Short Swing Arm RC Migration [Archive]

exFSAE

10-18-2008, 12:33 PM

How much is "crazy amounts"? And how much does your VSAL change during roll? What is the benefit of running a short VSAL suspension? (RMIT and SCCA A-Mod cars should have nothin to do with it)

The better question to figure out (and to be honest I don't have the best answer for this myself) is, what does RC migration even mean? What's the implication? I'm tempted to say it shows what the contribution of jacking force is from one tire vs the other.. in which case maybe You'd want it to fly to the inside tire (low jacking as a percentage to the high lateral force the outside tire generates)

If you can't say what exactly having a RC at the centerline of your car does, as opposed to having it toward the inside or outside tire, then who is to say that having a stationary RC (or RA) is better than one that moves?

Or, how relevant or accurate is a kinematic RC (I believe this is what you're analyzing?) with a significantly asymmetric suspension (ie one in which the chassis has rolled a good bit).

Claude will give you a lot of crap about a super short VSAL suspension, btw. His belief is that with the super short length the MOI about the instant center is super small relative to the MOI of the sprung mass. In roll then, the car will want to pick up the inside wheels.

I'm not convinced of it, but he is your head judge http://fsae.com/groupee_common/emoticons/icon_smile.gif

J.R.

10-18-2008, 09:48 PM

Originally posted by exFSAE:
How much is "crazy amounts"? And how much does your VSAL change during roll? What is the benefit of running a short VSAL suspension? (RMIT and SCCA A-Mod cars should have nothin to do with it)

They don't really have anything to do with it. I want it for the camber gain, but the 3" or so that the kinematic RC moves seems excessive. I honestly don't know the reasoning behind why I wouldn't want the RC to move all around, I just know that it is "Bad for drive-ability" if they move relative to each other. I started on suspension only a few months ago, so I still have some holes on my facts.

As far as the accuracy, I've been doing the calc's in SW, and am trying to write a matlab code to do it for me. I think they would be accurate, I'm not really sure why they wouldn't be?

You say that with the short VSAL you'll pick up a wheel? That's actually one of the things I'm going for. I want to pick up the inside rear, though, not the front. Other than loads of caster, I'm not really sure how to work the diagonal weight transfer to do that, hence why I need some place to look.

If I had my RC fly to the inside tyre, that increases the length of the outsides VSAL, and in turn decreases it's camber gain with roll, which seems to be a negative to me. If anything, wouldn't I want it to go to the outside tyre, causing the inside to gain less negative camber and the outside to gain progressively more? Perhaps I just answered my own question, but I'll have to look back at my tyre data.

And to think, I switched to suspension from engine because I thought it would be easier.... http://fsae.com/groupee_common/emoticons/icon_wink.gif

exFSAE

10-19-2008, 07:43 AM

You want short FVSA for camber gain... in pitch presumably? You'll have close to zero in pure roll. Short FVSA kinematic package we came up with a couple years ago had a 0.01 deg/deg roll camber rate (in theory anyway, before heinous amounts of compliance).

How do you know RC movement is bad for drive-ability? What about it is "bad" ? Don't just take Claude or Haney or Smith's words for this one. There's no gospel in motorsport.

You are calculating kinematic RC's in SW and Matlab.. and I assume then you'd use the resultant RC heights/positions for some sort of load calculations. Instant centers are just an approximation.. and as the suspension becomes asymmetric (rolls) it isn't really showing reality. This (http://www.neohio-scca.org/comp_clinic/hand_out_reprints/Vehicle%20Dynamics2007.pdf) is a good read.

Regarding picking up a wheel.. that's Claude's claim, not mine http://fsae.com/groupee_common/emoticons/icon_smile.gif I'm not convinced until HE proves it.

Suspension's tough. A-arm design is the easy part. Bellcrank geometry is a bitch.

Picking up the inside rear.. running a solid rear end then? I always wondered about that. Even if you can manage to lift it mid corner, that would only be briefly at max cornering no? (Assuming your driver is really crankin). I'd have to think the car would be really tight on entry and exit with the wheel still down.

Zac

10-19-2008, 08:22 AM

if you're that concerned about kinematic roll center migration, you can always just fix your virtual swingarm length (think drag race ladder bars turned sideways).

J.R.

10-19-2008, 02:33 PM

Wow, exFSAE, that paper is absolute gold. Although now all of the generalizations that I had in my head are called into question, about three weeks before I need a complete suspension design done, AHHH! But this is definitely going to help give me a better idea of what is actually happening.

So, another question. We can determine the lateral forces based on tyre data, how do we detrine the vertical forces the tyres exert? Is this purely based on lateral load transfer, and this is an input for lateral force? Or do we take the lateral force x sin(IA)?

This is kind of confusing because in the article diagram 3, they show for the left wheel a lateral force of -200, and a vertical force of +49, (vehicle axis system). Or is the -49 coupled with that arrow supposed to be a +49 down? Any inside insights? Thanks.

flavorPacket

10-19-2008, 04:17 PM

RC migration is not exclusively a function of any swing arm length. The relative lengths of the upper and lower arms have much more to do with it.

J.R.

10-19-2008, 04:21 PM

Originally posted by flavorPacket:
RC migration is not exclusively a function of any swing arm length. The relative lengths of the upper and lower arms have much more to do with it.

That's not really true. RC = f(FVSA Length). FVSA Length = f(Relative Link Lengths)

Out of curiosity, if you're a Rally Fan (quote) why FSAE and not Baja? Magst du gerne asphalt fahren auch?

rjwoods77

10-19-2008, 04:33 PM

I don't really think lifting the inside rear tire should be a goal. You only need to unload it some percentage versus the outside for what you are trying to do which is figured out through all the graphs, calcs, etc. If you want to lift the inside tire then throw the whole rear suspension out and run a pure swing arm. This is effectively what cal poly slo uses. Their shuttlebox doesn't translate on the pivot in roll which makes them lift the inside rear. They could put a spring/damper on that shuttlebox but have chosen not to do so. Point being is that they don't always have it up in the air the entire time so the inside rear does contribute to handling for some region on the handling envelope which means they have worked out how much suspension geometry induced understeer they need up to a point and it might not really effect them that bad. Wether you run a solid axle of a spooled IRS you will have this effect. We knew going into the solid axle that this would be an issue and that it would be worked out to a usable setup through spring rates, roll bars and dampening. This idea of "lifting the inside tire" as a built in suspension trait really needs to be questioned because that isn't the answer to your handling goals. The question should be how much inside rear tire grip can I get away with before it hinders my ability to get the car handling the way I want. This gets into what Pat Clark was getting into in his article. "Lifting the inside tire" is avoiding this question and quite frankly a cop out to understanding what you need to understand. Also if you think about it the on a solid rear axle such as ours the tires are linking and have zero camber and toe. When you lift the inside rear that means the wheel/tire centerline is no longer parallel with the ground which has an equal put opposite effect on the outside tire. Last time I checked these tires we use aren't most efficient with positive camber on the outside tire.

J.R.

