Race car suspension Class

[Ranger Mike]

Kozy I absolutely agree with you on the Gs. but...as I have said from the start.
.this is simply a method to get close to proper spring set up..not to be a blessed algorithm ..never meant it to be...

and the 1.15 G is not absolute as tracks differ..just a useful baseline
As you intimated to me the true diagonal weight transfer is complex..add to this the driver influence is all over the map due to tapping the brake pedal thus impacting the diagonal weight.
I use this as a start point...as long as all are aware of this it can be useful.

 

[Kozy]

I guess as a first pass kind of thing it works well as it's really adding a nice factor of safety. If you can hold that figure then you're not going to get into any trouble. Just seems a very different approach that is all, but then I guess I have to remember that I'd be coming from the angle of having penned out all the algorithms...

For a clean sheet I would probably start with with working out the ride frequencies.
Start out with a target hz of 2x lateral acceleration in G. So 1g = 2hz.
From that, figure out the load transfer and roll angles using the app.
Add in the bar specs, and tune to get the desired roll angle and load transfer distribution.

I've updated my app to display the break down of the load transfers, geometric, unsprung and elastic, as well as the distribution F/R. I would be interested to hear your thoughts on it.

My take on Smith's situation:

The front ride frequency is 2.5hz, the rear is 1.7hz.
Without any RC effects or bars, that puts the car at about 67% front roll stiffness and 59% front load transfer at 1.15G.
Front load transfer 549lbs, rear 284lbs of which 524lbs and 259lbs is elastic respectively.
Putting 524lbs into a 403lb/in wheel rate gives 1.3" deflection up front and 259lbs into a 200lb/in wheel rate also gives 1.3" wheel deflection at the rear.
Thus the chassis roll angle over a 64" track is 2.3°.

The springs alone look to be 90% of what you need.

I am not really sure on how the bars are rated though, to assess their effect on things. Is that rating per degree of roll, or per degree of twist in the bar?

 

[smithdl4]

Kozy;
As I understand it the bars are rated in LBs per 5 degrees of twist per their manufacturer. Also, I think the wheel rate for both ends would be smaller as the springs are canted over 9 degree which changes the actual spring rate the suspension sees, correct?

Ranger Mike;
The rear has an MR of 1 due to its construction. It is an SLA IRS. The spindle is designed to have a bolt that runs through the bottom that captures the coilover shock, the outer spherical bearing, and the toe control arm and is parallel to the chassis (or pretty much). My question about bars was not on how to compute their force but their force applied to the suspension. It attaches at the same point on the suspension as the spring and since it is a spring also it just seemed to me you could not use the full rate but would need to multiply its rate by the same MR to arrive at a reduced rate like the spring. Maybe I'm just over analyzing.

Gentlemen;
Thank you both for taking the time to answer my questions as I try to really understand what's going on with suspensions. Dave

 

[Kozy]

Yes the bar gets its force reduced by the installation ratio the same as the spring does.

Still a little confused about the rate though. If it is 355lb/5degrees, that would suggest that the arm length has not been accounts for correct?

My calculations on bar rate take the diameter, length and arm length to give a rate in lbs per inch deflection at the end of the arms, which is directly comparable with the coil spring rate.

 

[Ranger Mike]

Kozy..
I agree on ARB rate. There are numerous applications. Hollow and solid and many types of linkage. My assumption is the driver knows the ARB rate as hooked up on the vehicle. Ref the calc on weight..please bear with me..you say we have 524# weight transfer to the front..right?

Smith.. thank you for the clarification. Also if you note the attached pic on prior page the spring angle mount correction factor is there ..pretty small for 10 degrees.

 

[Kozy]

There is 524lbs load transfer 'from the inside front tyre to the outside front tyre'. That 524lbs is always over the front axle though.

Similarly, there is 259lbs transfer from the inside rear to the outside rear, however again it stays over the rear axle and does not shift diagonally.

 

[Ranger Mike]

Thanks Kozy..my point is you need to counter this with the ARb and the spring..right? So would you not need a 262# ARB and an effective wheel rate of 262#? A 350# spring times the .765 motion rate is 267# ..right?

 

[Kozy]

You've lost me again I think. Where have you got the 262lb wheel rate/ARB from?

The load is dependent on the springs. If you swap those rates and put 700lbs/in at the rear and 200lbs/in at the front, instead of 524lbs across the front and 259lbs across the rear, you'd get 111lbs across the front and 697lbs across the rear.

