Four wheel steering

[clausen]

Hi there,

Have many FSAE cars used it? To counter steer or compensate for rear slip angle?

What are people's thoughts?

Paul Clausen
Uni of Adelaide

[clausen]

Hi there,

Have many FSAE cars used it? To counter steer or compensate for rear slip angle?

What are people's thoughts?

Paul Clausen
Uni of Adelaide

[woollymoof]

Haven't used it, but it is very complex. Make sure that you can prove it with numbers and lots of them.

[Scott Wordley]

I heard UTA were trying to implement it for the US comp. Might be ready for Australia 03. I'd be interested to hear how UTA are doing it straight mechanical or hydro/pnumatic controlled?

Maybe Ken or Dr Bob could explain it some more

Regards,

Scott Wordley & Roan Lyddy Meaney
Monash FSAE Wingmen
http://www-personal.monash.edu.au/~fsae

[Bob Wright]

yeah, im interested to see what they run it off- is it from steer angle with understeer gradients mapped in to give a near zero chassis slip/yaw angle? or is it trickier using slip angle sensors as well?

I could see a big benefit in transient time on the tight corners for fsae cars with 4ws. I think most of the complaints about understeer in these cars is literally the distance required to generate the chassis slip to steer the rear wheels to steady state cornering- which can take two or three wheelbase lengths on the track, even for a good fsae car. in a corner with a radius of only 7 or 8 meters its like trying to set up a school bus to race at Monaco.

We're trialing a left/ right brake bias (rolling pedal and independant l/r circuits)to try and get a bigger yaw moment on corner entry, but it still looks pretty half assed compared to a 4ws working properly.

I cant wait to see if the UTA car does it well- if it does we'll copy it for sure http://fsae.com/groupee_common/emoticons/icon_wink.gif

any details would be appreciated

Bob Wright
Monash University
Australia
http://www-personal.monash.edu.au/~fsae

[clausen]

If for instance we wanted to steer the rear wheels in whatever way according to slip angle, couldnt we just use a lateral accelerometer as a reference? I mean everything else being the same (changing tyre pressure would throw it out) slip angle is roughly proportional to cornering force isnt it?

I'm only just starting in this game. Turn in a pretty common problem is it?

So maybe some counter steer in the transition followed by neutral or some "same steer" mid corner might help?

Paul Clausen
Uni of Adelaide

[Bob Wright]

lateral acceleration can be the same for whatever radius corner your on- so the car needs some other way of knowing (slip angle, steering etc)

Bob Wright
Monash University
Australia
http://www-personal.monash.edu.au/~fsae

[clausen]

Oh I see what you mean Bob

I was just thinking about the case where you would use rear wheel steer to compensate for slip angle so that the rears would be really going straight instead of at the slip angle.

Obviously a lot more complicated when using countersteer to help turn-in.

[Bob Gliege]

Check the SAE published reports. One day while doing some cleaning of our facilities I came across a published SAE paper written by our old faculty advisor, Prof. Dale Caulkins. The report tested 4 wheel steering on a formula SAE car. I don't remember the details of the report, but it concluded that the 4-steering system they tested didn't improve the performance of a formula SAE vehicle. This was done back in the mid '90's, and unfortunately Prof. Caulkins passed away in 1999. If you can't find it, myself or someone else involved with UWFSAE might be able to find it. It MIGHT be document number 942523 "Lateral Response of a Formula SAE Race Car." I'm not sure that is the same paper, but it could be. I don't remember if the system they used actually turned the rear wheels, or if it was based upon suspension geometry. I just thought I would let you know that there might be some papers out there that might help you. Best of luck,

Bob Gliege
UWFSAE '01-'03

[Charlie]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by clausen:
Oh I see what you mean Bob

I was just thinking about the case where you would use rear wheel steer to compensate for slip angle so that the rears would be really going straight instead of at the slip angle.