10-19-2008, 04:51 PM

Originally posted by rjwoods77:
I don't really think lifting the inside rear tire should be a goal. You only need to unload it some percentage versus the outside for what you are trying to do which is figured out through all the graphs, calcs, etc. If you want to lift the inside tire then throw the whole rear suspension out and run a pure swing arm. This is effectively what cal poly slo uses. Their shuttlebox doesn't translate on the pivot in roll which makes them lift the inside rear. They could put a spring/damper on that shuttlebox but have chosen not to do so. Point being is that they don't always have it up in the air the entire time so the inside rear does contribute to handling for some region on the handling envelope which means they have worked out how much suspension geometry induced understeer they need up to a point and it might not really effect them that bad. Wether you run a solid axle of a spooled IRS you will have this effect. We knew going into the solid axle that this would be an issue and that it would be worked out to a usable setup through spring rates, roll bars and dampening. This idea of "lifting the inside tire" as a built in suspension trait really needs to be questioned because that isn't the answer to your handling goals. The question should be how much inside rear tire grip can I get away with before it hinders my ability to get the car handling the way I want. This gets into what Pat Clark was getting into in his article. "Lifting the inside tire" is avoiding this question and quite frankly a cop out to understanding what you need to understand. Also if you think about it the on a solid rear axle such as ours the tires are linking and have zero camber and toe. When you lift the inside rear that means the wheel/tire centerline is no longer parallel with the ground which has an equal put opposite effect on the outside tire. Last time I checked these tires we use aren't most efficient with positive camber on the outside tire.

Good points. The tough part about figuring out just how much to unload the inside tyre is that there is no longitudinal data for these tyres. SOOOO, I have no idea what forces are pushing on the axle in the wrong direction. One of the things that I need to do is compare our tyres to a similar 13 and "guess" what the longitudinal forces are that I am dealing with and then find out the losses that I will have to worry about. I didn't really consider the positive camber issue on the rear before, thanks for pointing that out. With our weight distribution this would have been a significant oversight. Any chance your willing to come out for a day and finally give your suspension talk so we can argue about this stuff in person?

exFSAE

10-19-2008, 05:01 PM

Originally posted by J.R.:
Wow, exFSAE, that paper is absolute gold. Although now all of the generalizations that I had in my head are called into question, about three weeks before I need a complete suspension design done, AHHH! But this is definitely going to help give me a better idea of what is actually happening.

So, another question. We can determine the lateral forces based on tyre data, how do we detrine the vertical forces the tyres exert? Is this purely based on lateral load transfer, and this is an input for lateral force? Or do we take the lateral force x sin(IA)?

This is kind of confusing because in the article diagram 3, they show for the left wheel a lateral force of -200, and a vertical force of +49, (vehicle axis system). Or is the -49 coupled with that arrow supposed to be a +49 down? Any inside insights? Thanks.

The tires themselves don't generate vertical force. The jacking effect on the sprung mass is a fallout of your suspension bar arrangement (I believe.. though stuff like overturning moment of the tire would likely have an effect).

Bottom line, and I believe Mitchell even commented, RC's aren't the thing you should be super worried about.

What will help you (or hurt you..) much more with these cars than camber curves and RC's.. are things like chassis rigidity, suspension stiffness, and things like roll and bump-steer. It is almost difficult to design a-arm geometry that will junk a FSAE car handling.. not too hard to blow everything else!

Compliant chassis is going to make it a real pain to change the balance of the car without massive spring and bar changes. Be chasing your tail while tuning. Compliant suspension is going to cost heaps of grip and make the car lazy. Excessive roll and bump-steer in directions you don't want will make it wildly more unpredictable than an imaginary roll center moving around.

Plus, as I said, getting your bellcrank geometry is going to be way more pain in the ass than a-arms. Have fun http://fsae.com/groupee_common/emoticons/icon_smile.gif

http://i292.photobucket.com/albums/mm19/exfsae/short_fvsa_front.jpg
Been there done that...

rjwoods77

10-19-2008, 05:51 PM

JR,

Speak to Edward about that one. I assume that a scaling thing might give you something to look at. I would at least have all the work done and the understanding so when questioned you can say "if the curve says this then....." if you don't have the data to show whats up. The positive camber thing is all about how much you actually lift the tire. It is possible to not lift the tire and have almost no load on it. Just push on the main roll hoop high point and watch the car slide on the concrete. As far as a talk as you know my knowledge is more from a vehicle integration view. I don't pretend to know the ins and outs of suspension. Mike has done more actual engineering calculation work on the car than I have. I just know how to make the required geometry manifest itself in a very well integrated design. I just know where and what to look and not look at. I am very knowledgeable about rocker arm geometry however. If you guys want me to take a look at things and give you an opinion I would be more than happy too. I figured nobody calling me was sign so.....

Force based roll centers have been discussed here a whole bunch so look around.

I commented about the "lifting the inside tire" because I it seems the new crew(starting last year) has an almost obsession on a couple ideas that don't make much sense and/or are a simplistic view on things out of lack of wanting to look into them that I have commented about before out of frustration from not being listened to. The other big one was the false idea that the CVT should have a high engagement speed. Our vehicle is unique in the way it works and examples from karts and snowmobiles don't really apply to what you are doing. Kart chassis rely on axle bending and chassis stiffness to make the car do what they want it to do. Olav Aaens book, while a great resource, is almost purely devoted to drag sleds and road course SCCA cars that operate in ways our car does not. I have looked at things simplistically as well but always with an eye on what needs to be understood and not with what needs to ruled out.

murpia

10-20-2008, 03:28 AM

Originally posted by exFSAE:
The tires themselves don't generate vertical force. The jacking effect on the sprung mass is a fallout of your suspension bar arrangement (I believe.. though stuff like overturning moment of the tire would likely have an effect).
This point is really key when analysing / understanding jacking forces! You aren't increasing vertical load on any tyres in steady state, due to jacking. Yes, there might be a transient effect you can exploit but that's it.

It's the same as dampers, yes you can increase the vertical load on a tyre temporarily with a momentum transfer through a damper, but over time you must maintain the steady state vertical loads so you must at some stage reduce the vertical load in compensation.

So, do FSAE cars actually reach steady state, or they just transitioning from one transient to another? And you answer to that might strongly influence how you interpret a 'roll centre' or any other geometric or kinematic concept...

Regards, Ian

flavorPacket

10-20-2008, 12:58 PM

Originally posted by J.R.:That's not really true. RC = f(FVSA Length). FVSA Length = f(Relative Link Lengths)

This is false. Check your books. FVSA is defined by your IC y and z positions. IC position is controlled by the slope of the arms, not their lengths (unless one arm is so long that it goes through the projected line of the other arm).

As for why I don't do Baja, it's the same reason most of us don't: the series has ridiculously low standards for engineering and is not nearly as competitive as FSAE.

Mike Cook

10-20-2008, 05:54 PM

A few things that come to mind:

RC height determines jacking forces magnitude. This jacking forces will tend to lift that end of the car in a steady state corner. However, because our RC's are usually pretty low, jacking forces are small. For instance,

RC ht = 1"
outside load = 225
inside load = 75
track = 50
COF = 1.5

Vertical force from outside tire (upward) =
(1/25)*225*1.5 = 13.5lb

Vertical force from inside tire (downward) =
(1/25)*75*1.5 = 4.5lb

Net jacking Force = 9 lb

Suspension rate (heave) = 200lb/in --> front end lifts 9/200 ~ .05" I will let you decide whether or not that is significant. I guess if the rear moves in the opposite direction, than the turn in performance might be different than SS, but there going to be different (and rightfully should be) anyways....