In this scenario, you'd need to add a significant front ARB to even it up. The wheel rates from the springs would be 115lbs/in and 700lbs/in F/R, so you'd need something in the order of 685lbs/in contribution from the bar to equalise the load transfer.

This seems fairly typical of MR racecar setup, where the rearwards weight distribution requires stiff rear springs to set the ride frequencies correctly, and then use a large front ARB to set the load transfer distribution so that it does *NOT (Edited to add that) suffer dangerous oversteer.

The opposite is also true of performance FWD cars. The heavy nose requires stiffer ride springs in front to support, but this alone would cause a lot of understeer. Solution: Fit a big rear ARB to take a large portion of the load transfer across the back axle to neutralise understeer.

 

[smithdl4]

Kozy;
The bar rate includes the arm length. It is a tubular bar made by Speedway Engineering. Its 37 1/2 inch x 1 1/4 inch x 0.188 wall tubular bar in front using 12 inch arms. Rear is 37 1/2 inch x 1 1/4 x 0.095 wall
tubular bar using 10 inch arms. They measure using an Intercomp sway bar tester. That's as much as I can tell you about how it's rated.

Appreciate the clarification on the ARB rate when attached other that at the ball joint.

I believe Ranger Mike got the 262 value from his last statement in an earlier post to me (593 previous page) where he mentioned that the ARB should equal the springs or something on that order. Half of your 524 equals 262 for the ARB therefore the spring rate should be close to 262lbs supporting the total of 524. Using Ranger Mikes example then the 350 lb springs equate to 267 lbs rate when reduced by the MR. Total rate would be 529 lb to handle the 524 lb weight transfer. Am I on track Ranger Mike? Dave

 

[Ranger Mike]

ok break time.........
please let me reply upon reflection ,,,,,

 

[Ranger Mike]

Smith..,you are correct...you want the total ARB and front spring package to equal the maximum amount of weight coming forward AND amount of aero downforce you will be experiencing, If you have 524 pounds moving from the LF to the Right Front..you would want to split this so ARB handle 50% and RF spring 50%. 524 / 2 = 262 so a 350# spring with .765 MR= 267#
to my way of jaded cynical thinking I prefer my method of finding the total weight being slung around at 1.15Gs, figure out max amount coming forward. Then add in diagonal weight and the total amount is divided by 3...it will be close ..we still have not added in aero down force so if you are a little stiff...you should know it at tune and test day..THIS WORKS FOR ME..but not best correct number as Kozy can tell you..

 

[Kozy]

Mike have you tried entering some setups you are familiar with on my calculator and comparing the notes to what you get? I'd be interested to hear what you think of the figures.

 

[Ranger Mike]

kozy..will do!

 

[rick7343]

Interesting stuff!

How do you determine how far you want the suspension to dive or compress under the weight transfer ?

 

[Kozy]

Kozy;
The bar rate includes the arm length. It is a tubular bar made by Speedway Engineering. Its 37 1/2 inch x 1 1/4 inch x 0.188 wall tubular bar in front using 12 inch arms. Rear is 37 1/2 inch x 1 1/4 x 0.095 wall
tubular bar using 10 inch arms. They measure using an Intercomp sway bar tester. That's as much as I can tell you about how it's rated.

OK so I bunged those stats in with the rest of the setup.

It takes the roll angle from 2.3° with springs alone, down to 1.07°. FLLTD is 51.2% which assuming the weight distribution is 49% seems nicely balanced. Tyre stagger may affect this though.

Overall, it looks pretty stiff, but I have no idea what sort of roll angle is reasonable at 1.15G. These figures are assuming 0 roll center heights, above ground RCs will add some more roll stiffness.

 

[Kozy]

Interesting stuff!

How do you determine how far you want the suspension to dive or compress under the weight transfer ?

Is that at me or Mike?

From a design standpoint, if you've got camber curves for the suspension, you can figure out what sort of wheel displacements are going to keep the tyre in it's optimum range. That is getting pretty advanced though.

From a testing standpoint, you'd tune the camber with a tyre pyrometer. If you can get a nice temperature profile without maxing out the adjustment in the negative direction, then the car has enough roll stiffness. If at maximum negative camber you're still leaning on the outer edge hard, then you need more roll stiffness. Add roll stiffness then re-tune until you get where you want to be. Do not add more roll stiffness beyond this point though, you want as little as you can get away with really.

