Remedying understeer

[romaier]

Hi all,

I've spent much time perusing this forum but never really posted. Now, however, I have a question regarding an understeering problem.

I'm designing the front suspension of an off-road Baja SAE vehicle (in conjunction with the rear end team) for a design project at Uni. The lastest generations have had a nasty understeer problem, particularly at turn in. Some of the guys believe that by adding more roll in the front relative to the rear by means of a lower roll center will help this.

I, however, am not sold on this. I am a novice though, but I have been reading up on books such as "Race Car Vehicle Dynamics" by Miliken & Miliken in order to get a better feel for this. Am I misunderstanding something here? The old car has a wheel base of 64" with the new model shortened to 61". Also, the track is 58" being narrowed to 56". What would you all recommend as a remedy for this understeer problem?

Thanks

 

[GregLocock]

Well, what does a bicycle model tell you to do?

You might want to increase the front roll stiffness to increase the 'bite' of the outside wheel.

You might want to soften the front roll stiffness to increase the cornering stiffness of the outside wheel.

You might want to reduce the ackerman to get the outer wheel working harder

You might want to increase the ackerman to get more turn in drag from the inner wheel.

Interesting isn't it?





Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[Tmoose]

What's the F/R weight distribution?
What kind of tires do you / can you use?
Got any pictures?

 

[Aday]

I think Greg is saying...the solution(s) to your problem is dependent on the specific design of your vehicle and operating conditions.

I will say that reducing the front RC will reduce steady-state Understeer, but may make your turn-in problem worse...

 

[cibachrome]

The presumption is that the front is the problem. Often lack of turn in is a rear problem cause by the solid diff and not enough nose weight. And too much grip on powdered dirt. Under braking, not enough rear bias can restrict your yawrate.

For your job as front suspension release engineer, make sure the static and dynamic geometry makes best use of the front tire grip on the surface you want to excel on: dirt, hills, or the spalled concrete rocks. Plenty of wheel travel and a good shock compression front/rear ratio will get the car to turn.

Mske a tire tester using a local railroad siding and an A frame that you can measure reaction forces under various steer, camber, load and pressure settings. The results will startle most judges. You may find that you need a different tire profile up there if she still won't cut.

 

[carcrazy]

I too am a frequent reader of this site and have learned alot from all the posts but just when I think I finally understand something I get confused.

Aday said "I will say that reducing the front RC will reduce steady-state Understeer, but may make your turn-in problem worse..."

I thought that for a simple model with all thing being about equal (tires, cg location, etc) lowering the roll center on one end of the car would increase the steady state weight transfer across that axle and therefore that axle would have reduced cornering power.

For the suggestion above, wouldn't the steady-state understeer of the vehicle increase?

Thx

 

[Aday]

Raising the RC will increase the amount of load transfer that is carried through the suspension links, therefore increasing the percentage of total lateral load transfer carried at that end.

Unfortunately every vehicle dynamics book I have seen gives an explanation of RC that is incorrect to varying degrees. Most of the confusion seems to stem from the name 'roll center' which incorrectly implies that the vehicle will literally roll about this point. However there are some very good SAE papers on the subject that are accurate. Assymetric Roll Center's by Bill Mitchell comes to mind...

 

[Warpspeed]

Greg has given you one clue.

The other clue is front/rear static weight distribution.

Is there no traction at the front because the outside front tire is insufficiently loaded (no bite), or is it because the outside front tire is being grossly overloaded ?

Try to figure out why that tire is coming unglued. It either requires more weight transfer in roll at the front, or less weight transfer in roll at the front, depending what the problem is.

 

[AlbertHung]

I see "suspension links" or "suspension linkages" mentioned often, but I'm having trouble figuring out what components of a suspension are considered links/linkages. I know springs/shocks/swaybars are not, but then what are? I suppose the control arms, but what about their bushings? Balljoints? Tires?

Sorry if this is slightly off topic, but I ask because as I understand it, raising the roll center will increase the amount of load transferred through the springs/shocks/swaybar, which in turn causes more roll. I think I've been able to make the distinction between "roll center" and what causes a vehicle to roll, but it still makes sense to me that raising the roll center will cause the vehicle to roll more at that end.

 

[GregLocock]

Arms are links.

Raising the roll centre at one end will increase the roll resistance of that end of the car and will make it roll less, normally.

Draw a free body diagram to demonstrate this.




Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[Tyler33]

I'm also a member of this group with Rob. Weight distribution is about 40% front 60% rear. The car understeers because there is no bite, especially on gravel. The roll axis is basically parallel to the ground at about 5" above ground. The CG is about 22" high. We have low speed compression damping adjustment on the rear shocks to increase rear roll stiffness and eliminate the need for a sway bar.

I was reading about the center of percussion in Chassis Design. How low can you go with the k^2/ab ratio? It seems our wheelbase is a bit too long (62.75"). I think our k^2/ab ratio is about .75.

Do you guys see any problems with running parallel SLA in the front?

 

[murpia]

"We have low speed compression damping adjustment on the rear shocks to increase rear roll stiffness and eliminate the need for a sway bar."

OK then, what's your roll stiffness distribution front to rear, ignoring this damper effect? And how does this change through the corner, once you factor in the dampers? What's this going to do for your understeer?

