quick suspension question

[Horace]

Hello,

I know there are already a lot of threads on suspension and yes I have used the find function but I cannot find the answer to this question:

On most fsae cars or race cars in general, the lower A arm is usually horizontal. However, our team has designed a suspension setup that has the lower A arm sloping up as it goes from the inside to the outside (wheels). Is there any negative effects of this? We've used susprog to design this and the data looks pretty good. The only thing that might possibly be a problem is that in a bump or droop the RC moves up and down a bit since our fvsa length is small. However in the roll it doesn't move much.

Thanks again,

Horace

[Horace]

Hello,

I know there are already a lot of threads on suspension and yes I have used the find function but I cannot find the answer to this question:

On most fsae cars or race cars in general, the lower A arm is usually horizontal. However, our team has designed a suspension setup that has the lower A arm sloping up as it goes from the inside to the outside (wheels). Is there any negative effects of this? We've used susprog to design this and the data looks pretty good. The only thing that might possibly be a problem is that in a bump or droop the RC moves up and down a bit since our fvsa length is small. However in the roll it doesn't move much.

Thanks again,

Horace

[B Hise]

Be careful with only paying attention to roll geometry. Its good to have stable RMD through roll, but pay attention to pitch as well. Do the coupled analysis for situations like corner entry and exit and you'll see that a lot of vertical roll center movement can cause the car to change balance (US/OS) significantly through the corner.

If youve already built the frame, you can try running a stiff suspension, using anti-drive/anti squat, tuning on the dampers, or just driving around it.

-Bryan

[Jersey Tom]

What data looks pretty good? How much is "a bit" of RC movement?

[nos]

Another question along similar lines.

Ive jumped in the deep end and am designing the suspension for this years car, with no guidance from the previous guys... (although i've spent all my spare time at work getting up to speed on past posts on this forum - some of which are gold!)

We've run single keel set-up in the past, and i thought the benefit of this was to limit RC movement. However from analysis ive done on susprog, i get less RC movement with a dual keel set-up and no degradation of camber gain curves.

So my question to add to what is a reasonable RC movement, is why i'm getting the above scenario?

Cheers, Dean
UNSW

[B Hise]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">So my question to add to what is a reasonable RC movement </div></BLOCKQUOTE>

You can do it analytically, with a good data system or both. By relating your TLLTD to US/OS you can select an appropriate range. For example between 47% and 49% front might be a good balance for your car. But because your roll centers move vertically in pitch, your TLLTD is between 45% and 52% front for the springs you have on the car. Then you would stiffen the car up in pitch to get the car back in the range that you want. You may find during testing that the right TLLTD% is maybe 1.5% higher, but thats what adjustable sway bars are for. Obviously there is a lot to picking suspension geometry, but this is a good start.

I highly recommend getting some tire data and firing up Matlab and/or spending alot of time this summer with a DAQ and your 07 car.

Hope this helps

-Bryan

[Horace]

Jersey Tom,

The RC movement based on roll looks pretty good. About 100-130mm movement horizontally max on 5 deg of roll. And vertically was about 10mm max on 5 deg of roll.

Also, the camber gain from roll and steering looks pretty good and from pure estimation, it seems like they both add up to zero in ~6deg steering.

[B Hise]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">5 deg of roll </div></BLOCKQUOTE>

thats alot of roll.

How do your RCs move in bump/pitch?

[mtg]

I'm convinced that kinematic roll centers are barely worth considering.

At OptimumG, Ranjit looked into force based roll centers (which is what really happens when you're driving) and most of the time the real roll center is far from the kinematic roll center.

You'll get more benefit out of designing a suspension to keep the tire alignment what you want throughout suspension travel than designing it for a certain kinematic roll center control.

The non-horizontal lower control arm just means you will have more track width change as the suspension moves.