Longitudinal load transfer and pitch motion is a little more tricky to calculate as it requires knowledge of pitch center height. Roll center heights are hard enough to find! Again, pretty advanced stuff.

 

[Ranger Mike]

Kozy..I got to dig up my set up sheets and enter the data on your calculator and will be doing this ..this weekend..The track condition will dictate the suspension travel. The total CAMBER build also dictates the travel. it is all about tire contact.
In racing, we can assume a certain degree of ideal conditions, or at least more ideal than public roads. In a stock street car, even notoriously "bumpy" race courses feel glass smooth compared to most public roads. In these conditions, the purpose of the spring can be focused to maintain maximum and consistent contact of the tire with the relatively much smoother road surface. Under these conditions, very little wheel assembly travel is required. The spring can be optimized for smaller wheel travel conditions. For example, a CART or Formula 1 race car driven on smooth courses may only have 1/4 to 1/2" of total suspension travel! On our Formula Car I bump steer it on 2 inch bump and two inch dive. The linkage type and layout influences the camber build so a full blown Indy type suspension may have 1.05” spring compression on the LF in a right hand turn. The LF tire would build 1.3 degrees and the RF may have 1.3” droop and the RF tire goes to 3 degree camber.

On the stock car suspension and considering the pitiful condition of the tracks these days we need at least 3 inch travel +/-. You will not have this much travel required on the track but entering and exiting the race track to pit road , the scales and the access roads around the track is where you need the travel. Some tracks are literally 3 inches higher than the infield and all you need is to be forced off the track by a shunt or race hazard and you are in the ruff. The desired camber build is 4.25 degree negative camber in 3 inch of bump travel on right front. Left front should have .5 to .75 degree positive camber per inch of rebound travel. I bump steer over +/- 3 inch travel..keep it under .040” per inch.

 

[rick7343]

KOZY: is that calculator supposed to be interactive? When I download it, there is double typing and it won't let me change any numbers... Maybe my old browser?

Thanks!

 

[Kozy]

New link here:

http://blackartracing.zxq.net/Chassis/Load Transfer 2.php

The inputs in cyan should be editable yes. The results in green are not.

What browser are you using? I use Chrome/Firefox. It will not work on Internet Explorer.

It should look like this:

 

[rick7343]

That would explain it! thanx!

 

[Kozy]

Yes IE is a PITA to script for. Requires it's own subset of code or something like that (I am a total n00b at web development) which to be honest, I can't be bothered with.

I could do with a popup warning advising visitors to upgrade their browser though!

 

[Kozy]

Can't go back to edit previous posts, but the old link no longer works.

New one here. http://blackartracing.zxq.net/Chassis/LoadTransferDC2.php

 

[Ranger Mike]

I made a mistake

My previous post about using diagonal weight is WRONG.

Kozy
I plugged in the numbers on the latest version calculator. Your calc said total load transfer 801 lbs. I came up with 813 on my post on physics forum.
Your calc said 534 lbs. front end 271 rear..I came up with 521 lbs. front and 292 rear weight.
My previous post about using diagonal weight is WRONG.

I made a major mistake by not adding in t he ARB motion rate which is huge. If anyone wants to know specifics look on page 22 post 358 of this post. I apologize for this error. Disregard my calculation of diagonal weight transfer.
My other disconnect on this is the heavy amount of rear weight transfer but the effective spring rate that works is only 95 pounds.
We have between 271 and 292 pounds of weight coming from the left to the right side of the car. Then it hit me like a ton of bricks. Only 95 pounds was slinging through the roll centers the rest was straight lateral transfer through the rear axel center line to the outside tire contact patch. Duh!
I had to add up the UNSPRUNG weight and it proved out.
Kozy I think your calculator is a very good tool to use.

UNSPRUNG Weight
coil spring front 17# each
coil spring rear 11# eacj
Ford 9 inch floater rear end 133#
Quick change rear end 208#
brakes front and rear ( disc) 20#
Rear Trailing Arms 28# each
Spindle, hub, backing plate assbly 35# each

the following are 50% sprung and 50% unsprung
ARB ( sway bar) 15#
Tubular upper A-arm 5#
Shock absorber (damper) 4# but depends on mounting .
Could be 1 # sprung 3# unsprung if hung like stock production automotive.
If you reverse you have 3 # sprung 1# unsprung.

 

[Kozy]

Christ, that's a lotta unsprung weight!

