How to Optimize Steering Arm Angle for Rear Steer and Ackermann in FSAE Design?

[mamta8]

Hi !
I am doing the steering of our FSAE(formula racing car). will be using rear steer with a lot of ackermann.

One particular strategy i came across for finalising stering arm angle is, you vary your steering arm angle (angle seen in plan between steering arm and uprights) so that your turning center is rearwards with respect to rear axle ,for some values of inner wheel steering angle and forwards for others.Please guide me if this is the right way of doing...

WAITING FOR A QUICK ADVICE.

thanks !

 

[Ranger Mike]

whoa there partner...do you know the proper chamber and caster for the steering set up?
have you calculated the true roll center from the suspension components?
are you familiar with Bump Steer? How much bump do you have in droop and rebound? How much ackerman are you running?
what it the ackerman differential between the inside and outside steering arms when in a turn?
do you have access to steering plates. ( these are metal plates that the permit proper measurement of degrees of rotation. )

If you do not know the answers to the above I think you are at risk of screwing up the steering geometry by changing a key component..which I have done my self...and results will be an ill handling race car...could even be fatal if you super stress a critical component..

if you are really interested in making the car go..first you got to master the factory setting then start making changes to improve the performance...DO NOT LISTEN TO SOME CLOWN IN THE PITS HOW HAS " THE HOT SET UP"...

 

[Cyrus]

Sorry, can't help you here and if I could I wouldn't since you are our competition.

 

[Ranger Mike]

Racers are a COLD hearted lot! :D

 

[xxChrisxx]

if you are really interested in making the car go..first you got to master the factory setting then start making changes to improve the performance...DO NOT LISTEN TO SOME CLOWN IN THE PITS HOW HAS " THE HOT SET UP"...

I assume he's designing from scratch, and thus has no factory setting Mike.

To OP, Steering and suspension geometry isn't exactly my thing but I'm going to be designing a chassis to F.Ford specs so I suppose this is time for me to start reading up myself.

What program are you using to calculate you suspension geomentry? The idea of varying it so you will get an acceptable range of turning centres sounds fine to me, but as Mike said this can sometimes have a detrimental effect on bump steer.

Why are you running so much ackermann? There is an autotest round iirc, is it for that?

 

[Ranger Mike]

i'll be happy to start a refresher class on this ifin you'all think it will help
could cover suspension layout and designing mounting points..instant center, roll center,

 

[xxChrisxx]

That would be helpful, I understand all the basics and principles and kind of know what I am talking about. The only thing I'm majorly lacking is experience in designing and what exactly I would want the car to do.

I'm going to start by designing the suspension anyway, so I can then design the chassis pickup points around them so I won't have to compromise the geometry.

I'm also thinking the aim of the game is to have as consistent steering as possible. So I am looking for zero bump steer and as little lateral roll centre movement as possible.To OP: sorry for hijacking the thread. Maybe we should start a new one. "Ranger Mike's Suspension Masterclass"

 

[mamta8]

I assume he's designing from scratch, and thus has no factory setting Mike.

I am a girl ,by the way...

It's our first car

the results for camber, cater and roll center ,i have attached..done by the suspension guy.
100% ackeramnn would not be a good option ,i guess ,as that would help the inner and outer wheel to roll freely but at low cornering speeds. For high cornering speeds the tires would develop high slip angles ,results in tire wear and ofcourse high temperature ,if run with 100% ackermann.

I would like to have akermann for low speeds/steering angles , parallel for straight running and anti-ackermann for high steering angles. But cannot figure out right now! how to achieve this...

can anyone of you , give a strategy for going about designing the steering

 

[mamta8]

I think 2 things needs to be done :
1. finding steering arm angle and also the angle that your tie rods would be making with the rack.
2. finding the tie rod lenghts

could anyone of you give a process for the same or some advice.

thanks

 

[mamta8]

Sorry, can't help you here and if I could I wouldn't since you are our competition.

It's fine Cyrus

But i thought...

KNOWLEDGE ALWAYS INCREASES ON SHARING

 

[xxChrisxx]

I am a girl ,by the way...

Oops, sorry bout that.

the results for camber, cater and roll center ,i have attached..done by the suspension guy.

That looks suspiciously like a susprog output (that brings back nightmares). I'm clueless about ackermann and how to achieve what you want. Looks like its back to reading Milliken for me.

