I have wondered about this for awhile and thought I'd ask around.
I see that some people like to buy a pre-packaged kit like the TVS from Hotchkis or PST or something similar, but I'm still curious about it.
When a person sets out to improve the handling of their car, their personal tastes will differ from the next guy. I'm interested in knowing how these engineers arrive at the specific parts that become part of the kit. For example, with engines, we expect the exhaust valve to be smaller than the intake valve because it always is. In the case of suspensions, the front roll stiffness or spring rate is always higher than the rear. I mean in terms of our front engine/rear drive cars.
I guess I am asking, is there a percentage of roll stiffness that they use to determine what to use? If the front is say...a *100, is the rear a *75? More at the rear causes oversteer, less will result in understeer?
Another hypothetical:
3500 lb 72 Duster with all iron 360, battery in trunk.
.92 Torsion bars 1.250 front sway bar KYB shocks.
Stock leafs with 1 additional leaf, .75 rear bar, KYB shocks.
I haven't been able to drive the car on the street yet for many reasons but I am just interested in the science of all of it. I don't want to just do what others did, I want to learn the math of it. Do people start off setting the car to lean/roll as little as possible while giving a decent ride, then fine tune the chassis with small changes?
I am interested in whatever anyone can spill on the subject.
Thanks, Jeff

Time to read

I have this

http://www.amazon.com/Build-Ultimate-Super-Street-Mopar/dp/B00538MYA8

For starters, the the whole bag of goodies Carroll Smith's Engineer to Win series

http://www.amazon.com/Engineer-Motorbooks-Workshop-Carroll-Smith/dp/0879381868

The MP oval and circle track books have some interesting information.

I own the car on the cover of the Ultimate Super Street Mopar. And that is a very good resource for information. Andy F's book in another and if you can find them, the old Direct Connection Circle/Oval track books have lots of stuff too. The Mopar Performance versions of those books are watered down and missing a lot of the information you are after.

Sorry Andy, forgot about your book

I agree, some time library time would be in order. I have read a plethora of books including those already mentioned. My takeaway, for the non engineer, the best book, bar none is:

http://www.amazon.com/Think-Fast-Racers-Why-To-Winning-ebook/dp/B00IJ2MB0I

One common thread thru out the book, which many newbies seem to ignore is, because they are looking for the holy grail, is, "everything effects everything else". You can't get around that as your set-up becomes more sorted out.

Andy's book gives an easy and simple answer for it all.

Another easy answer is simply set your car up to fall on the neutral handling line as shown below from the Mike Martin book, Mopar Suspensions.



Of course, getting front weight percentage is easy. Its all the calculations that go into figuring out roll couple front and rear that make it tricky. Martin's book gives info on how to do this, as do the other books listed above and the Mopar Chassis and Oval Track books also have info on it. Where it gets tricky is you can't just plug in rates. You have to find motion ratios of the sway bars front and rear and rear springs. T-bars are 1:1, so that part is easy. Once you have all this, then yes, its possible to set up a formula that show you where your car falls and then you can target the percentages on the chart above. I created one for my Challenger so I can see what changes may do to it, but I also have a configuration on my car that is different than stock using many non stock components, so it doesn't easily translate to other vehicles. I also set my calculator up to back into required leaf spring rates as these are the easiest part to customize. There are only so many t-bar and sway bar pieces out there, so we are limited on how wide a range of front end set ups we can get, whereas leaf spring can more readily be modified to match the front system.

The theoretical handling line is fine, for a starting point. But each driver has his idiosyncrasies which will make every setup a bit different even on otherwise identical examples.

This

The way engineers think about it is controlling the tire patch. The tires with the heaviest load are more likely to be overloaded than the tires with less load. On a nose heavy car the front tires take more load than rear tires. You can fix that by putting bigger tires up front but almost nobody does that since it looks weird.

The whole problem of handling for a nose heavy car then revolves around the question of how do you keep the front tires from being overloaded.

Some Mopar guys don't even like the 11.75 brake swap since it widens the front track width. But widening the front track width is exactly the right thing to do when you have a nose heavy car. If you aren't going to widen the front track width, and you aren't going to run big tires up front then you basically have a pretty car that isn't going to corner. The only other option is to spend a ton of money and turn your nose heavy Mopar into a lightweight race car with 50/50 weight distribution.

If you can achieve something close to 50/50 then your options open up. If you can move a little more weight to the rear and have a 55/45 car then you can have a pretty car that does corner. It might look like a 911 but it will have bigs and littles and will transfer weight to the rear on take off and maintain good weight distribution under heavy braking. Basically the perfect setup which is why the 911 has been killing it on the racetrack for many, many years.

The theory is one thing but a '72 Duster with a cast iron engine and stock suspension is such a horrible pig in terms of handling that theory doesn't apply.

If you want to learn handling theory then start with something nice and simple and square like an entry class pavement car. Something with a 50/50 weight distribution, rigid chassis and tuneable suspension at each corner. Then you can see what happens as you adjust spring rates, roll rates and shock valving.