[Horace]

B Hise

I just got to school and now have the data right infront of me. Here's what it is:

Static RC height is 30mm

At +/- 25.4 bump/droop, the RC height is -1.66 to 62.20. Problem is that I don't know how to interpret the bump/droop. I don't know what units it is in since I changed the units of the program from mm to in and the 25.4 stays to same. Is it measured in degrees or angle? I'm using Susprog 3D.

Just incase it is measured in degrees, I'm guessing 2 deg would be the max bump. And at that point the RC height is 27.49 to 32.52mm for +/- 2deg.

mtg,

That's exactly how I approached my first suspension design. I designed it to keep the tires aligned. The geometry looks a little too different compared to other teams and I really want to be conservative on my design since it's a first year car. I didn't pay too much attention to RC movement, but on roll here is how much it changes

Static RC 30mm

left wheel

1 deg, -53.75mm offset (horizontal), 29.80mm height

2 deg, -106.90mm offset, 29.21mm height

For the perfect suspension, is the RC only supposed to move horizontally and not vertically? How do you guys interpret the horizontal RC movement?


Thanks very much,

[Horace]

I also want to mention since our RC change is a little on the high side probably, we have a small front view swing axle length (800mm). We did this to have the camber gain we wanted.

Our track width is 1200mm.

Oh and all the data is for the front suspension

[Horace]

And by a "not very conservative" design I mean this kind of shape

http://render2.snapfish.com/render2/is=Yup6aQQ%7C%3Dup6...0Pe%7C/of=50,590,233 (http://render2.snapfish.com/render2/is=Yup6aQQ%7C%3Dup6RKKt%3AxxrKUp7BHD7KPfrj%3DQofrj 7t%3DzrRfDUX%3AeQaQxg%3Dr%3F87KR6xqpxQQ0lxQoGxe0lx v8uOc5xQQQ0laQGlJPeoqpfVtB%3F*KUp7BHSHqqy7XH6gXPle %7CRup60Pe%7C/of=50,590,233)

You might have to paste that into a browser

It seems like most teams have the bottom A arm more horizontal and the top A arm less drastically inclined

[JHarshbarger]

IMO, it looks fine. The reason that most teams have a lower a-arm that is almost horizontal is to minimize tire scrub in bump. Some people also say that a little scrub will help heat the tires up to temp. Scrub is one output that you will find in Susprog3D under both roll and bump.

The changes in RC seem to be fairly normal. Understand that the RC you're looking at is taken from the ground, so when you're at 25mm of bump (that is wheel travel distance not angle), it is understandable that your RC is going to drop about 25mm also. I your case, your RC moves more than your chassis moves, which is not neccesarily a bad thing.

Everything seems fine so far, just keep messing with the numbers to get the geometry and characteristics you want. For a first year car, I would focus more on making the car easy for the driver to control, easy to build, and reliable. Since you have a good suspension program at you hands, you should be able to do all these things while still having a pretty good suspension design.

Good luck!

[Fyhr]

By the looks of it you have a rather high cambergain, how much caster are you running?

Also, you can find near zero RC movements throughout the range, two of our teammates have written programs to find these, one in matlab, the other in java.

[ed_pratt]

Hi Horace

I'll second JHarshbarger's post, it looks fine to me too. I'm in a similar situation designing the suspension and chassis for our 1st class 3 entry. My bottom links are sloping in the other direction, it doesnt matter really - as long as it works for you'r design as a whole.

Remember that although the suspension is important, it's more important to make a package that works together and is easy to drive.

With regards to your numbers, just a bit of fiddling about with the geometry should help you tie down that vertical RC movement in roll.
The most sensitive joints for this are the lower, inner pickups - it might just be a millimetre or two movemtn to get what you want.

hope that helps a bit

Ed

[Jersey Tom]

Bewaaare of short swing arm lengths. Ours was pretty aggressively short, like &lt; 30". We thought we were hot shit.. really smooth track at MPG, camber sensitive tires, so minimize camber change in roll to an absolute minimum and bank on little bump travel over the perfect smooth asphault.