Now, what happens if the RC moves laterally?
well, the jacking force will move laterally which will also cause a roll moment. Lets say the RC moves 4". Roll moment due to jacking forces would be 4*9= 36in-lb. What is our roll moment due to cornering force? Well it will be
(300*1.5) = 450lb*9in = 4050in-lb, so the roll moment due to RC migration is roughly 1% of the total roll moment. I will let you decide whether or not that is significant.

So what matters?

Well, in terms of absolute performance (regardless of setup):
Toe's and cambers that the tires like.

Now to make the whole car work, we need to not only make absolute grip, but make the car balance right. To do this, your LLTD needs to be right. So i'm a big fan of getting your Front RC to rear RC height ratio correct. If they are not, than you can compensate for this with springs and bars which will work too, but for instance, lets say I run to high of a front RC, then I will need to drop front spring rate a lot which will hurt pitch. Also, RC's will have a direct effect on entry/exit performance while springs won't, but I don't have a lot of first hand experience testing that - usually I can use the dampers to get us in the ball park here.

A few other things for thought....

Over the summer we drove a fellow competitors car. Their car is very basic and not very good at all-they usually finish towards the end of the pack at competition. With us driving it, it was 3 seconds off of our times. Now their engine wasn't very good, so you figure maybe a second if the engine was improved, which puts them 2 seconds off our times. Starting at first place at michigan for instance, if you were two seconds off the pace, how many places would you loose? 5? 10? Lets not forget the big picture. Spending too much time on geometry will hurt you if you don't have time to tune it.....

Jersey Tom

10-20-2008, 08:08 PM

Good post, Mike.

J.R.

10-20-2008, 09:12 PM

That's really helpful, I just want to clarify some things.

Originally posted by Mike Cook:
A few things that come to mind:

RC height determines jacking forces magnitude. This jacking forces will tend to lift that end of the car in a steady state corner. However, because our RC's are usually pretty low, jacking forces are small. For instance,

RC ht = 1"
outside load = 225
inside load = 75
track = 50
COF = 1.5

Vertical force from outside tire (upward) =
(1/25)*225*1.5 = 13.5lb

Vertical force from inside tire (downward) =
(1/25)*75*1.5 = 4.5lb

Net jacking Force = 9 lb

RCh/(tf/2) * lateral force at tyre (out, in) = jacking (o, i)

Now, what happens if the RC moves laterally?
well, the jacking force will move laterally which will also cause a roll moment. Lets say the RC moves 4". Roll moment due to jacking forces would be 4*9= 36in-lb. What is our roll moment due to cornering force? Well it will be
(300*1.5) = 450lb*9in = 4050in-lb, so the roll moment due to RC migration is roughly 1% of the total roll moment. I will let you decide whether or not that is significant.

(300*1.5) = 450lb*9in = 4050in-lb
Lateral Force both* moment arm = total moment

Is that 9 in supposed to be a 4 in? I can't find where the 9 would have come from.

So what matters?

Well, in terms of absolute performance (regardless of setup):
Toe's and cambers that the tires like.

Now to make the whole car work, we need to not only make absolute grip, but make the car balance right. To do this, your LLTD needs to be right. So i'm a big fan of getting your Front RC to rear RC height ratio correct. If they are not, than you can compensate for this with springs and bars which will work too, but for instance, lets say I run to high of a front RC, then I will need to drop front spring rate a lot which will hurt pitch. Also, RC's will have a direct effect on entry/exit performance while springs won't, but I don't have a lot of first hand experience testing that - usually I can use the dampers to get us in the ball park here.

Most LLT occurs from CGh and t(f,r) as far as I've seen. Does this mean just fine tuning it with geometric effects?

Another dumb question but something I can't find the answer to: Two types of load transfer are mentioned, elastic and inelastic OR geometric. Is this the difference, I'm assuming elastic = track based, inelastic = suspension geometry based?

A few other things for thought....

Over the summer we drove a fellow competitors car. Their car is very basic and not very good at all-they usually finish towards the end of the pack at competition. With us driving it, it was 3 seconds off of our times. Now their engine wasn't very good, so you figure maybe a second if the engine was improved, which puts them 2 seconds off our times. Starting at first place at michigan for instance, if you were two seconds off the pace, how many places would you loose? 5? 10? Lets not forget the big picture. Spending too much time on geometry will hurt you if you don't have time to tune it.....

Definately understood with the tuning, our car should have had more of a focus on that last year. Gremlins in the wiring harness kept us from tuning until... brake test at VIR? http://fsae.com/groupee_common/emoticons/icon_wink.gif We want a completed design for our entire car by Thanksgiving, constructed by winter break and running by march (lots of snow in Buffalo keeps tuning till then anyways http://fsae.com/groupee_common/emoticons/icon_wink.gif )

I know I'm asking for a lot. I really like to fully understand a problem before I call my design 'final', and these are things that I haven't been able to wrap my head around. I really appreciate all the help, feel free to tell me if I'm asking for too much!

exFSAE

10-20-2008, 09:25 PM

Correct on jacking force. The line from the contact patch to the instant center shows how much jacking compared to lateral. If the slope is 1 in 4, for each 4 lb of Fy, you get 1 lb Fz on sprung mass. In a compliance test you can apply pure lateral force to a stationary tire and see the change in Fz..

I imagine 9 is CG height.

Wrt elastic and geometric.. you know your total LT is a function of CGH. Can't change that. Percentage that is rolling moment is proportional to distance from roll axis to CG. Non-rolling overturning moment (which I believe is what you'd call geometric) is proportional to the remainder of that distance, roll axis to ground. Sum is CGH. Roll axis inclination is a balance tuning tool for that part of the WT, whereas spring and bar split are for elastic. Keep in mind what happens when the car pitches..

This is good though. Great questions and good discussion. The massively important thing is to try to make sense of why everything does what it does. With SO many things that change the feel of the car, you need to know how to separate it all out.. for those odd situations when disconnecting an ARB has no effect on balance change.. when bolting on a different set of tires changes the car balance to absurd understeer.. when stiffening the rear ARB seems to make the car oversteer less, etc.

Mike Cook

10-21-2008, 05:47 AM

Originally posted by J.R.:

Is that 9 in supposed to be a 4 in? I can't find where the 9 would have come from.

Most LLT occurs from CGh and t(f,r) as far as I've seen. Does this mean just fine tuning it with geometric effects?

Another dumb question but something I can't find the answer to: Two types of load transfer are mentioned, elastic and inelastic OR geometric. Is this the difference, I'm assuming elastic = track based, inelastic = suspension geometry based?

The 9" was assuming the CG_ht was at 10" and thus the value h=9=(cg_ht-RC_ht). This would be your moment arm from the location the lateral force is applied at the chassis to the CG in front view.

If you go back to my math for jacking forces, you see that the outside tire produced 13.5lb up and the inside produced 4.5 down at the chassis. Well these forces have equal and opposite forces acting at the tires. So while the net force on the chassis was only 9lb when cornering, the outside tire will get 13.5lb heavier and the inside tire will get 4.5lb lighter. So what are the new loads on the tires?

outside = 225+13.5 = 238.5
inside = 75-4.5 = 70.5

but now the total weight increased which can't be true -> 70.5+238.5 =/ 75+225.

