Side View ICs and Pitch Questions

[nickerss]

Can anyone explain the proper way to determine the side view instant centers? I've read Milliken and Carroll Smith, and any attempts they made just confused me. How important is this point anyway? I've searched quite a bit on here and haven't found any meaningful discussion about them. I'm not really planning on running any anti-dive or squat, but I'd like to make sure I'm not creating a pro-dive/squat situation.

I'm using the OptimumK software right now, and I've got things fairly well under control for bump, roll, and steering inputs, but anything related to pitch throws everything out the window. I end up with outrageous things like 40 degrees of negative camber and 80 inch front view roll center movements. I don't think my suspension points can possibly be that bad, so maybe it's a completely separate issue with the software, but I just don't understand the side view stuff much anyway. Can anyone give me some ideas?

[nickerss]

Can anyone explain the proper way to determine the side view instant centers? I've read Milliken and Carroll Smith, and any attempts they made just confused me. How important is this point anyway? I've searched quite a bit on here and haven't found any meaningful discussion about them. I'm not really planning on running any anti-dive or squat, but I'd like to make sure I'm not creating a pro-dive/squat situation.

I'm using the OptimumK software right now, and I've got things fairly well under control for bump, roll, and steering inputs, but anything related to pitch throws everything out the window. I end up with outrageous things like 40 degrees of negative camber and 80 inch front view roll center movements. I don't think my suspension points can possibly be that bad, so maybe it's a completely separate issue with the software, but I just don't understand the side view stuff much anyway. Can anyone give me some ideas?

[terra_dactile]

Hello nickerss,
With OptimumK you should see a pitch axis and roll axis, try and make your geometry so that these axii stay within the range you deem acceptable, if these axis are traveling at Mach speed with a little variation in pitch angle or roll angle then you might want to worry about how you car will react. Having zero antis is not a good idea, since that is exactly what will happen when you brake or accelerate, the pitch axis will travel very far very fast.
Remember that the pitch axis also influences the effect of inetria on your suspension system.
If your pitch axis is in front of you front wheel axis, you will have all four springs in droop when braking, imagine braking and having your car rise instead of diving.

This is just what I have learnt speaking with Claude Rouelle, try out one of his seminars they are worth every penny.

Jude Berthault
ETS FSAE 2003-Current
Team Captain & Vehicle Dynamics Leader
2007 Formula Student Autocross Winners

[nickerss]

Thanks Jude. I understand the effects of antis, I'm just wondering to what extent teams use them. When I was reading Milliken, there were a few arguments against spending a lot of time designing for them (something about their usefulness essentially disappearing as ride height changes, as well as loading the a-arms, which means they need to be bigger). I'm not inherently opposed to running antis, I just got the sense that they might not be as effective as one might think.

My real question is how to really determine the side view instant center. Sure, I can see the pitch axis in OptimumK, but I don't understand where it is coming from. I know that you project a line through the upper ball joint and the upper a-arm chassis point (but going which direction?? There are two choices...) and do the same thing with the lower points. In Milliken, it seems to me that they arbitrarily choose how far away the IC is, and what inclination angle it has from the tire contact patch. I haven't found any nice pictures like I have for the front view stuff, which makes a lot more sense to me.

Ross

[JHarshbarger]

To create the side view instant center on a dual a-arm suspension, draw a line in the side view from the inboard points of the upper a-arm and another line from the inboard points of the lower a-arm on one axis. Where those lines intersect would be the side view instant center. If your inboard a-arms are parallel to each other from the side view, you are not going to have any anti-dive because you side view instant center would be infinitely far away. Anti- calculations is a little more complicated to calculate, but since you have a suspension program you don't have to worry about that. You should still understand that your CG creates a moment about the side view instant center to create pitch. Read through RCVD's side view instand center section one more time to more thoroughly understand it.

Determining how much anti-geometry is right for your car is somewhat of a guess at your point. What you need to consider is how much you car is going to dive under braking, how much your car is going to squat under acceleration, and what your suspension is at in those situations. If your car only allows for 1 inch of bump and at 1.3 g's of deceleration your front end bottoms out, you can correct that with anti-dive. Also, as Jude mentioned, you transients while not be as exagerated and drawn out. However, don't go crazy with anti-geometry. Since your transients will be reduced, too much anti-geometry will make your car unstable and twitchy.