Went to test. Picked spring and damper rates pretty well, but even though the car was almost perfectly balanced mid-corner with no roll bars or anything, had real bad turn-in understeer. Like bad to the point where it would just plow through a slalom. Cranked the compression damping up quite a bit, problem went away. Was only a patch though.

Claude pointed this out at competition..

You start bringin that fvsa in, and the moment of inertia of the unsprung mass about the IC goes WAY down.. goes with radius square. To the point where that inertia is now considerably less than that of the chassis.

So you go into a turn, tires develop some force which is reacted through the wheel nut to the suspension. Line of action of that force relative to the IC (which is now fairly low to the ground) makes a moment that wants to lift the inside tire relative to the chassis. Since the moment of inertia is so damn low now, it can start to do that and lose a ton of grip before the chassis overcomes its inertia in the turn, you get all the weight transfer, and the suspension is planted. Boosting the compression damping makes it more ddifficult for the inside wheel to develop motion relative to the chassis and "lift," so its an effective patch.

Crazy stuff, this suspension business.

[Drum]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content"> At +/- 25.4 bump/droop, the RC height is -1.66 to 62.20 </div></BLOCKQUOTE>

This means the RC moves more that your CoG vertically! That's impressive, I like the way you think.

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content"> we have a small front view swing axle length (800mm). We did this to have the camber gain we wanted.

Our track width is 1200mm. </div></BLOCKQUOTE>

Does this make sense? Maybe it is just me, because I'm tired and full of Wisconsin cheese at the moment (it is ORANGE), but if you have a FVSAL of 800mm, and your track width is 1200mm, wouldn't this make your RC move a little less than the vertical movement of your CoG? I haven't been on the forums for a while and I'm a bit rusty on RCs, but I would have thought that if your FVSAL is longer than your half track, you get a RC that doesn't move as much as your CoG vertically.

Either way, with a lot of vertical RC movement relative to the ground, you will get a lot of camber gain (which I have absolutely nothing against at all http://fsae.com/groupee_common/emoticons/icon_smile.gif ).... BUT have you asked yourself what camber will the front wheels be at when you are braking heavy into a hairpin. When front of the car is diving right down, and you need the maximum braking traction from the front wheels, what does the contact patch look like? Will your tyre choice (and rim choice) be able to accomodate these situations?

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content"> About 100-130mm movement horizontally max on 5 deg of roll. </div></BLOCKQUOTE>
When looking at RC movement, I would concentrate on 2 degrees of roll to start with on an FSAE car. I wouldn't worry too much even if you have 100-130mm of horizontal RC movement at 2 degrees of roll. My thoughts on this are that whether you move the front inboard fore pickup point down 1.3mm or the rear inboard aft pickup point up 0.6mm is pretty irrelevant. Sure, it may have big effects on horizontal RC movement, maybe it will cut it by 50%, but are you that confident in your manufacturing abilities that your actual car will match Susprog that closely? For all the FSAE teams I've seen or been in, there are bigger fish to fry. Get it in the ball park and control tolerances as best you can. Worry about camber gain first, or spring rates, or damping ratio, or arb contribution, or TYRE CHOICE, or DRIVER TRAINING, or castor trail, or mass (or lack of), or ride height, etc.

Pat Drum
RMIT Racing Fan

[Drum]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content"> So you go into a turn, tires develop some force which is reacted through the wheel nut to the suspension. </div></BLOCKQUOTE>

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content"> had real bad turn-in understeer. Like bad to the point where it would just plow through a slalom. Cranked the compression damping up quite a bit, problem went away. </div></BLOCKQUOTE>

Jersey Tom, I think this could be more of a case of just getting more 'bite' on turn-in rather than the IC theory. Stiffening up the front end will help get your tyres working at the first instant of the corner and will in turn improve initial turn-in.