But remember, we had a 9 lb jacking force on the chassis. The spring rate was 200lb/in (heave), and the chassis raised .05". When the chassis raises, its springs unextend, which means their load on the tires decreases. And we know it has to decrease by 9lb (statics). So 4.5lb is removed from each tire. So our true wheel loads are:

outside = 238.5-4.5 = 234lb
inside = 70.5-4.5 = 66lb

Thus the inside lost a total of 9lb, and the outside gained 9lb. So our weight transfer through the suspension is 9lb. This is called your inelastic weight transfer.

What happens if we move the RC upward? Well the jacking force will be higher, and as we saw, <STRIKE>the jacking forces is equal to the in-elastic weight transfer. </STRIKE> EDIT: This isn't always the case, I think its only the case when the COF is symmetric, if you run through some cases where when the inside tire becomes unloaded, yet still produces the same cornering force you will see what I'm talking about........END EDIT

Of course what else happens if we move the RC up? Your roll moment will decrease, which means the car will roll less, which means that there will be less weight transfer through your springs and ARB (this is called your elastic weight transfer).

But it is important to note that, lets say I move my RC from 1" mto 2", than my inelastic WT doubles from 9lb to 18lb (a net 9lb increase). As I just mentioned, increasing our RC ht will decrease our elastic WT. But it will not decrease by 9lb. It will decrease by the ratio of front roll stiffness to total roll stiffness times the inelastic WT. So if we have 50% front roll stiffness and 50% rear roll stiffness, the elastic WT will be (50/(50+50))*9 = .5*9=4.5lb. If we move the RC from 1" to 2" the net change is:

Outside = 234(previous weight)+9(inelastic)-4.5(elastic) = 238.5lb
Inside = 66(previous weight) - 9(inelastic) +4.5(elastic) = 61.5lb

So when we move up the RC, inelastic WT increases, elastic WT decreases (but not by as much) and the new WT increases.

If the car is too loose, we need to tighten it up, which means we need to transfer more weight at the front. To do this, we could rasise the front RC by changing the ride height, or increase spring rate, or increase swaybar rate.

scott_rfr

10-22-2008, 12:59 PM

Mike, not sure about your "in-elastic wt" comment. Your saying that the inelastic wt is the jacking effect and all other is elastic? During steady-state this might be right however during the corner entry/exit transient phase your inelastic wt which is transferred instantly (well relatively) via the linkages then your elastic part transferred via the springs.

Mike Cook

10-22-2008, 06:51 PM

Scott, thinks for getting me to think about this a little more. I edited my above post a little bit.

When you turn in, the cornering forces are reacted through the suspension links and instantly transfer weight and produce jacking forces. In addition, these lateral forces will make the CG roll out of the corner, with only the springs to resist its rolling. As the chassis rolls, and the outside spring compresses and the inside spring extends, more weight is transferred. However, how fast this happens depends on the amount of roll inertia the vehicle has.

So:
in-elastic WT = instant
elastic WT = delayed and dependent on roll inertia, spring rates, max roll angle, etc.

In my above post, I got thinking about jacking forces a little more. Jacking forces are really caused by having different corner forces produced by the outside and inside tires. If the cornering forces produced by the inside and outside tires are equal, there will be no jacking forces but there will still be in-elastic WT. This only happens if the COF for the outside and inside tires are not equal - (think nascar). Anyways, I need to think about this a little bit more. Interesting topic....

scott_rfr

10-22-2008, 07:48 PM

Mike, No problem.

The way I look at jacking is I use the KRC. I draw a force line from tire contact patch to KRC. I draw Fy along this force line. The vertical component of with is your jacking force. The sum of the jacking force is then applied to the KRC. This jacking force will "lift" the car. Making this a little more complicated the jacking force will cause displacements in the CG thus transferring more weight. So this answers the questions about having the KRC moving laterally, as the KRC moves laterally the slope of the force line is then changing.

The COF for inside and outside tires is going to vary even in an fsae car during cornering as you transfer load the COF on outside goes down, COF inside goes up this is all due to tire load sensitivity. Interesting if you have tire data to write a matlab/simulink script to see what going on.

ben

10-23-2008, 06:19 AM

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 exFSAE:
The tires themselves don't generate vertical force. The jacking effect on the sprung mass is a fallout of your suspension bar arrangement (I believe.. though stuff like overturning moment of the tire would likely have an effect).
This point is really key when analysing / understanding jacking forces! You aren't increasing vertical load on any tyres in steady state, due to jacking. Yes, there might be a transient effect you can exploit but that's it.

It's the same as dampers, yes you can increase the vertical load on a tyre temporarily with a momentum transfer through a damper, but over time you must maintain the steady state vertical loads so you must at some stage reduce the vertical load in compensation.

So, do FSAE cars actually reach steady state, or they just transitioning from one transient to another? And you answer to that might strongly influence how you interpret a 'roll centre' or any other geometric or kinematic concept...

Regards, Ian </div></BLOCKQUOTE>

I'm not sure I agree. surely if I have 75% of lateral force from the outside tyre and 25% from the inside tyre (for example) then I have a net vertical force acting on the sprung mass that will unload the springs to maintain vertical equlibrium as Mike's pointed out.

Surely this force is present in steady state mid-corner?

Ben

BilletB

10-23-2008, 10:33 AM

BilletB

10-23-2008, 10:42 AM

In my above post, I got thinking about jacking forces a little more. Jacking forces are really caused by having different corner forces produced by the outside and inside tires. If the cornering forces produced by the inside and outside tires are equal, there will be no jacking forces but there will still be in-elastic WT. This only happens if the COF for the outside and inside tires are not equal - (think nascar). Anyways, I need to think about this a little bit more. Interesting topic....

I think you may want to think about it a little more. Jacking forces are not created by having non-equal forces at opposite tires. Not at all. Jacking forces are just a result of lateral forces acting at the tire contact patch. How the normal forces on your tires effects the lateral force will then effect the jacking force but the actual creation of jacking forces doesn't have anything to do with un-equal lateral forces between tires. Don't forget that as your suspension rolls each side of the suspension is no longer a mirror of the opposite side so n-line slopes change and even equal normal tire loads won't create offsetting jacking forces.

J.R.

10-23-2008, 11:14 AM

Originally posted by BennyHL:
Ben,
I think Ian was getting to the point that these jacking forces can increase net load on the tires in transients (net load being a total normal force on all 4 tires greater than the vehicle's weight), but during steady state, while these jacking forces are still present the net sum of the normal force on your tires will always sum to the vehicle's weight.

Can anyone confirm this? I guess it makes sense since tyres can produce more lateral force than they have normal load, and the normal load is not changing, but it is still rather confusing. By this same logic, with an alternate suspension design you would actually be able to reduce the loads that were on the tyres (no reason to do so, just a concept clarifier) Also, if the forces are acting through the suspension linkages, why would steady state vs. transient matter. You could arrange the linkage to provide more normal load in steady state based on the same principal if my logic is correct.

scott_rfr

10-23-2008, 11:16 AM

BennyHL

How can the total vertical force on the tires be greater than the weight of the car plus any aero? If an extra load during the transient phase is pushing down on the tires it has to come from somewhere you can just add weight from nowhere???

Scott

BilletB

10-23-2008, 11:22 AM

Originally posted by scott_rfr:
BennyHL

How can the total vertical force on the tires be greater than the weight of the car plus any aero? If an extra load during the transient phase is pushing down on the tires it has to come from somewhere you can just add weight from nowhere???