A stock '72 Duster is a disaster and you have to stop the bleeding before you can even start to think about good handling. You need to add a bunch of roll stiffness just to keep the tires from folding over turning a corner. The stock tires are way too small for handling and the spring rate is way too soft. The weight distribution is horrible so it will never handle very good unless you spend money to take weight off the front.

So I'm not sure what answer you're looking for. If you want theory then there are plenty of books. If you want to make your Duster handle okay (like a new Honda Accord or something in that league) then ignore the theory and just call Firm Feel and do what they tell you do. If you want a daily driver where you can apply handling theory then sell the Duster and buy a used Z06 or something along those lines.

Is that graph posted earlier copied out of your copyrighted book, without your express written permission, and if you don't take prudent action to prevent this type of ,violation you risk it becoming public domain, and you know all this as lawyer, if I am correct? I know you are nice guy, but fair is fair.

If there is an exception, I'd like to know it, because there plenty of stuff I'd like to put up.

Thank you for clearing that up, thought it was posted originally without author notation, but the copyright issue still exists.

Get a grip JCC

Look up fair use exception

[censored] eyeroll

Doesn't apply here, IMO, maybe it grips yours.

Clear enough?

Dude, just click "ignore this user", you won't miss much.

Thanks, I will.

Thanks everyone. I have some reading to do!
Jeff

Jj, where are you located in the left coast/prc?

I don't have an issue if people post stuff from my books as long as they say where it came from.

There was one person on here who claimed some of my pictures as his own and I got on him about that but that was the only problem I've had.

Ironic how NONE of the above requirements were met in this case, it was not transformative, it was not "commented upon", it was not "criticized". It appeared to me the main end result here, intended or not, was the post precluded the OP from purchasing the quoted authors book and getting his desired information from authors copyrighted graph, rather then a generic formula, IMO.

Ostrich's don't miss much either, you can quote me on that.

Tony and AndyF, this is not about you guys.

I would disagree that none of those conditions were met. One could say the posting of the graph transformed the conversation from one of "go read this" to a graphic example of what needs to be achieved. Part of commentary in the discussion to direct the OP to a document that would assist in his goals. Removing the graph and additional commentary, we are left with the references below to Mike Martin's work without divulging the entirety of the chapter contents or methods, thus providing limited purpose. As such, the OP will need to do additional research on the topic, which may lead them to a search for this or any of the other books mentioned in the thread as the OP says he now has some reading to do. So I'd say we fell within fair use guide lines as described in the definition above.

jcc-

The information presented in Mike Martin's book (the theoretical handling line and formulas) was originally presented in one of the Mopar Oval Track books, which I believe is now out of print.

I'll let ya'll decide how that plays into the copyright discussion.

My main issue with the theoretical handling line is the formula is never clearly stated in either book. And working backwards from some of the suggested setups in the oval track book doesn't give consistent results. My main issue is if the wheel rates of springs are truly supposed to be doubled. No one spells it out in either book.

Regardless, the theory is sound.

But changing roll centers, etc, would change the line. And the original line was developed for circle track mopar B bodies if I remember correctly. So maybe the line is a little different for an A body.

So really, if one wants to go all the way, you need a complete chassis setup program (or to go hire someone like Ron Sutton to do it for you).

The formula as I've used it:

Percentage of Front Roll Bias: Front Roll Couple / (Front Roll Couple + Rear Roll Couple)

Front Roll Couple = (Torsion bar wheel rate * 2) + rear sway bar wheel rate
Rear Roll Couple = (Leaf Spring wheel rate * 2) + rear sway bar wheel rate

Torsion Bar wheel rate = listed rate
Front Sway Bar wheel rate Opt 1 = rate as quoted from Hellwig or Hotchkis
Front Sway Bar wheel rate Opt 2 = sway bar rate * (Motion Ratio^2)
Motion Ratio: (sway bar attachment point from pivot point) / (ball joint from pivot point)
Leaf spring wheel rate = (listed rate) * ((leaf spring width / track width)^2) * 2
Rear Sway Bar wheel rate Opt 1 = *not sure if Hellwig & Hotchkis quoted numbers are wheel rate or sway bar rate - will need to measure and calculate old way*
Rear Sway Bar wheel rate Opt 2 = sway bar rate * (sway bar end link width / track width)^2

solid sway bar rate = 500,000D^4 / (0.4244*A^2*B + 0.2264*C^3)
A = Main bar centerline to end link - 90* (think ~8.5")
B = Main bar width (think ~30")
C = Main bar centerline to end link - at whatever angle the bar is (think ~10")
D = Main bar diameter (think 1.25)
*if bar is hollow, calculate rate for solid of that size, then subtract rate of a bar that would be the diameter of the hollow diameter
*sway bar rate formula from the Fred Puhn book - How To Make Your Car Handle.