[murpia]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by JHarshbarger:
If your inboard a-arms are parallel to each other from the side view, you are not going to have any anti-dive because you side view instant center would be infinitely far away. </div></BLOCKQUOTE>
My understanding is that if you have parallel inboard wishbone pickups, but they are inclined to the horizontal, then anti effects will be in fact be present. This is because the line of force action from the contact patch / wheel centre (depending on what anti we are evaluating) to the infinite instant centre will also be inclined.

Regards, Ian

[terra_dactile]

Hi Nickers,
I could be wrong but is seems like the way your evaluating the problem is as if the front and the back of the car are seperate, however this is not the case, obviously the real case would be as in the roll axis analysis using force pased pitch axis, however that would complicate thing emensly, In the book given during the omptimum G seminar it show how to find your pitch and roll axii with planes that intersect depending on your suspension gemetry, we used the planes in a Catia model to verify the results of Optimumk and to understand the question your asking, trying to understand a 3d problem in 2d is very difficult.
You will see that RCVD, Suspension analyzer and Tune to win all have there own idea of what anti squat/dive is.

Hope this help in some way,

Jude Berthault
ETS FSAE 2003-Current
Team Captain & Vehicle Dynamics Leader
2007 Formula Student Autocross winners

[JHarshbarger]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by murpia:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by JHarshbarger:
If your inboard a-arms are parallel to each other from the side view, you are not going to have any anti-dive because you side view instant center would be infinitely far away. </div></BLOCKQUOTE>
My understanding is that if you have parallel inboard wishbone pickups, but they are inclined to the horizontal, then anti effects will be in fact be present. This is because the line of force action from the contact patch / wheel centre (depending on what anti we are evaluating) to the infinite instant centre will also be inclined.

Regards, Ian </div></BLOCKQUOTE>

I stand corrected. The amount of anti is due to the line created from the contact patch to the IC, and if the a-arms are both at the same angle, the contact patch/wheel center to IC will be at that same angle and you will have anti effects. Thanks, Ian.

[nickerss]

Thank you to everyone for the helpful insight.

I'm still curious to see how much anti teams are running though. From what I've found, most teams run none at all, some run maybe 10-15%, and one or two run large amounts (up to 100%). Sure, our nose dove under braking last year, but we never bottomed it, so the only benefit I see is keeping the camber a bit nicer and very slight reduction in roll center movement.

I would just say what the heck, 15% can't hurt, but I'm worried about ease of manufacturing. We don't have much in the lines of chassis jigging equipment, so making sure that both sides of the car have the a-arm pickups at the same angle would pose a problem (obviously I don't want the left diving more than the right or something like that).

Ross

[C.Zinke]

I like the discussion. The LR07 has 100% AntiSquat and AntiDive. Until now i haven't seen any practical disatvantage. We have testet alot and i am the opinion that the overall concept works good, so the LR08 will still have 100% AntiX.

One thing: Its important to keep in mind that a AntiX effect can be generated by moving the inner suspension points in lateral direction.

[BenB]

I never have done this before, but I was just thinking a little about it. It seems like you could calculate what you need for AD/AS:

You probally know the spring rate you want which was determined by your desired ride and roll rates. So if you have a target for how much pitch (or dive/squat) you want then you can tune your antis until the springs you are already using give you that desired pitch.

The only problem is its not very easy to adjust this so hopefully your numbers are correct.

[nickerss]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by C.Zinke:
I like the discussion. The LR07 has 100% AntiSquat and AntiDive. Until now i haven't seen any practical disatvantage. We have testet alot and i am the opinion that the overall concept works good, so the LR08 will still have 100% AntiX.

One thing: Its important to keep in mind that a AntiX effect can be generated by moving the inner suspension points in lateral direction. </div></BLOCKQUOTE>

Could you please elaborate on how an anti effect can be generated by moving the inner points laterally? I'm only familiar with the previously explained way of inclining the inner points relative to the horizontal (in the side view).

[mtg]

All of the "anti" business didn't really make sense to me until I started thinking about it exactly like front view IC's. It's the same thing, only looking from the side instead of the front.

Instead of thinking about a percent anti-dive, try considering it as a percent geometric load transfer longitudinally. As you increase the percent geometric load transfer (ie increase the anti-dive, etc), the longitudinal load transfer happens faster (same total magnitude). Your wheel path also changes, which has an effect on mechanical grip.

[C.Zinke]

Its not that complex. In my opinion it is not useful to design the lateral ICs or longitudinal ICs seperate in 2D. Things become clearer when you start in 3D. You create planes with the 3 points of each AArm of one wheel. If you intersect the planes of the upper an lower AArm you will get the instant axis for your wheel. If you intersect this axis with the plane of your wheel you will get the longitudinal IC for this wheel.