Obviously a lot more complicated when using countersteer to help turn-in.<HR></BLOCKQUOTE>

From what I understand, you have to have slip angle to have grip. You want slip angle. You can't make the rear tires at zero slip angle with rear steer. http://fsae.com/groupee_common/emoticons/icon_confused.gif

-Charlie Ping
Auburn University FSAE (http://eng.auburn.edu/organizations/SAE/AUFSAE)
5th Overall Detroit 2003
? Overall Aussie 2003. http://fsae.com/groupee_common/emoticons/icon_smile.gif

[clausen]

From what I understand, you have to have slip angle to have grip. You want slip angle. You can't make the rear tires at zero slip angle with rear steer. http://fsae.com/groupee_common/emoticons/icon_confused.gif

In the scenario that im talking about, the tyre would still have its normal slip angle relative to the wheel, but then you steer the wheel towards the inside of the corner by same angle as the slip angle. Therefore the rear goes exactly where you want it instead of in the slight drift that you always get because of the slip angle. I was reading that this is what Williams were trying to acheive with their 4ws F1 car back in the active days.

Regards

Paul Clausen
Uni of Adelaide
(Australia)

[woollymoof]

Clausen, I get what your are saying, but have you considered that this situation might be very unstable?

[gug]

whats wrong woollymoof, afraid of a bit of wheelspin? http://fsae.com/groupee_common/emoticons/icon_wink.gif (i know an unstable car is in general slower, im just joking)
im really interested in 4ws, just cause i generally think that the best way to go around a corner cant be by just keeping the rear wheels straight.
but ive only started f-sae, and really dont know much, especially about suspension design and tyre mechanics. im in the line to read the "to win" series, and also a text by milliken, called racecar vehicle dynamics, but does anyone know of a good source on the internet for learning about suspension? ive found a few very basic sites describing slip angle, but nothing advanced enough. also, i probably wouldnt know if someone was bullshitting.

read an article on the benneton 4ws system in f1, and they used rear steering angles of about 5 degrees i think. would these steering angles be larger in f-sae? i imagine trying to turn at the most 5 degrees would require a very precise system, and would be tough to set up.

"I come from a land down under,
Where beer does flow and men chunder"

[Angry Joe]

Lehigh's cars have been steerable by the rear wheels for years. We use a device called a gas pedal.

In all seriousness, you're better off putting the time and effort into learning to be a better driver, unless you're really not busy enough as it is.

Lehigh Formula SAE Alumni
Team Captain 2002-2003

www.lehigh.edu/~insae/formula (http://www.lehigh.edu/~insae/formula)

[Charlie]

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>Originally posted by clausen:
In the scenario that im talking about, the tyre would still have its normal slip angle relative to the wheel, <HR></BLOCKQUOTE>

Can you explain that? I don't understand the statement.

<BLOCKQUOTE class="ip-ubbcode-quote"><font size="-1">quote:</font><HR>but then you steer the wheel towards the inside of the corner by same angle as the slip angle. Therefore the rear goes exactly where you want it instead of in the slight drift that you always get because of the slip angle. <HR></BLOCKQUOTE>

If you steer the same angle as the slip angle, you'll simply get a new slip angle. The front tires are at a slip angle too, yet no amount of driver correction (save going REALLY slow) can correct this. Same with the rear. Maybe I am missing something?

[woollymoof]

Internet learning is very dangerous, stick with the books. You said that you are new to FSAE and because of this I think you should try to fully understand 2WS and everything to do with tyres and suspension geometry first before embarking on such a huge project.
Don't believe everything I say, I'm just an engine guy, but I went to Claude Rouelle's seminar the other weekend and at the moment you have a lot to learn and understand. Sit back and have a look at any designs that you might have at the moment and find answers to questions about why you did anything and everything. Judges will insist on answers to these questions.

[clausen]

Charlie,

Picture one of the diagrams that expain what slip angle does. you have the direction that the wheel is 'facing' (in a 2ws car, the rears are in line with the chassis. You also have, due to the slip angle, the direction that the wheel is moving in, which is a few degrees toward the outside of the corner. What williams were doing was turning the back wheel so that the slip angle vector is in line with the chassis.
Similar thinking to the arguement for anti-ackerman.