 

[mamta8]

ya it's a susprog output ..

ok ! please tell me will reading MILLIKEN , give answers to most of my questions on designing...? do i have to go through the whole book or some parts will do.. (lack of time)

 

[Cyrus]

ya it's a susprog output ..

ok ! please tell me will reading MILLIKEN , give answers to most of my questions on designing...? do i have to go through the whole book or some parts will do.. (lack of time)

I must ask, are you guys using any books? This is an odd question for you to be asking if your trying to compete seriously. It sounds like you're just designing something and hoping for the best.

That's what our car looked like for those of you who are unfamiliar with SAE. By "our" I mean our school, I wasn't on the design team but will steal credit for their victory shamelessly.

 

[mamta8]

I have gone through tune to win (tires and steering )..have started with RCVD...

I have done theory part but seriously am in a fix as to how start the practicle thing..

 

[xxChrisxx]

That's what our car looked like for those of you who are unfamiliar with SAE. By "our" I mean our school, I wasn't on the design team but will steal credit for their victory shamelessly.

Christ Cyrus, can your head fit through standard width doors?

ya it's a susprog output ..

ok ! please tell me will reading MILLIKEN , give answers to most of my questions on designing...? do i have to go through the whole book or some parts will do.. (lack of time)

Well, no book you are going to read will give you answers to design questions diectly. Suspension is like aero, its a bit of a black art, and to be honest one that I never fully got to grips with. I always liked the big metal lump that is the powerplant.

It'll tell you what changes can be made, and had a detailed section on Ackermann. It's kind of up to you just to pick what you think will be best.To be honest i'd just keep it simple as, in my experience, when you start to make things overly complicated its when things start to go wrong.

Just try to keep the steering consistent, and avoid bump steer. A really simple steering that feels good is far far better than a complicated one that had a bad feel.

 

[KLoux]

Cyrus, are you still in school or are you a recent grad? I may have raced with you... Ever go with Drexel U. to any autoXs at Ripken Stadium?

Anyway, back to Ackermann:

First note: If you've got the steering arms intersecting behind the rear axle (with the wheels pointed straight ahead) then you've got anti-ackermann (the inside wheel will be steered less than the outside wheel).

Second note: The "traditional" percentage Ackermann is only valid when the steering link (tie-rod) and the steering arm are at right angles and when the wheels are pointed straight ahead. This isn't to say that it doesn't still give you some index as to how your wheels will steer when you turn the wheel, but you shouldn't consider this the end-all-be-all. Instead, I recommend plotting the actual steer angles and the difference between the inside and outside wheels, throughout the range of travel. I have found that this is very useful. Changes to the steering geometry can actually change the shape of the curve, not just shifting the curve up and down, for example. It is not a good idea to quantify the car's steering behavior with just one number...

If you have some tire data (which you should - check out the FSAE Tire Test Consortium for some high-quality, low-cost tire data), you can determine the "optimal" slip angles for each wheel of your car in a particular turn. To get the "exact" lateral acceleration, I suppose this is an iterative process, but I would start by assuming that your car will pull 1.3-1.5 G in the corners (without aero). This is also where you would account for your camber angles (after the car has rolled, plus the gains from KPI and caster) RCVD has everything you need to know about calculating the load on each tire from here. You can make it really complicated by getting involved with dampers, turn-in, etc., but to start, just look at steady-state turns.

Now that you know what your slip angles "want to be," you can plot steer angle vs. turn radius for your max. performance corners (assuming you're operating at your ideal slip angles). I recommend looking at only the two outside wheels - since they've got the majority of the weight of the car, these will be the dominant factors in determining the path of the car - now you just want to get the inside wheels to contribute as much as possible to the cornering force. What angle does the wheel need to be at to contribute as much as possible to the cornering force?

It shouldn't be difficult to plot these angles throughout the range of rack travel.

When looking at the tire data, don't forget to account for the slip angle when going from Fy to cornering force (there will be a "drag" component as well as a "useful" component).

Sorry, this is a little rushed... hope this helps!

-Kerry

 

[Cyrus]

Christ Cyrus, can your head fit through standard width doors?

It's hard to get race helmets that fit, .

KLoux, nah. I was in the Vehicle Dynamics Class with them, but I wasn't on the design team. I'm just proud of their hard work. My office is in the same building as their shop. Those guys were in there building that car every day till 3am+ for months.