If you move your inner suspension points of your upper AArm lateral, you will chance the Plane of it, so the Intersection (the instant axis) of the wheel chances and therefor the longitudianal IC.


@Matt, what is the conclusion of your post?

[mtg]

My conclusion is that it doesn't make sense to use the terms anti-dive, anti-squat, pro-dive, etc.

It makes more sense to think about the percentage of geometric load transfer. Of course, that opens up another topic about the relevancy of kinematic roll/pitch axes/centers. To make a long story short, I don't think kinematic roll/pitch axes matter because the real roll/pitch axes aren't the same.

So my real conclusion is to analyze the car with the real load transfer that actually happens with units that make sense. Sounds crazy, I know. Percent anti-dive sounds like something that was made up before suspension systems were well understood.

[murpia]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by mtg:
My conclusion is that it doesn't make sense to use the terms anti-dive, anti-squat, pro-dive, etc.

It makes more sense to think about the percentage of geometric load transfer. Of course, that opens up another topic about the relevancy of kinematic roll/pitch axes/centers. To make a long story short, I don't think kinematic roll/pitch axes matter because the real roll/pitch axes aren't the same.

So my real conclusion is to analyze the car with the real load transfer that actually happens with units that make sense. Sounds crazy, I know. Percent anti-dive sounds like something that was made up before suspension systems were well understood. </div></BLOCKQUOTE>
A very sensible conclusion, if you ask me. Much like the situation with the relevance of 'kinematic roll centre height' (there are plenty of threads around...)

I wonder how practical it would be to analyse & describe the behaviour of what I would describe as the 'compound' true 3D instant centre of each wheel? (For sure we would need ADAMS or a similar multi-body simulation). Given that rarely does a tyre generate a purely longitudinal or lateral force with respect to the sprung mass, the true total reaction forces from the tyres will sort of 'swing around' the sprung mass as the corner proceeds from braking-&gt;turnin-&gt;apex-&gt;exit. I have a feeling that the smoother this 'swinging' the better for the vehicle.

Can anyone suggest a good visualisation & quantification method, or comment on my opinion / description?

Regards, Ian

[terra_dactile]

Hi Matt,
Do you mean to evaluate with sensors on a car ( strain guages push rod for instance) what the real load tranfer is and then use this as a guide line to compare theoretical and real pitch or roll axii depending on the G value read by data aquisition at the same time or have i misunderstood your comment? What I find hard is that when everyone speaks of the force based roll axis etc.. this means you need to have your car built before you can evaluate what happens with all of the compliances etc.. You need K&C testing data etc.. But how do you set your kinematics prior to the build process inorder to have a reaction that you can fine tune once the car gets built? Isnt controlling the pitch and roll axis with kinematics theoretically better then just saying its not reality so who cares where they are located, i guess you have to start somewhere and iterate untill it works best for your car.

Jude Berthault
ETS FSAE 2003-Current
Team Captain & Vehicle Dynamics Leader
2007 Formula Student Autocross Winners

[Kyle Walther]

jude,

i think matt is just contesting the term "anti-drive" saying that it's not the geometry resisting the diving of the vehicle. it's just i matter of taking a percent of the total weight transfer and putting it through the geometric path not through the sprung path. and that the units should represent that.

but i could be way off

The judges at Cali last year asked many times if i believe that the car actually pitches or rolls about these calculated axises. I said no. i'd like to know what someone else thinks.

[mtg]

I think I'm just going to hire the Oklahoma guys to translate my thoughts into meaningful sentences from now on.

Kyle, yes that's what I meant. I could get anal and say I was talking about anti-dive, not anti-drive http://fsae.com/groupee_common/emoticons/icon_smile.gif

My "long story short" part was about the real axes versus kinematic ones. They are not the same. If you were to, say, change the front spring stiffness, your pitch axis moves. Kinematics don't tell you that.

[js10coastr]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Kyle Walther:

The judges at Cali last year asked many times if i believe that the car actually pitches or rolls about these calculated axises. I said no. i'd like to know what someone else thinks. </div></BLOCKQUOTE>

Yes and no... it's pretty clear that a car with a RC close to the CG has less roll than one with a long RC-CG arm. Does the car roll exactly around the roll axis? No, not exactly. The roll axis is just a mathematical representation of what is happening.

For FSAE, it's like asking if the hydro dynamic damping of the fuel makes a difference in handling. I'm sure it has some effect, but at this level, it's really outside of the scope.

My non-pro opinion is to know you're kinematics well enough and be aware of force based RC.