Woolymoof, good point about stability. Hadnt even thought about that http://fsae.com/groupee_common/emoticons/icon_smile.gif It could all end up with a car with a rear end that oscillates around like crazy! Scary thought.

For FSAE applications I'm just wondering if a little countersteer on turn in might be useful, but, like someone else said, it can all be done with the right chassis balance and throttle/brake steering.

Regards

Paul Clausen
Uni of Adelaide

[clausen]

Charlie, also the same thing happens on the front, the driver just compensates for slip angle automatically. Under hard cornering the wheels are not aligned tangentially to the corner, the slip angle directions vectors are.

Not sure if im making myself any clearer or not http://fsae.com/groupee_common/emoticons/icon_smile.gif

Regards

Paul Clausen
Uni of Adelaide

[Bob Wright]

steering wheel steers the front wheels- front wheels steer the chassis- chassis steers the rear wheels. That all takes distance, time, and yaw inertia. The tire always need slip angle to make cornering force but the chassis doesnt unless you tie the rear wheels to it.

the driver automatically compensates for the change in front wheel slip angle relative to the ground as the chassis starts to get a slip angle by steering out of the corner. If the chassis has 0 slip, then the driver doesnt need to do that at the front, and the computer doesnt need to do that at the rear- but that all sounds a bit simple. its just about how close you can keep the turn centre to a normal line form the cg. If the rears dont steer enough then there is chassis oversteer (my definition) and if they steer to much there is chassis understeer- with neutral being the best as the transient inertia thing is the lowest.

its all about what rate the fronts vs rears build cornering force relative to a point on the normal line from the cg that is the driver intended turn centre (roughly proportional to steer angle). If the f/r forces buid evenly toward the same point, then all you need to do is give the f/r forces enough difference in magnitude at the start and finish of each transient to get the rotational velocity of the car set for that corner.

The complex bits are all in the detail. Drive load makes moments that have to be accounted for (oversteer on small radius and understeer on big radius) and different tires at different temps have differnt cornering stiffness etc..

I think benetton had a dial that was a simple multiplyer for rear steer to compensate for tires going off and on.

to get the best combination of car and driver for the fsae comp, as Joe said, you are better off keeping it 2ws. But i really hope someone does the groundwork on 4ws to change that soon.

until then it looks like big wings, left/ right brake bias and 4wd are all we have left to make massive improvements in

Bob Wright
Monash University
Australia
http://www-personal.monash.edu.au/~fsae

[Denny Trimble]

Interesting discussion! If I remember correctly, the UW 4WS paper Bob mentioned described "in-phase" and "out-of-phase" direct mechanical linkage of front and rear wheel steering. I remember seeing the '93 car with this linkage on it.

The paper claims out-of-phase is faster through a low-speed slalom, up to a certain speed, at which in-phase if faster. In summary, I believe it claimed an active system transitioning from out-of-phase to in-phase would be required to be of benefit.

I also believe the new GMC/Chevy truck with 4WS uses an active system.

Also, look for the "no steer by wire" rule in 2004. This would kill any active 4WS systems, I believe.

Hope this helps.

University of Washington Formula SAE ('98, '99, '03)

[Pavan Dendi]

UT Austin had 4ws on our 2002 car. It is switchable from parallel to opposite steer but not dynamically, and it is based off of steering angle. When we drove with it enabled, we always had it on opposite steer since we always run tight courses.

It significantly reduced steering effort, and it was simply terrific through the slaloms. The problem with it is that you cannot power out of corners. You have to pretty much have the steering wheel straight before you could get any real acceleration without the back end stepping out.

We liked geting on the gas early so 4ws was abandoned for the 2003 car. Maybe turning the rear wheels less would have helped the problem, but personally, I don't see 4ws shortening lap times that much for FSAE. For wider and higher speed tracks, it has the potential for large gains. Maybe this was why it was banned in F1.