Pat Drum
RMIT Racing Fan

[js10coastr]

I think I'm with "Patty Cakes Drum" on this one. The damper (or dampener) section in RCVD has some good graphs of loads due to various parts of the suspension vs time. It's also a "rule of thumb" that it takes "half a revolution of the wheel" for it to develop lateral load.

On the other hand... it's more likely due to the hydrodynamic damping of the fuel.

[Horace]

I have changed my suspension design several weeks ago to something with fvsa further than the opposite wheels just to be on the safe side, knowing that there is still a lot about suspension that I do not understand. It now has less camber gain, which I now personally think is not as important as keeping RC's steady.

Drum - can you explain why having the RC moving more than the CoG vertically is a good thing? I did not do that intentionally, but rather as a trial thing. I would think it has something to do with jacking and that the RC change will off set the forces set by the bump/droop. But the way the numbers come up it seems to react a force opposite of what you would want. I think I'm missing something here.

This is probably what will happen to me in hell. Me, Susprog3d, and studying suspension until my brains blow into pieces.

[js10coastr]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Drum:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content"> So you go into a turn, tires develop some force which is reacted through the wheel nut to the suspension. </div></BLOCKQUOTE>

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content"> had real bad turn-in understeer. Like bad to the point where it would just plow through a slalom. Cranked the compression damping up quite a bit, problem went away. </div></BLOCKQUOTE>

Jersey Tom, I think this could be more of a case of just getting more 'bite' on turn-in rather than the IC theory. Stiffening up the front end will help get your tyres working at the first instant of the corner and will in turn improve initial turn-in.

Pat Drum
RMIT Racing Fan </div></BLOCKQUOTE>


So I was actually sleeping and then was awaken by an epiphany moment...
...if you put the two posts by "Patty Cakes Drum" together you might have another answer. If it's corner entry that you're having problem with, the car is probably pitched forward a bit, causing the tires to go into negative camber, and then you end up with a pretty small contact patch because of the short FVSAL. Increasing the compression probably helped to keep the nose up.

[Drum]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Drum - can you explain why having the RC moving more than the CoG vertically is a good thing? I did not do that intentionally, but rather as a trial thing. I would think it has something to do with jacking and that the RC change will off set the forces set by the bump/droop. But the way the numbers come up it seems to react a force opposite of what you would want. I think I'm missing something here. </div></BLOCKQUOTE>

Sorry, didn't mean to misguide you. Whether it is a good thing or a bad thing relies heavily on the choices you make to complement your vertical RC movement. Personally, I like the idea of short FVSAL on FSAE cars with lots of camber gain given the nature of the tracks. Perhaps this isn't desirable on a 100hp car where you may struggle to get the power to the ground, or maybe isn't desirable where you have lower profile/stiffer sidewalls on the front tyres which will not conform very well to the road surface with lots of camber gain in braking. In my opinion, big camber gain is not desirable on winged cars either given that your camber will then also be a function of your speed.

Having a RC moving at a ratio close to 1:1 relative to your chassis means a couple of things. The elastic forces transmitted to your springs will not vary as much, but you will inevitably have more variation in your geometric forces.

Gotta run.

Pat Drum
RMIT Racing Fan

[Jersey Tom]

Hey I'm just sayin what was straight out of Claude's mouth. He was adamant about it, and his explanation seemed reasonable to me. I'm in no position with regard to experience to argue much with him.

[Buckingham]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">So you go into a turn, tires develop some force which is reacted through the wheel nut to the suspension. Line of action of that force relative to the IC (which is now fairly low to the ground) makes a moment that wants to lift the inside tire relative to the chassis. Since the moment of inertia is so damn low now, it can start to do that and lose a ton of grip before the chassis overcomes its inertia in the turn, you get all the weight transfer, and the suspension is planted. Boosting the compression damping makes it more ddifficult for the inside wheel to develop motion relative to the chassis and "lift," so its an effective patch. </div></BLOCKQUOTE>

Not having the IC so close to the ground will reduce that moment arm. Our 2004 car was extremely quick in the slalom and had fvsa length equal to half-track.