Scott

I'll give you a hint: vertical accelerations. Once you get it, do NOT get too engulfed by the wonderfulness of the concept. When it all comes to light you'll find the increased normal force turns out to be a temporary fluctuation and often times that fluctuation when exaggerated will lead to more net harm than net good. So, this is all a transient analysis and transients are funny and difficult things to accurately model and they are often painfully fleeting.

BilletB

10-23-2008, 11:31 AM

Originally posted by J.R.:
You could arrange the linkage to provide more normal load in steady state based on the same principal if my logic is correct.

This is what you really cannot do. Steady state tire loads are a function of your track width and cg height (given static weights and some lateral acceleration) and has nothing to do with suspension geometry beyond any wild effects it has on possibly moving your cg. Look at it again. You won't change normal tire loads with suspension geometry! You'll just change how that load is shared between the tire and chassis.

murpia

10-23-2008, 01:57 PM

Originally posted by ben:
<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 exFSAE:
The tires themselves don't generate vertical force. The jacking effect on the sprung mass is a fallout of your suspension bar arrangement (I believe.. though stuff like overturning moment of the tire would likely have an effect).
This point is really key when analysing / understanding jacking forces! You aren't increasing vertical load on any tyres in steady state, due to jacking. Yes, there might be a transient effect you can exploit but that's it.

It's the same as dampers, yes you can increase the vertical load on a tyre temporarily with a momentum transfer through a damper, but over time you must maintain the steady state vertical loads so you must at some stage reduce the vertical load in compensation.

So, do FSAE cars actually reach steady state, or they just transitioning from one transient to another? And you answer to that might strongly influence how you interpret a 'roll centre' or any other geometric or kinematic concept...

Regards, Ian </div></BLOCKQUOTE>

I'm not sure I agree. surely if I have 75% of lateral force from the outside tyre and 25% from the inside tyre (for example) then I have a net vertical force acting on the sprung mass that will unload the springs to maintain vertical equlibrium as Mike's pointed out.

Surely this force is present in steady state mid-corner?

Ben </div></BLOCKQUOTE>
If we're happy to define steady-state mid-corner as a situation of constant lateral acceleration and zero suspension movement (i.e. all transients have decayed and all velocities are zero except the gross vehicle velocity) then we must be able to draw a free body diagram with no D'Alembert forces in it except the aforementioned constant lateral acceleration.

With no vertical component possible in that lateral force, the only possible equilibrium is for the 4 contact patch vertical forces to exactly equal the weight of the car (neglecting aero forces, as we have been doing).

But, there's no reason that the steady-state mid-corner vehicle should have it's centre of gravity in the same place as the straight-line vehicle. It could be higher, due to the jacking effect of the suspension.

Energy (a force acting along a displacement) can be expended during the corner-entry transient to raise the centre of gravity of the vehicle, and is then recoverable as the centre of gravity lowers again on corner-exit transient.

Or looked at another way (as BennyHL states) the jacking forces accelerate the sprung mass vertically during the corner-entry transient to steady-state mid-corner. So, you would expect the sprung mass to accelerate downwards again as the car exits steady-state mid-corner and the vehicle will lose some total vertical contact patch load during the exit transient.

This again indicates to me that jacking forces should be treated similarly to low-speed damper adjustments, when it comes to chassis tuning - what you gain on the way in, you lose on the way out (or vice versa). That's not a criticism, just a realistic assessment of the situation. The good engineer uses all the tools in the box, and uses them appropriately and with understanding.

Regards, Ian

Mike Cook

10-23-2008, 08:46 PM

I think you may want to think about it a little more. Jacking forces are not created by having non-equal forces at opposite tires. Not at all. Jacking forces are just a result of lateral forces acting at the tire contact patch. How the normal forces on your tires effects the lateral force will then effect the jacking force but the actual creation of jacking forces doesn't have anything to do with un-equal lateral forces between tires. Don't forget that as your suspension rolls each side of the suspension is no longer a mirror of the opposite side so n-line slopes change and even equal normal tire loads won't create offsetting jacking forces.

Correct, and I think this was what I was basically getting at...

Carry on...

CappyUMD

10-23-2008, 08:46 PM

Keep in mind that cg vertical acceleration is out of phase with cg height. The cg acceleration becomes negative when the vertical velocity decreases. That is, before the cg has finished rising. Jacking may improve inital turn-in response, but you will see a reduction in normal load even before mid-corner. In my opinion, dividing a corner into more than 3 segments (entry,mid,exit) of car behavior doesn't improve drivability.

Typically, the jacking forces are probably too small to confuse a driver or be used as a tuning tool.

BilletB

10-24-2008, 12:42 AM

Originally posted by Mike Cook:

Thus the inside lost a total of 9lb, and the outside gained 9lb. So our weight transfer through the suspension is 9lb. This is called your inelastic weight transfer.

Mike,
This may account for the load transfer not through your suspension springs due to jacking but remember there is also some other load transfer that occurs that would be classified as 'inelastic'.

ben

10-24-2008, 01:04 AM

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 ben:
<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 exFSAE:
The tires themselves don't generate vertical force. The jacking effect on the sprung mass is a fallout of your suspension bar arrangement (I believe.. though stuff like overturning moment of the tire would likely have an effect).
This point is really key when analysing / understanding jacking forces! You aren't increasing vertical load on any tyres in steady state, due to jacking. Yes, there might be a transient effect you can exploit but that's it.

It's the same as dampers, yes you can increase the vertical load on a tyre temporarily with a momentum transfer through a damper, but over time you must maintain the steady state vertical loads so you must at some stage reduce the vertical load in compensation.

So, do FSAE cars actually reach steady state, or they just transitioning from one transient to another? And you answer to that might strongly influence how you interpret a 'roll centre' or any other geometric or kinematic concept...

Regards, Ian </div></BLOCKQUOTE>

I'm not sure I agree. surely if I have 75% of lateral force from the outside tyre and 25% from the inside tyre (for example) then I have a net vertical force acting on the sprung mass that will unload the springs to maintain vertical equlibrium as Mike's pointed out.

Surely this force is present in steady state mid-corner?

Ben </div></BLOCKQUOTE>
If we're happy to define steady-state mid-corner as a situation of constant lateral acceleration and zero suspension movement (i.e. all transients have decayed and all velocities are zero except the gross vehicle velocity) then we must be able to draw a free body diagram with no D'Alembert forces in it except the aforementioned constant lateral acceleration.

With no vertical component possible in that lateral force, the only possible equilibrium is for the 4 contact patch vertical forces to exactly equal the weight of the car (neglecting aero forces, as we have been doing).

But, there's no reason that the steady-state mid-corner vehicle should have it's centre of gravity in the same place as the straight-line vehicle. It could be higher, due to the jacking effect of the suspension.

Energy (a force acting along a displacement) can be expended during the corner-entry transient to raise the centre of gravity of the vehicle, and is then recoverable as the centre of gravity lowers again on corner-exit transient.

Or looked at another way (as BennyHL states) the jacking forces accelerate the sprung mass vertically during the corner-entry transient to steady-state mid-corner. So, you would expect the sprung mass to accelerate downwards again as the car exits steady-state mid-corner and the vehicle will lose some total vertical contact patch load during the exit transient.

This again indicates to me that jacking forces should be treated similarly to low-speed damper adjustments, when it comes to chassis tuning - what you gain on the way in, you lose on the way out (or vice versa). That's not a criticism, just a realistic assessment of the situation. The good engineer uses all the tools in the box, and uses them appropriately and with understanding.