_____________________
UT Austin FSAE
http://www.me.utexas.edu/~fsae

[woollymoof]

Have a look at this, I haven't actually looked at it myself so I don't know how good it is. http://210.101.116.115/fisita/pdf/G344.pdf

[woollymoof]

and this http://210.101.116.115/fisita/pdf/H257.pdf

[gug]

most people here seem to be discussing an active system that would have to have quick turn rates. how do you think this could be actuated? is there a servo thats both big enough and fast enough to do it? especially since there would now be torque steer, like in front wheel drive cars.

also Pavan, how did you guys justify going with 4ws to the judges?

outa curiousity, anyone driven the 4ws honda prelude?
http://www.preludecrazy.com/images/mylude/front_side4.jpg

i am the grand master of procrastination. if i was this good at kung fu, i could kill you with a look.

[clausen]

Woolymoof,

You legend,

Regards

Paul Clausen
Uni of Adelaide

[Angry Joe]

When we investigated this, we considered a large stepper motor. I believe they made sufficietly strong and quick units, however they were quite heavy

Lehigh Formula SAE Alumni
Team Captain 2002-2003

www.lehigh.edu/~insae/formula (http://www.lehigh.edu/~insae/formula)

[Travis R]

IIRC, the judges in the design presentation in 2002 seemed uninterested in the 4ws, so we didn't have to justify it at all!... one of the reasons it was dropped for '03.
Like Pavan mentioned the car was awesome through the slaloms. The car had enough rear steer that if you missed the cone with the front tire, then you would miss the cone with the back tire. The car was a breeze to get through the tight stuff. The only time the car has been unstable was when I've taken it out to local autocrosses where they have much higher speeds. And that's with the rear steer set to parallel. But that's mainly because the suspension/steering geometry was designed to turn quickly at low speeds (1st and 2nd gear VS. 4th and 5th)
You really have to think about the kind of course you drive on at competition. If you can get a short enough wheelbase or a narrow enough rear track, then you wouldn't need 4ws. Our '03 car had a narrower rear end than '02, another reason we dropped it.
Reliability! Our system was hydraulic, and was exposed on the rear of the car. One misplaced foot while pushing the car around the paddock could cause a leak. I was personally called at 2am the day before the autocross to fix one such leak... Sure I was pissed, but I was the also one of the drivers for autocross, and I knew I wanted it active. http://fsae.com/groupee_common/emoticons/icon_smile.gif We made an aluminum "shield" to protect the components.
Having a "dynamic" system that can change from opposite to parallel steer is almost pointless. There are maybe 2 corners in the autocross/enduro that have high enough speeds to justify parallel steering. Everything else is so slow and tight that opposite would be more desirable.
Good luck

Travis Rouse - Test Pilot - The University of Texas (Austin)
http://www.ls1power.net/Travis/RX7/autox4_slide.jpg

[D-Train]

Digging up an old thread, i know...

I just noticed Firenze Race Team ran what looks like a fully active rear steering rack (it's electronically actuated anyway). Anyone know how well it did?
EDIT: Just checked the results. 2nd overall in both Formula Student and FSAE-Italy.

For the record, Swinburne ran a 4WS system last year ('05) at the Aus event. I can't remember who else has successfully ran them, but there's been a few more.

Pics of the Firenze rear end:
http://pcm.dmti.unifi.it/formula_student/ITA/IMMAGINI/2005/car/P6240195.JPG

more at http://www.firenzerace.too.it/

[markocosic]

They were not second in Formula Student UK 2005 - Helsinki were.

Something also tells me (nagging feeling, but I didn't take pictures or notes) that the system was removed and locked for the dynamic events (they were certianly having trouble with it/taking it to pieces on the day) - you'll want to check with the team whether it actually ran on the day I think.

http://web.mit.edu/cosic/www/FSAE-UK-2005/target328.html

[D-Train]

Oops... must have misread that somehow.

It looks like it was enabled in some of their footage (http://pcm.dmti.unifi.it/formula_student/ITA/Media.htm) - maybe I'm just seeing what I want to see.