 

[xxChrisxx]

Stuff...

That is a really good post.

EDIT:

And all this talk makes me jealous, the bean counters at my Uni just didnt want to put any money or time or effort into a Formula Student team. We had willing and able people but simply no backing.

[/rant]

 

[mamta8]

Hey Chris !
thanks for that advice ..

KLoux !
you have told me more than i was expecting from this forum (after reading the first few replies )... THANKS A LOT

will come back with some specific questions..

 

[mamta8]

100% ackermann is not a good option for FSAE cars as the inner angd outer wheels roll freely but at low slip angles ..BUT THE COMPETITION TRACKS SET UP ALLOWS ONLY FOR LOW SPEED CORNERING and ON THIS THE TIRES ARE AT SMALL SLIP ANGLES SO THEREFORE i think 100% ackermann is a GOOD OPTION...

 

[xxChrisxx]

100% ackermann is not a good option for FSAE cars as the inner angd outer wheels roll freely but at low slip angles ..BUT THE COMPETITION TRACKS SET UP ALLOWS ONLY FOR LOW SPEED CORNERING and ON THIS THE TIRES ARE AT SMALL SLIP ANGLES SO THEREFORE i think 100% ackermann is a GOOD OPTION...

Welcome to the black art of suspension and steering.

Rule 1. Everything is a compromise.
Rule 2. EVERYTHING is a compromise.
Rule 3. Swearing at Susprog helps it work.

 

[KLoux]

100% ackermann is not a good option for FSAE cars as the inner angd outer wheels roll freely but at low slip angles ..BUT THE COMPETITION TRACKS SET UP ALLOWS ONLY FOR LOW SPEED CORNERING and ON THIS THE TIRES ARE AT SMALL SLIP ANGLES SO THEREFORE i think 100% ackermann is a GOOD OPTION...

Actually, high-speed vs. low-speed corners are more to do with the lateral acceleration than the speed of the car (kind of a misnomer, I guess). Any kind of a racecar (even one that doesn't go "fast") should be designed for high-speed corners. The difference is in the slip angles: In low-speed corners, the slip angles are negligible (as is the lateral force and thus the lateral acceleration), but if you want to create lots of grip (which hopefully you do :-), your slip angles won't be negligible. Off the top of my head, I'll say that if you're running Hoosiers or Goodyears, your cornering force will peak at slip angles anywhere from 5-15 degrees. So you might be tempted to say "5 degrees, that's pretty small, isn't that negligible?" But NO! We're not talking about sin and cos here - the Fy vs. slip angle curve is very non-linear, especially in the range where the tires will be max. performing.

Have you been to any FSAE competitions? A couple of years ago when I was involved, our data system recorded speeds in the 40s and 50s (MPH) at competition. So it's certainly not as fast as a what a "real" racecar might see, but I would hesitate to call it slow...

Also, in my previous post, I forget to mention that it is important also to consider toe. Toe is something that is commonly changed during the tuning process, but it is important to consider what toe or range of toe you expect to run because it does affect the way the steering system works.

-Kerry

EDIT: And yes, EVERYTHING is a compromise :-)

 

[KLoux]

Don't have much time now, so this will be a quick one:

Carroll Smith's opinion in Tune to Win (or maybe one of the other ... to Win books - I can't remember) is that since a car is hard to push in the pits when it doesn't have 100% Ackermann, you must want 100% Ackermann. I disagree (but his books are still invaluable).

Determining the amount of Ackermann to design into a car comes from looking at the tire data. As the load on the tire increases, the slip angle at which the tire creates maximum cornering force changes. If as the load on the tire increases, the optimal slip angle increases, then you probably want positive Ackermann (outside wheel steers less than inside wheel). If this is the other way around, then you probably want reverse Ackermann. I say probably because it can depend on the actual numbers - and if you're designing for a specific turn radius (for say, a skidpad event?) you can always "perfect" your "static Ackermann" by playing with your toe setting, even if you're way off with the design of your steering geometry.

-Kerry

 

[mamta8]

hey kerry !

i am really short of words in order to appreciate your posts...they are wonderful

anyways my cup is still empty ...but do fill my cup with that knowledge of yours...
i know that's me being really greedy...thanks

best regards
mamta