Regards, Ian </div></BLOCKQUOTE>

Ok so the vertical acceleration is zero, but the magnitude of the jacking force is still having an effect on the sprung mass position relative the ground - quite important if you've got an aero car I suspect?

Ben

murpia

10-24-2008, 01:27 AM

Originally posted by ben:
Ok so the vertical acceleration is zero, but the magnitude of the jacking force is still having an effect on the sprung mass position relative the ground - quite important if you've got an aero car I suspect?

Ben
Not as important as you might expect, since aero cars also have quite stiff wheel rates. Vertical sprung mass movement will be roughly proportional to wheel rate, i.e. a stiffer car jacks less for equal jacking force.

Regards, Ian

PS Swing arm + soft springs http://fsae.com/groupee_common/emoticons/icon_smile.gif (http://uk.youtube.com/watch?v=VjqL-2yx_4Y)

J.R.

10-24-2008, 05:27 AM

Originally posted by BennyHL:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by J.R.:
You could arrange the linkage to provide more normal load in steady state based on the same principal if my logic is correct.

This is what you really cannot do. Steady state tire loads are a function of your track width and cg height (given static weights and some lateral acceleration) and has nothing to do with suspension geometry beyond any wild effects it has on possibly moving your cg. Look at it again. You won't change normal tire loads with suspension geometry! You'll just change how that load is shared between the tire and chassis. </div></BLOCKQUOTE>

So from this I gather you don't believe in jacking loading at all. Or does the LT calcs that mike cook did only apply in transient cornering? The jacking effects are definately not the "Just dependent on lateral accel, cg and track" which is what I thought as well. According to the jacking calcs that were done, it is actually possible to transfer more load to the outside tire with jacking forces. This would add to the traditional LLT. Or do these jacking forces have to balance front to rear with the total LLT, similar to adding an ARB?

Mike Cook

10-24-2008, 05:48 AM

JR, read through this post again.

The calculations I wrote up basically apply to steady state conditions. What I got at later on in my posts was that I made some assumptions which probably don't change the results very much. I assumed that the car was symmetric, COF was even from side to side, etc.

Jacking forces can be present any time the car produces lateral force.

If a jacking force is applied to sprung mass, the mass will accelerate up. The load on all 4 tires = M*A. Since A goes up, the load on all 4 tires will actually be more than weight of the car. This is a transient effect. As the sprung mass accelerates up, the springs are extending which takes load off of the sprung mass and the system goes into equilibrium at some new ride height (this is basically when the car is steady state).

I tired to show in my earlier posts that the effects of jacking and roll center movement is just not that significant to the cars handling. For an aero car with ground effects, the jacking forces could be important, but I guess that's why most people run fairly low roll centers.

To give an example: Lets suppose you have a car which is really tight. You want to free it up (reduce rear grip/increase front grip) so you raise the rear end. As the rear of the car moves up, the RC moves up, the jacking forces will also increase too. But the most noticeable difference is that weight transfer in the rear of the car will increase, and front weight transfer will decrease.

CAPPY: Don't you got some fluent to be running http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

J.R.

10-24-2008, 06:07 AM

Originally posted by Mike Cook:
To give an example: Lets suppose you have a car which is really tight. You want to free it up (reduce rear grip/increase front grip) so you raise the rear end. As the rear of the car moves up, the RC moves up, the jacking forces will also increase too. But the most noticeable difference is that weight transfer in the rear of the car will increase, and front weight transfer will decrease.

In the calcs that you did there was no interdependence between front and rear, so how to jacking forces at the rear balance with the front? Is it a ratio similar to ARB calcs?

J.R.

10-24-2008, 06:34 AM

New thought as well. People always talk about diagonal load transfers due to caster and KPI, are these load transfers due to the lift/squat that these angles cause, multiplied by the spring rate of the respective corner? I was thinking about the jacking force things effecting a lift to the chasis, and a redistribution of weight, so this concept (caster, KPI lift = DWT) makes sense to me, but I'm not sure if I'm correct in assuming that.

exFSAE

10-24-2008, 07:28 AM

I can't imagine caster moves much weight around. I could be wrong, I've never checked it on scales. But how much are you talking about, compared to multiple hundreds of pounds of LLT?

BilletB

10-24-2008, 09:49 AM

Originally posted by J.R.:
So from this I gather you don't believe in jacking loading at all. Or does the LT calcs that mike cook did only apply in transient cornering? The jacking effects are definately not the "Just dependent on lateral accel, cg and track" which is what I thought as well. According to the jacking calcs that were done, it is actually possible to transfer more load to the outside tire with jacking forces. This would add to the traditional LLT. Or do these jacking forces have to balance front to rear with the total LLT, similar to adding an ARB?

I think you're still missing the point. I don't know what you mean by 'I don't believe in jacking loading' because I hope I made it clear that I believe jacking forces are real and present!

Remember jacking itself does not change total LLT or have anything to do with LLTD. The only way lateral load transfer can be effected by jacking is if your jacking force is significant enough to really cause the cg of your sprung mass to move. The idea that jacking can create a momentary increase in net Fz tire force during transients is true, but theoretically it's immediately followed by a net decrease in Fz tire force. This comes from the reactions of what it's doing to your sprung mass. Once you reach steady state corner the jacking forces DON'T go away, but it is not changing the total LLT or LLTD. In fact the portion of the LLT that is attributed to jacking is sprung mass load transfer that is no longer available (or taken away if you will) for tuning via front and rear roll stiffnesses.

Try working it out and reading through some of these replies again.

1)Steady state TLLT is a function of your car's cg height and trackwidth given some lateral accel.

2)Jacking forces themselves are geometric forces on the sprung mass and are created from Fy contact patch forces and have no ability to directly change TLLT or LLTD. They only can effect load transfer by effecting the parameter of cg height stated in #1.

J.R.

10-24-2008, 12:10 PM

My confusion is this. Mike Cook did calculations to show that jacking forces change the loading on your tyres from a pure calculation using Ay, track, and cg. If his calculations are correct, then this shows that you have a change in lateral load transfer. This seems to make sense to me, since if you take a force going into bars that are not horizontal, and they transmit force, then there is a vertical force component, correct?

I think where the confusion comes from is my misuse of the term "load Transfer." I am using it in the sense of changing Fz as seen by the tyres. So, in effect,

Fz = static tyre load +/- jacking forces

Jacking forces meaning the steady state force caused by the geometrical orientation of the suspension members, reacting lateral forces through non- horizontal bars and creating a vertical force, which unloads the springs, and comes to a steady state value.

This is what I meant when I said that jacking forces cause a change in lateral load transfer, and I admit I mis-spoke.

So, is everyone in agreement with saying that during steady state, mid cornering, there is additional loading on a tyre due to jacking forces, as per Mike's calc's on the first page?

BilletB

10-24-2008, 01:55 PM

Originally posted by J.R.:
So, is everyone in agreement with saying that during steady state, mid cornering, there is additional loading on a tyre due to jacking forces, as per Mike's calc's on the first page?

No, not in agreement...

