Best steering column torque for car control

[murpia]

Anyone have an opinion on the minimum / maximum steering column torque for good car control?

I searched around the forum a bit and found that 70 ft-lbs is too heavy (http://fsae.com/eve/forums/a/tpc/f/125607348/m/81710797121) but no-one's commented on what's too light...

Obviously the steering wheel rim diameter is a factor also, but I would have thought most wheels are in the 300mm (12in) range?

Regards, Ian

[murpia]

Anyone have an opinion on the minimum / maximum steering column torque for good car control?

I searched around the forum a bit and found that 70 ft-lbs is too heavy (http://fsae.com/eve/forums/a/tpc/f/125607348/m/81710797121) but no-one's commented on what's too light...

Obviously the steering wheel rim diameter is a factor also, but I would have thought most wheels are in the 300mm (12in) range?

Regards, Ian

[Biggy72]

We've always used 10" steering wheels. I think before you hit a torque being too light, you end up with too much steering input. Last year we had to turn the wheel wayyyyy too much and it tired out the drivers as fast as having too high of a torque input. And it was much harder to drive because of it. Try to keep the wheel close to only moving 180º lock to lock.

This is all fine and good, but I'm not sure how much of a scrub radius and stuff like that you're looking for. Our first car had a pretty large scrub radius and while it had good feel, it would wear the driver pretty quickly.

[murpia]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Biggy72:
This is all fine and good, but I'm not sure how much of a scrub radius and stuff like that you're looking for. </div></BLOCKQUOTE>
Just interested in opinions or data on column torque / steering wheel rim force ranges that drivers are most comfortable with. The rest I can deal with...

Thanks for your info anyway!

Regards, Ian

[Biggy72]

You could just make a setup with a couple pulleys and weights, then have someone make about as many turns as there are in an endurance course and see how they feel. Compared to some of the measurements people are told to go find themselves this seems like one of the easier ones.

[John Valerio]

that wouldn't really work because self-aligning torque (i.e. steering effort) is a function of slip angle and therefore changes considerably based on driving style and how hard the driver is pushing. it's quite the sensation when the steering goes light right before the tires break away.

[murpia]

Derek: It's not so much how tiring the steering is for the driver, as you say that's easy enough to simulate with a few weights.

It's more as John describes, feeling the self aligning torque through the other forces (weight jacking and other steering / suspension geometry effects).

For example, let's assume that the self aligning torque represents 50% of the steering wheel force. There must be a preferred range where it's easier to feel that 50%?

Regards, Ian

[Steve O]

I wrote out a longer more technical post but decided against it as I have a better solution. As Derek suggested its not a bad idea to know what your drivers can handle without being fatigued... its sort of a monte carlo analysis of the drivers assuming worst case conditions. After that I'd say play around with something between that number and that of what it takes to turn the average power steering assisted wheel. You obviously want it higher than that because there is not much feel with power steering, and you want it lower than what would fatigue the driver under worst case conditions. To give you a ball park of where you should be, at the highest resistance through the corner you should be looking at 15ft/lbs of torque at the drivers hands. Remember this is a peak number though, not a linear number. To reduce wandering you will need to create a little bit of zero g steering effort and you want it to initially DECREASE as you enter the corner. This is a little unsettling at first but is essential to let the driver know he's started to corner and the weight transfer has yet to take place. This keeps the driver from oversteering on entrance to the corner (do to the sudden shift of weight across the roll couple). As the weight transfers and you pull harder g's the torque should roll up to that 15ft/lbs giving the driver the progressive feel that he anticipates as he goes into the corner and this will decrease with traction decreasing as anticipated.

Hope that helps...unfortunately as you noted steering is a very dynamic event and changes with drivers which makes it a very difficult thing to approach; however, if you approach it from the cars abilities you should find yourself in a good area to start. If the 15ft/lbs is too light for your drivers go up to 20... if its too heavy, get new drivers. It will be helpful if you plot your Alignment related torque vs. normalized slip angle to give you an idea of how much of a change each driver will see based on their approach to corners.

Steve

Edited Note:
11-15ft/lbs is an average for most road track cars. I aired on the high side because our cars are smaller and lighter and we don't travel as far. However, you do also have to consider our cars aren't quite as comfortable to drive in some other respects since they are generally not so perfect prototypes.

[John Valerio]

interesting approach steve. you dont have to reveal anything about your setup, but how are you achieving an increasing steering effort with increasing cornering accel? typical aligning torque curves peak at around 3 degrees slip angle depending on normal load, whereas lateral force peaks much later than that. the aligning torque even shifts negative at a point that is dependent on normal load, something i got to experience in a car without enough mechanical trail. the way i see it, if the driver is experiencing increasing steering effort with cornering accel then he or she isn't going fast enough.