I appreciate that baja tyres on dirt might not behave 'classically' in this respect, but at least you should take a look...

Regards, Ian

 

[Tyler33]

I wish I could answer your questions, murpia. Do you have any suggestions on how to measure or calculate those things?

 

[SusTestEng]

"We have low speed compression damping adjustment on the rear shocks to increase rear roll stiffness and eliminate the need for a sway bar. "

That sounds like a bad idea! I am going to take a wild shot at this and say that you are trying to control the rear too much initially. I bet when you drive it, you can't feel the rear moving or rolling too much, But I bet you feel the front trying to roll and stroke the suspension. If you roll ballance is more toward the rear, initially the vehicle won't turn in or respond very well, it wants to stay going straight, then it may snap to an oversteer condition later on in the turn. I would strongly suggest using a rear stabilizer bar and not try to increase your roll stiffness with dampers. It never works, and is not a very good idea for off road setups.

As an OEM development engineer, when I set up the basic roll ballance and spring rates in a car, I don't use ANY damping. I litterally drill a hole in an old damper and drain ALL the oil out so all I am left with is a place holder for the spring and the residual friction of the damper. The vehicle will behave a little strange, but you can tell right away if your roll balance is off, or if your front to rear spring balance is off. You basically use pitching rate, ride frequency, and roll balance as a judge to see what needs to be changed on the car. After this is done, then I start adding damping force to the system. I may come back and make slight changes to the springs and sway bars, but only very minor changes.

 

[BillyShope]

Take a look at the equations at the top of page 683 in your RCVD and you'll see that your friends are on the right track. But, the other parameters present indicate there are other possible solutions (as Greg indicated).

http://home.earthlink.net/~whshope

 

[bentwings1]

I know you are looking for equations and geometric solutions and we went thru a lot of the same head scratching, we finally made our best estimate and began road testing.

Here is how we handled the sand buggy. It has about 58% rear 42% front weight distribution and about 16 inches of suspension travel. It has no anti roll bars. We set the suspension relatively soft all around with soft to medium dampening. Fully adjustable shocks. Front and rear a-arms. The car is about 90 inch wheelbase and 66 inch track. From memory. Liken to a giant R/C car. It has 150 hp and in crusing gear about 9:1 overall gear ratio. It has a live rear axel, no diff. It is capable of over 110 mph as it sits on typical ATV trails. Very fast to say the least.

The driving style that works the best is to enter the corner very fast and quickly let off the throttle so the front end dives or loads then steer and hammer the throttle. The result is that the front end digs in hard then the rear end just drives it thru the corner. You can do multiple S turns very easily with this technic. However if you try to just power thru a turn you will go off the trail as the front end pushes or understeers. We have not tried it on pavement as there are no paved areas on the trails.

We tried at least half a dozen spring rates and the entire range of shock adjustment. For the most part it is the driving that determins how this car handles. We do use 2-3 springs for a progressive rate. The car is grossly out of its element on ATV trails but considering that it weighs about the same it is only larger and more powerful.

As SusTest.. says you really cannot feel suspension changes. We have tried making changes and not telling the driver on purpose and he cannot tell the difference. We have video taped and you can see some difference in roll after multiple passes thru the same corner. It is more sensitive to ride height than spring/dampening. I'd like to try anti roll bars maybe this summer.

99 Dodge CTD dually.

 

[Tmoose]

"It has a live rear axel, no diff."

In the big snow storm a few weeks ago our Toro snow blower
(Vintage 1971) reminded me that a non-diff axle with decent traction REALLY likes to go straight, to the extent the Toro is a one handed machine except on wet ice, which is fortunate since the drive control is either on or off, and is controlled by the shift lever on the dashboard, so trying to inch forward, like to chew thru a heavy drift or pull close to a car, requires driving with one hand on and manipulating the shifter. For serious turning they provide a large knobbed control beside each grip to disengage either wheel independently to allow power turns.

Don't/didn't buggies have split rear emergency brakes to allow similar induced turning.

 

[patprimmer]

VW based dune buggies used the seperate hand brake cables coupled to a pair of levers or a special lever that applied one or the other wheels depending on direction of movement of lever. Later more sophisticated hydraulic set ups were used.

They needed to only be used in moderation to avoid differential gear damage. As well as steering, they could be used to stop a spinning wheel and transfer power to the wheel with traction, just like used in farm equipment tractors.



Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[Tyler33]

Thanks for the info guys. We have sort of arrived at the same thing that bentwings described. We need to transfer weight to the front outside tire on turn in. This will also take some weight off the rear inside tire allowing it to slip easier. It is our hopes that the low speed compression damping adjustability in the rear will allow us to cause the front end to roll more, initially. We want to avoid using a sway bar if possible due to the extra weight, complexity, and means of failure.

Here's a question, will roll center locations have any effect on what we're trying to do? Right now we're trying to decide if it is better to raise the front roll center to gain a better camber curve. Initially we had decided that a higher roll center in the rear and lower in the front would be ideal for our new vehicle (in comparison to the previous one).

 

[GregLocock]

yes raising the front roll centre is the fastest way to transfer weight to the outside front wheel.

Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.