's to a R&P.

It is not a perfect system by a long shot but does a pretty decent job of offering steering geometry design freedom and significant lowering of the COG.

Again I am passing on details of student work here. While I am present at design reviews and advise in a few areas I may be short on a few of the details. I have driven the car and found it to be quite a good steering system, with no noticeable problem with slop or compliance. Sitting in the car you wouldn't be aware that the mechanism is any different to a well setup R&P apart from having a little less slop than most I have driven. One real bonus is that over time the R&P designs tend to wear pretty bad, both in the UJ's and rack and pinion itself. In this system the gears are generously sized and with 4 contact points there has been no noticeable wear over 2+ years. It still has the same performance as when it was installed.

But for those still in doubt do the calcs/design/construction/testing and see for yourself.

Kev

[MCoach]

The leash being selecting tires that are capable of generating a high enough Mz/steer gain to save it, something I never found quite good enough in the 13" R25B tires to like them.

[tromoly]

Kevin, I want to say I saw one of the Canadian Hybrid teams with a similar set up. They did so to clear their front electric motors.

Not my picture, this is from Formula North 2013, could this be the car you're thinking of?