Here is what Mike posted:

Mike:
If you go back to my math for jacking forces, you see that the outside tire produced 13.5lb up and the inside produced 4.5 down at the chassis. Well these forces have equal and opposite forces acting at the tires. So while the net force on the chassis was only 9lb when cornering, the outside tire will get 13.5lb heavier and the inside tire will get 4.5lb lighter. So what are the new loads on the tires?

outside = 225+13.5 = 238.5
inside = 75-4.5 = 70.5

I looked over the calculations.. And I struggle following logic in forum posted math, but here is where I find the error.

I hate roll centers, especially kinematic roll centers. Mike seems to be summing forces at his. Does this really happen? <--devil's advocate. I believe that it's more accurate to model the jacking force as loading or unloading that side of the chassis and forget this non-existent, imaginary, and made-up term, "rollcenter". A net jacking force doesn't lift or lower the chassis in a purely pitch mode.

Using Mike's numbers it works out like this. 225 lbs on the outside tire. My upwards jacking force is 13.5 lbs. This adds 13.5 to the tire Fz. But, that 13.5 lifting force unloads my wheelrate by the same 13.5 lbs. So that decreases Fz by 13.5. So...
Fz on tire: 225 + 13.5 - 13.5 = 225

Whatever force we gain from jacking is directly negating that much spring force because our Fz component of Fy acts at the tire and so does our wheelrate. It works the same for negative jacking and/or looking at the inside tire. This is all very much analogous to anti-squat and anti-dive. Same damn thing.

CappyUMD

10-24-2008, 10:33 PM

Yes, let's be very clear that in steady state, the normal forces of the tires must equal the weight of the car. Otherwise, the car would continue accelerating through the ground or into the sky.

The reason that jacking can cause transient changes in normal force is because lateral forces will generate jacking forces almost instantly, but springs require displacement (time) to change their reaction forces. Reaction forces of the springs will not decrease until the ride height increases. So there is a period when spring reactions have not changed, but jacking forces are present and add to spring forces.

When the vertical velocity of the cg is decreasing, the decrease in normal load due to relaxation of the springs outweighs the increase in normal load due to jacking (total normal load on tires is less than the weight of the car).

A car with damping will eventually settle to the equilibrium ride height (which is higher than ride height when driving straight). At equilibrium, the cg is not accelerating vertically, and so the normal forces of the tires add up to the weight of the car.

J.R.

10-25-2008, 10:40 AM

Your no actually reading what I'm saying. Benny, wrong calcs, Crappy, read what I said.

http://s420.photobucket.com/albums/pp290/jr226/?action=view&current=jacking.jpg

FBD (http://s420.photobucket.com/albums/pp290/jr226/?action=view&current=jacking.jpg)

I'm am convinced that jacking forces cause a change in wheel loads, but THE TOTAL OF ALL FOUR WHEELS = VEHICLE WEIGHT.

Look at how the forces act. The Red arrows are forces that directly act on the car. Fy, Weight and Fz. Now the way that Fy is transmitted is through the suspension links. For the bottom links as drawn, there is no vertical force acting, but the top links have a vertical force component, voila jacking. Now these forces will NOT contribute to a total increase in wheel loads, but they will cause the car to rotate, since the forces act as moments at the CG. These forces will also cause an increase on the load of the outside wheel, and a corresponding decrease on the load of the inside wheel. Now, how much this will effect a change in wheel loads depends heavily on the geometry at that moment, including steady state, but the net sum of vertical forces must = the total weight of the vehicle for SS no acceleration. So, because of this, the forces that are reacted through the upper suspension links in this picture (in real life, both top and bottom would cause vertical force since they wont remain horizontal) will cause an increase in the load that the outside tyre has to produce, and a decrease in the loading that the inside tyre has to produce, therefore, CHANGING WHEEL LOADS, not causing a load transfer (subtle difference). But the point is that jacking will cause a different Fz at the tyres due to linkage design, EVEN IN STEADY STATE.

Looking at our calculated load transfers for U buff, this year we should have according to theory w/o jacking, had ~25lbf on our inside front. When we test in skidpad/ high g corners, we pick up the inside wheel. Explanation? There are other forces taking normal load from the inside tyre. Answer, jacking and possibly diagonal due to caster.

If anyone wants to disagree, please read what I have actually said, and not just what I have been saying before, I think I have it right now, but I could be off base in assumptions somewhere.

Wesley

10-25-2008, 10:50 AM

How is a change in wheel loads not simply load transfer if vehicle weight is the only force acting vertically?

J.R.

10-25-2008, 10:55 AM

Originally posted by Wesley:
How is a change in wheel loads not simply load transfer if vehicle weight is the only force acting vertically?

b/c the load is coming from a vertical jacking force, not a lateral acceleration. I don't know what exactly is correct in terms of terminology, that's just how I separate the two in my head.

BilletB

10-25-2008, 11:51 AM

Originally posted by J.R.:
Your no actually reading what I'm saying. Benny, wrong calcs, Crappy, read what I said.

http://s420.photobucket.com/albums/pp290/jr226/?action=view&current=jacking.jpg

FBD (http://s420.photobucket.com/albums/pp290/jr226/?action=view&current=jacking.jpg)

I'm am convinced that jacking forces cause a change in wheel loads, but THE TOTAL OF ALL FOUR WHEELS = VEHICLE WEIGHT.

Look at how the forces act. The Red arrows are forces that directly act on the car. Fy, Weight and Fz. Now the way that Fy is transmitted is through the suspension links. For the bottom links as drawn, there is no vertical force acting, but the top links have a vertical force component, voila jacking. Now these forces will NOT contribute to a total increase in wheel loads, but they will cause the car to rotate, since the forces act as moments at the CG. These forces will also cause an increase on the load of the outside wheel, and a corresponding decrease on the load of the inside wheel. Now, how much this will effect a change in wheel loads depends heavily on the geometry at that moment, including steady state, but the net sum of vertical forces must = the total weight of the vehicle for SS no acceleration. So, because of this, the forces that are reacted through the upper suspension links in this picture (in real life, both top and bottom would cause vertical force since they wont remain horizontal) will cause an increase in the load that the outside tyre has to produce, and a decrease in the loading that the inside tyre has to produce, therefore, CHANGING WHEEL LOADS, not causing a load transfer (subtle difference). But the point is that jacking will cause a different Fz at the tyres due to linkage design, EVEN IN STEADY STATE.

Looking at our calculated load transfers for U buff, this year we should have according to theory w/o jacking, had ~25lbf on our inside front. When we test in skidpad/ high g corners, we pick up the inside wheel. Explanation? There are other forces taking normal load from the inside tyre. Answer, jacking and possibly diagonal due to caster.

If anyone wants to disagree, please read what I have actually said, and not just what I have been saying before, I think I have it right now, but I could be off base in assumptions somewhere.

JR,
I read what you've written and drawn, but you're still not right on. Jacking forces are not going to change wheel loads.

In your FBD you show the car accelerating steady state to the right and have a lateral force acting at the wheel center and show a net downwards jacking from the outside tire. That lateral force happens at the tire contact patch and cannot just be moved to the wheel center. The scenario of your FBD with those a-arm angles and ss lateral accel to the right will create a net upwards jacking from the outside tire. This should be clue #1 something is off with the FBD.

You're making things a lot more complicated than they really are. I think we've all been on the same page all along about all vertical forces summing to weight in steady state. However, there is no changing load on the tires in steady state cornering by what is dictated by cg height and trackwidth (ie. if you want to prove that wheel loads are modified by jacking you need to show how jacking lowers or raises the cg). If you want to keep with the complicated FBD's draw the lateral force at the tire contact patch and think about the force of the spring and how it acts at the wheel. Jacking forces and your springs are just simply sharing the load of the sprung mass.