 

[Ranger Mike]

yes..that was what was giving me fits, kozy. Coil overs take some of it out but 50 pound tire / wheels and that heavy differential add up quick. Your calculator was a very good deal and confirms my post on calculating the weight ( without diagonal figures).

 

[Ranger Mike]

All the chassis tuning in the world will not help when the car is 3 seconds slower than the number one car qualifying. A compression test told the story. Two out of four cylinders were down 15 % psi. If we had a V8 the impact would not be as dramatic but we were getting pulled two car lengths on the straights...No way can you win races when this happens. Now the decision is to race or not race the car at the final event. I do not like losing! But a day at the track beats a week at work, right??

 

[rick7343]

Just order 4 more cylinders and fix the others!

 

[Ranger Mike]

I did not go to the track the last race weekend of the year. We won both days...With the engine that had two cylinders 15% low...and we found a front ARB that was cracked. Now wonder the cr was off. The front ARB sway bar arm was giving way and after swapping in a new one..we were back to the old fast time. Lesson to remember. Check for cracks, have regular maintenance scheduled, and never assume anything. Or don't have the crew cheif at the track..i will have to test this theory next year!

 

[Kozy]

Good effort Mike! A crack in the ARB would stump even the best engineers I reckon!

I've written up an article with built in calculations based on my model of load transfer, to complement the calculator I wrote. First part of a multi-part set of articles anyway.

http://blackartdynamics.com/Chassis_Articles/Tyres_and_load_transfer.php

 

[radracr]

Hi Ranger Mike! I got to say, I good friend of mine referred me to this site and…..WOW!

My question: How best to determine shock compression / rebound values front and rear? We have determined that our Motion Ratio is 2.0 on the Front A-arms. The car is a mid-70s A-body.

Thanks again for the thread and incredible insight!

And, I recently purchased the Circle Track Analyzer 3.6. (Yes, my application is circle track.)

 

[Ranger Mike]

Thank you much Radracr..kind words always good to hear. I assume you are running Chevy Chevelle A body. what is the race ready weight?

 

[radracr]

Actually it's a 76 Cutlass - but it's the same - 112" wheel base.

Class rules are 3200# MIN - I'm at 3270.
Class rules are 56% MAX left side - I'm at 55.8
Class rules are 47.5%MAX rear - I'm at 44% - 45.5 with a full 22 gallon fuel cell and 3380#.

Left side is running 3" offset 8" wheels.
Right side is running 2" offset 8" wheels with 1" spacers @ RF. RR = 1/2" spacer.
I have not checked track width.

We were running LF = 14/4, RF = 12/4 but it seemed that once we took out the anti dive and went to an 800# RF, 700# LF springs, the RF shock could not keep up - the RF would smack the track surface going into the turns. We were previously at, with anti dive 900# RF, 800# LF

 

[Ranger Mike]

Sounds like you are real close. The left side wt. is perfect. it would be great to move 65# to the rear or take it off the car period. You are at the classic point of making the front end a tad too soft. Sounds like 850 optimum Rf spring. Maybe #750 on LF. We always ran 76 shocks on the front on the monte carlo. Can you run 3way adjustable shocks?

 

[radracr]

Stock shocks in stock location. We can run rebuidable - hence the 12/4 and 14/4 - but not adjustable. A lot of the racers in the other classes are running the same shocks I am but they are outboard. I'm sure that the motion ratio plays a role.

The current thought process is to leave the rebound alone but increase the compression to 8, 9, or 10? Is there any way to determine this upfront before forking out the cash on a new set?

Also, we plugged in some 'rough' numbers and it said that our RF natural spring frequency is 1.51, while the rears are 1.53 ish. The LF was 1.1. I thought the RF was to be 2.0? Currently, RC = +1.7, 7.0 to the right. The car, just like the software says, has always been a little too oversteer and therefore no bite off.

Thanks,

 

[Ranger Mike]

Move the Rc back to 3 inch offset to the right. You have too much leverage at 7 inch offset. This lifts the left side of the car. You may have too much rebound on the shocks up front. Present set up means the rf spring is pancaked, the shock is tying the rf down and probably same on lf shock.
On corner off acceleration there is not enough lift at lf front to transfer weight to RR tire to hook up off the turn so is loose off.

bumping up compression is not the answer..fix RC first. then change springs 50 # then tune with shocks
but..dial in camber build and bump steer too