Wishbones attachment question !

[DaveW]

I find this thread has turned into an interesting topic. Let me explain why.

1. When playing with active suspension, we used warp offset to turn steer angle, with other measurements, into a "trajectory" demand. It worked well, improving the cornering performance. I have described the control law, and shown video evidence of the effect, elsewhere in F1T.

2. It is not well known, perhaps, that we added rear steer to some active vehicles. That also worked well when driven by a version of the same control law. Interestingly, in a turn, the rear steer angle moved quickly to increase the yaw rate, and then returned equally quickly to (close to) zero as the turn developed (not good for rear passengers, because the initial acceleration was in the wrong direction).

3. The rear dampers on an oval car are often set with compression digression on the left, and rebound digression on the right (the cars only turn to the left). I was told that set-up gave the drivers "confidence", but I interpreted it as giving the vehicle turn-in oversteer, hopefully backing off as the turn developed.

4. I devised a rig test procedure in an attempt to quantify the effect. We discovered that many of our "optimal" set-ups resulted in turn-in oversteer, but not always. We thought that roll center heights might have had an effect, but had no way of measuring them on a simple rig...

An extract from a previous post stated:

Rear roll centres should be higher than the front because the "CG axis" of the car is inclined up towards the rear

I was always told that this was done to make weight transfer happen more quickly at the rear to allow the rear to "catch up" to the front during turn in.

Not meant as a personal dig, but that sounds like just another cult saying. The sooner people move away from religious, hand-wavy explanations and use proper physics then the discussion can move forward. I think also a lot of people would be disappointed with how small the delay is between the geometric load transfer and the elastic load transfer, especially on stiffly sprung race cars.

I was happy with Lycoming's comment, and his explanation. Tim commented on the size of the delay caused by the geometry. My immediate thought was the effects of damper "shape" would also be small, but it can have a significant effect on cornering performance.... according to my observations.

So why am I interested? I would like to understand better the "physics" of the turning maneuver and develop our test procedure into a more useful tool. Hence any reference to Lycoming's explanation would be useful. It might be useful to know Tim's production cars that do not follow the trend.

BTW, perhaps anticipating a response from Greg, some engineers do not like (relative) turn-in oversteer on a road car....

[Tim.Wright]

The warp discussion is interesting Dave as its something I'm researching. I found the posts but the video link is broken. Any chance you could throw the vid back up?

I don't know what sign convention you were using but it looks like you were calculating the current path curvature (yawrate/vel) and comparing it to the path curvature of a neutral car (Tan(SWA)) and then amplifying that delta by a function of latacc and yawrate.

At the risk of going off topic, I believe that warp loads have a big influence on the response of a car. I have shown here that the roll distribution of a car with a traditional suspension setup can change its elastic LLTD in the range of 4% per mm of warp input at the ground.

Given that a driver can feel a 1% change in roll distribution setup, this is a massive change which is uncontrolled because its coming from the road profile.

Coming back to the roll centres, I'm still undecided as to whether they have any effect on vehicle performance of if they are only affecting the vehicle dynamics in the "tactile" range which the driver feels. Its also something I'm studying at the moment but for now I don't have an answer.

Have you tried to measure the roll centre heights on your rig Dave? I'm not sure how reliable it would be but if you are able to put a 3 axis accelerometer and 3 axis gyro on each upright you could derive the contact patch and wheel centre trajectories with respect to the chassis and from this you can calculated the roll centre height.

But from there I think you are a bit limited in what you can do on a 7 poster. Without the possibility to apply lateral loads to the wheels it will always be difficult to get the correct roll and pitch responses.

[Lycoming]

@Dave;

I can't cite a reference for that, it was told to me by a now-graduated senior at my school's FSAE team when I asked why the rear roll center was higher. The reasoning he gave was simply that it's load transfer occurring through the suspension links and the chassis (which are "infinitely stiff") as opposed to the springs/dampers. Tim, you said that the delay is small on stiffly sprung racecars, I presume damping factors into it as well? (though I guess stiffly sprung kind of implies lots of damping). Would you say that the delay is also small for lightly damped, softly sprung vehicles? Or does it simply not matter on those kinds of cars?

@Tim;

Your plot shows that going over a moderately large bump, say 5 mm, would result in a pretty large change in LLTD, would that be a correct interpretation? If so, wouldn't doing so mid-corner be catastrophic? I mean, I know fairly stiffly sprung racecars manage to make it through the last corner at Sebring fine most of the time, and an LMP1 would, I presume, have a very stiff warp mode. Is it just because such an event happens very quickly?

[DaveW]

The warp discussion is interesting Dave as its something I'm researching. I found the posts but the video link is broken. Any chance you could throw the vid back up?

Here, if you have the patience, is the video. Note the red car appears to "lead" with its rear axle, and hence establishes the turn quickly, whilst the white car appears to "lead" with its front axle.