G'luck!

J.R.

10-25-2008, 12:21 PM

Originally posted by BennyHL:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by J.R.:
Your no actually reading what I'm saying. Benny, wrong calcs, Crappy, read what I said.

http://s420.photobucket.com/albums/pp290/jr226/?action=view&current=jacking.jpg

FBD (http://s420.photobucket.com/albums/pp290/jr226/?action=view&current=jacking.jpg)

I'm am convinced that jacking forces cause a change in wheel loads, but THE TOTAL OF ALL FOUR WHEELS = VEHICLE WEIGHT.

Look at how the forces act. The Red arrows are forces that directly act on the car. Fy, Weight and Fz. Now the way that Fy is transmitted is through the suspension links. For the bottom links as drawn, there is no vertical force acting, but the top links have a vertical force component, voila jacking. Now these forces will NOT contribute to a total increase in wheel loads, but they will cause the car to rotate, since the forces act as moments at the CG. These forces will also cause an increase on the load of the outside wheel, and a corresponding decrease on the load of the inside wheel. Now, how much this will effect a change in wheel loads depends heavily on the geometry at that moment, including steady state, but the net sum of vertical forces must = the total weight of the vehicle for SS no acceleration. So, because of this, the forces that are reacted through the upper suspension links in this picture (in real life, both top and bottom would cause vertical force since they wont remain horizontal) will cause an increase in the load that the outside tyre has to produce, and a decrease in the loading that the inside tyre has to produce, therefore, CHANGING WHEEL LOADS, not causing a load transfer (subtle difference). But the point is that jacking will cause a different Fz at the tyres due to linkage design, EVEN IN STEADY STATE.

Looking at our calculated load transfers for U buff, this year we should have according to theory w/o jacking, had ~25lbf on our inside front. When we test in skidpad/ high g corners, we pick up the inside wheel. Explanation? There are other forces taking normal load from the inside tyre. Answer, jacking and possibly diagonal due to caster.

If anyone wants to disagree, please read what I have actually said, and not just what I have been saying before, I think I have it right now, but I could be off base in assumptions somewhere.

JR,
I read what you've written and drawn, but you're still not right on. Jacking forces are not going to change wheel loads.

In your FBD you show the car accelerating steady state to the right and have a lateral force acting at the wheel center and show a net downwards jacking from the outside tire. That lateral force happens at the tire contact patch and cannot just be moved to the wheel center. The scenario of your FBD with those a-arm angles and ss lateral accel to the right will create a net upwards jacking from the outside tire. This should be clue #1 something is off with the FBD.

You're making things a lot more complicated than they really are. I think we've all been on the same page all along about all vertical forces summing to weight in steady state. However, there is no changing load on the tires in steady state cornering by what is dictated by cg height and trackwidth (ie. if you want to prove that wheel loads are modified by jacking you need to show how jacking lowers or raises the cg). If you want to keep with the complicated FBD's draw the lateral force at the tire contact patch and think about the force of the spring and how it acts at the wheel. Jacking forces and your springs are just simply sharing the load of the sprung mass.

G'luck! </div></BLOCKQUOTE>

I'm 99% sure that the basic principles of statics (move a force, replace with a force and moment) still apply to the vehicle realm, so I see nothing wrong with moving the force to the wheel center. The moment will put a twist on the chassis.

"Complicated Free Body Diagrams" We're engineering students, we live with FBD. I am coming up with a logical reasoning for why jacking forces exist, and the only response that I'm getting is, "Well this is the way an equation looks in a text book so I must be right"

So a book says LLT is only based on Track, Ay and CGh. Okay, so what. How does this work? I have a FBD that says that there may be more going on than is suggested by a some book.

I may be completely wrong with this, but I don't understand the reasoning why I could be wrong. I'm not willing to stop at the point where I say that something is this way because thats what I read without actually understanding the reasons why things fall that way.

Am I complicating things? Yes. Does it need to be this complicated? Maybe. If I am right, then the extra force can be used to make minor adjustments in wheel loads. Coming back to using a SR, we need to unload the rear inside enough so that the Longitudinal force that the inside is producing is small enough to not be detrimental under cornering.

I think my lack of understanding comes from the spring region, which isn't in the FBD. That will supply the downward force on the wheel. The component of force which is on the outboard tyre compresses the spring more, and this in turn places more load on the wheel and removes it on the right.

BilletB

10-25-2008, 01:12 PM

Originally posted by J.R.:
I'm 99% sure that the basic principles of statics (move a force, replace with a force and moment) still apply to the vehicle realm, so I see nothing wrong with moving the force to the wheel center. The moment will put a twist on the chassis.
Yes! the basic principles do apply, thus you cannot trash that moment. It is now applied at the center of your upright and YOU NEED IT to continue calculations!!

Tell me how your new force on that outside tire nets a downward jacking on the chassis when it should net an upwards jacking force!

Originally posted by J.R.:
"Complicated Free Body Diagrams" We're engineering students, we live with FBD. I am coming up with a logical reasoning for why jacking forces exist, and the only response that I'm getting is, "Well this is the way an equation looks in a text book so I must be right"
You couldn't be more wrong. I have never read a race car engineering book in my life. Just referenced them to check and verify my equations. I build and design suspensions due to dominating factors and how it acts in the real world. Nothing wrong with a complicated FBD but you've already messed it up. It's overly complicated because you can quickly do the same thing analyzing instant centers and n-line slopes. These are the important things to look at to quickly get correct calcs. N-lines slopes and IC's. It is the heart of a double wishbone suspension analysis.

Originally posted by J.R.:
So a book says LLT is only based on Track, Ay and CGh. Okay, so what. How does this work? I have a FBD that says that there may be more going on than is suggested by a some book.
This is the big picture. The LLT as dictated by by CGh and trackwidth is a real world occurrence that cannot be altered. It is the fundamental and underlying dominant factor.

Originally posted by J.R.:
I may be completely wrong with this, but I don't understand the reasoning why I could be wrong. I'm not willing to stop at the point where I say that something is this way because thats what I read without actually understanding the reasons why things fall that way.
That's well and good. Just be sure all your reasoning is correct. At this point it is not. Once you work through it in the complicated but correct fashion you will achieve the same results as using IC's, n-lines, forces, and wheelrates!

Originally posted by J.R.:
I think my lack of understanding comes from the spring region, which isn't in the FBD. That will supply the downward force on the wheel. The component of force which is on the outboard tyre compresses the spring more, and this in turn places more load on the wheel and removes it on the right.
This is important. Remember, all your sprung weight must be account for in steady state through all 4 springs. If it's not in the spring, it can be a jacking force which is fine but then all it does is effect roll angle, not wheel loads. If it moves your CG then it causes a new LLT and the new amount of load transfer needs to be calculated for at the wheels, plus the new change in jacking forces. If that causes more CG movement then it continues on and on and on.

Eventually you should come to find out that if jacking really raises your cg it compounds around a corner. Jacking raises cg, more load is transfer, jacking now increases, cg raises, more load is transfered, jacking increases........ A case like that is a bad scenario. But if jacking doesn't raise your cg, it won't change Fz at the contact patch.