[Steve O]

I don't really have any suspension setup secrets so I will go through how to achieve some of the stuff I talked about.

You can get your suspension to do anything you want it to as long as you know the effects of each component of it. For instance if your roll axis is above your c of g your car will actually lean into the corner instead of out of the corner. This isn't typically done because you would freak your driver out!

The same effects can be created with steering effort, you just have to understand what actually affects the steering effort. As you noted trail is your primary factor of what you feel in your hands and negative trail doesn't make you feel so great when it isn't timed right. You are right, as you go into a corner and lateral g's increase, your trail is negatively affected and in some cases will go negative, and this decreasing trail does decrease steering effort. There are other things that affect steering effort though; for instance, pos. caster angle changes will increase your steering effort while decreasing straight line effort. Caster is a little tricky as that actually effects quite a number of physical and virtual dimensions in your suspension so be careful with this one. Camber has a huge effect on your trail/KP moment under small slip angle conditions (zero g lateral conditions down the straight away). Doubling your camber increases your zero g effort by 35%.

From there I am sure you can figure it out... it all then lies on your link geometry to cause appropriate camber and caster adjustments during bump and droop. Obviously though, the ability to setup your suspension like this will lie on your KPI and upright geometry.

Steve

[John Valerio]

steve, thanks for the reply. a few comments about your method.

yeah i agree that you can make your suspension react in a lot of different ways with the geometry, but most of the methods lead to undesireable side effects. i'll leave the high roll centres for another discussion, but to achieve a changing caster angle you need to introduce some antis into the geometry which may or may not be what you're looking for, also it will introduce another variable into your diagonal weight transfer and also the slope of the steer-camber gain curve will be a function of pitch and roll angles, whereas with constant caster the pitch and roll angles just provide an offset to the curve.

by "doubling you camber" i assume you mean going from for example -1 to -2, and yes this does increase your effort, but it makes it go more negative. not to mention that the "zero G" lateral condition only applies right up to the point where you turn the wheel, at which point the aligning torque shoots up and then back down as you sweep through the slip angle range.

all that aside, i think it's important for us as suspension designers to cater to our competition specifics. what i mean in relation to this discussion is that we, with a few exceptions, aren't race drivers. the more you can get your suspension to help the driver out the better, be it through linear rates, minimal jacking, etc. thats pretty much our logic behind using the aligning torque drop off as an impending break-away signal to the driver. smith and milliken both talk about it, and i've heard it from some other teams, so i know that we're not doing revolutionary work here, i just think it's a good idea.

[Steve O]

I agree with everything you said. The key with anything you change on your suspension is moderation. The thing is that either way, when you go into bump and droop, you are most likely changing both camber and caster if not physically, virtually. I am only saying that controlling this to what you want with proper geometry will give you a more desired natural feel. If you control the amount that your trail is reduced while cornering the increasing g force will increase on the moment about your kp and trail and create slightly more resistance as your corner harder (as long as your trail didn't reduce more than your lateral force increased). I do agree though with your first statement about speed though, as you hit corners faster, it will cause the higher g cornering to be easier. This is a desired effect though as you will probably be approaching traction threshold at higher speeds with the higher slip angle...looking back at your post I think you misunderstood this about what I was saying with my setup, I think we are pretty much in agreement on it. I am saying that it should increase through the initial portion of your slip angle, to keep you from oversteering initially, and as you get closer to your threshold slip angles your steering should get easy to give the driver the sensation of less traction.

My thing about the zero g lateral condition is only about applying some force to the wheel under straight line conditions to prevent wandering when attempting to go straight.

[John Valerio]

yeah sounds like we're on the same page. now all we need is the TTC Round 3 moment data to be fixed...

[Patrick]

SteveO,

You said that as you get closer to your threshold slip angle (slip angle to achieve max lateral accel) your steering should get easier. This will almost always be the case. The slip angle at which the maximum self-aligning torque is reached is generally smaller than the slip angle corresponding to the maximum grip. So, as a driver you should be aware... the steering getting light doesn't mean you're out of grip... it usually means that you're allllmost at the maximum lateral grip.

-Patrick DeGrosse Jr.
NASA - Jet Propulsion Laboratory

[Steve O]

I didn't say you run out of grip... it is a sign that you are approaching your grip threshold... how light the steering gets before you break free is obviously something that will change per car and your driver will have to be familiar with it.

Steve O