The consequence is obvious in hind-sight, the trick is to make a passive car do the same thing (without destroying rear grip on a rear drive car).... Arguably, achieving turn-in oversteer will reduce overall mechanical "drag" through a corner - which is, I think, why oval cars use it.

Have you tried to measure the roll centre heights on your rig Dave?

I haven't, for the reasons you stated.

I can't cite a reference for that, it was told to me by a now-graduated senior at my school's FSAE team when I asked why the rear roll center was higher. The reasoning he gave was simply that it's load transfer occurring through the suspension links and the chassis (which are "infinitely stiff") as opposed to the springs/dampers.

I found this. Are you there Ben?

BTW, F1 cars are said to be "stiff". This is a true statement for front axle heave & pitch, but not necessarily front roll. The rear axle tends to be relatively soft at lowish cornering speeds, anyway....

[firasf1dream]

guys, please tell me what should i study to understand what you're talking about :S !?

i am a mechanical engineering student, finished all my courses, still have one and now working on my project
but what you are talking about is very advanced and car's related, all i have studied about dynamics are 2 courses called Solid Dynamics and Solid Mechanics !

[Tim.Wright]

guys, please tell me what should i study to understand what you're talking about :S !?

i am a mechanical engineering student, finished all my courses, still have one and now working on my project
but what you are talking about is very advanced and car's related, all i have studied about dynamics are 2 courses called Solid Dynamics and Solid Mechanics !

All we are talking about is mechanics and dynamics.

I still stand by my advice not to read any automotive suspension books. But there is one good vehicle dynamics book which you need to buy and read before you should start looking at suspensions:
Milliken, Race Car Vehicle Dynamics

Read the chapter on the tyre then the one on steady state stability and control and you will 50% there. The knowledge in these 2 chapters is 90% of my work as a suspension engineer.

Then you can flick through somewhere near the middle and there is a picture of Dave.

[firasf1dream]

All we are talking about is mechanics and dynamics.

I still stand by my advice not to read any automotive suspension books. But there is one good vehicle dynamics book which you need to buy and read before you should start looking at suspensions:
Milliken, Race Car Vehicle Dynamics

Read the chapter on the tyre then the one on steady state stability and control and you will 50% there. The knowledge in these 2 chapters is 90% of my work as a suspension engineer.

Then you can flick through somewhere near the middle and there is a picture of Dave.

aha i have this book, Race Vehicle Dynamics and another one called Tune to Win
so i don't have to read the whole book ? can you please tell me exactly which chapters others than the one you named that i should read ?

so you are a Formula 1 engineer now ? or you were ?

[firasf1dream]

lol i found Dave Whitcomb's picture

[Tim.Wright]

lol i found Dave Whitcomb's picture

Wrong Dave (I think)

[firasf1dream]

aha so what do you say about my question ?

[Tim.Wright]

aha so what do you say about my question ?

I think you are asking too many questions.

If you really have finished all your courses, you should be at a level where you can flick through a book and decide for yourself what's important and what isn't. So do that.

If you want to learn about suspension, the best thing I ever did was to make a 3D wireframe model in excel where you can apply wheel loads and calculate the link forces. From this you will be able to learn pretty much everything you need to know about jacking, antis and roll centres (all of which are actually the same thing).

[firasf1dream]

aha so what do you say about my question ?

I think you are asking too many questions.

If you really have finished all your courses, you should be at a level where you can flick through a book and decide for yourself what's important and what isn't. So do that.

If you want to learn about suspension, the best thing I ever did was to make a 3D wireframe model in excel where you can apply wheel loads and calculate the link forces. From this you will be able to learn pretty much everything you need to know about jacking, antis and roll centres (all of which are actually the same thing).

well i did take a look at the book even before asking the question so i will tell u which chapters i decided to read and i need your opinion please

here are the chapters:
chapter 10 Race car Design
chapter 12 Chassis Setup
chapter 17 Suspensions Geometry
chapter 18 Wheel loads
chapter 21 Suspensions Springs
chapter 22 Dampers

these are the chapter related to Suspension that i saw that i should study, of course there is others like for Aerodynamics ...
so what do you think about that, is there others that i should read related to Suspensions ?

btw about my studies well very general and most calculations which are very useless when it comes to designing the suspensions and for a race car

and about the excel 3d wireframe ?! well excel is used for some calculations especially for management related so how come you use it for engineering ?!!

[firasf1dream]

hello guys,
any answers ?

[Lycoming]

You absolutely need to read the chapter on tires and steady state stability and control. The tire is pretty important to understanding vehicle dynamics. I would recommend chapter 8 and 9 as well. The chapter on springs is not really that useful unless you're designing springs. I think. I only skimmed it.

[firasf1dream]

aha ok great thanks a lot