Further credit to TC@HP2 - we've had a few email conversations about the theoretical handling line. There was even some discussion of making an excel for everyone to use, but we never did. Maybe TC@HP2 will still do that?


If you're serious about doing some of the number crunching on your own, you should look at getting the following:
-How to Make Your Car Handle - Fred Puhn
-How to Build the Ultimate Super Street Mopar - Mike Martin
-Mopar Oval Track book - I think the Mopar Oval Track Modifications book (available at Mancini)


Still, at the end of the day, the above formulas will only get you close. Be prepared to fine tune the combo (i.e. - buy more springs, sway bars, etc). If you work with Ron Sutton or one of the newer suspension setup programs that requires more inputs, then you can probably nail setup from the start. Those programs or Ron will require measurements every attachment point in your suspension.

I have read the Fred Puhn book, but wasn't aware of the Mopar Oval track book. I'd expect it to be biased to left turn-only cars though, right?

I understand that the Op didn't want an answer so much as a few suggestions on how to figure it out for himself. That is impressive to see...most people just want to copy from someone else since it is much easier.

Yes, the neutral handling line was compiled by Mopar engineers in the early 70s and was used in developing the Kit Car program through the late '70s. It was copied and expanded by Mike Martin into his book in '84. The Martin books are out of print now and can only be found at swap meets and Amazon. The Mopar Oval Track book is still publish, and like Mopar's Engine or Chassis books, contains a lot of good data. However, while some of it is becoming dated, some is still relevant, and you have to understand how to sort it out.

I thought Martin provided a good basic of roll couple calculations. The Mopar books, not so much on the calculations but they do provide piles of other data. However, I have only read Martin's 1984 book "Mopar Suspensions" and not his 1988 book "Ultimate Super Street Mopar" so I do not know if his last book was repackaging of the first, all new material, or a combo of the two. In the first book, Martin did not get into the Nth degree details of adjusting rates by motion ratios. He simply stated overalls, which, if that's all you used, could be a more simple way to arrive at a number for a relevant starting point, but would definitely shift the neutral line one way or the other from what Martin published.

Martin does point out right on that page posted above that there are many contributing factors that will impact the relative position of the neutral line. Even Mopar said that that the line can have a wide range of variables but it can be used to create the fastest route to a starting position of parts that handle reasonably well compared to simple trial and error.

While the original observed data was all B body stock cars at Atlanta, verification of the data was done across multiple tracks, including road courses, and applied to the kit Car program which had a several different wheel base lengths and weight breaks it was involved in. So the theory is sound and is applicable to any chassis regardless of wheelbase, weight, or application.

Uhcoog1 your format right above does seem to cover everything necessary to arrive at some good numbers with the most precise calculations. Like you, I've used multiple books to understand the variables and develop the formulas I have. From my Chevy and Ford days, I have an extensive collection of Steve Smith Racing books that have developed many similar methods that seem to have picked up where Mopar and Martin left off.

The thing about geometry in a stock based suspension is that it applies to both left and right turning cars. The transition to an oval track car is accomplished by weight bias, wheel offsets, spring rates, and shock rates. Typically you want the suspension motion to be symetrical from side to side to avoid radical unsettling and create predictable handling should the car get unweighted or turned around, which happens often. So it is very easy to apply oval track theory it to a street car that turns left and right. All you do is equalize everything left and right.

But even with a formula, you have some homework to do and further tuning will be required. Or, like Uhcoog1 says, you contact Ron Sutton and he does all this work for you, on your car, with your preferences in mind. Or spend a few hundred $$ on a software app that you can plug all the variables in to.

This is a great thread, but it looks like there is a typo in the formula. I'm not real sure, but it looks odd to me, so I thought I'd ask.

In the Front Roll Couple formula, it says to add the REAR sway bar wheel rate.
That's supposed to be, add the FRONT sway bar wheel rate, correct?


Tav

Nope, makes sense to me...

So then the front sway bar rate doesn't affect the front roll couple?

Still seems odd.

He included a formula on how to calculate the front sway bar rate, but then didn't use that front bar rate in either roll couple formula.

You are correct, that is a typo. Front roll couple should be front springs and front sway bar, rear roll couple is rear springs and rear sway bar.

THANKS!
Previously I had just duplicated the stock setups, but it's become apparent that there's room for improvement, even without drastically changing the cars original appearance.
I've been studying up, and this thread has been quite helpful.

Best I could determine was that there was no single Neutral line for all situations, but rather a line that could be established and used for guidance. I summarized this here http://www.heritech.org/cuda/mgcudah.html regarding my own car. If interested in the spreadsheet contact me off-list.

Thanks you, I enjoyed reading that.

OK, I have another question about one of the formulas.
"Front Roll Couple = (Torsion bar wheel rate * 2) + front sway bar wheel rate"

I see that it's torsion bar wheel rate times 2.
I imagine that's because there are two torsion bars & two wheels.
That makes sense.

I see that it's front sway bar wheel rate.
That's the sway bar rate for a single front wheel.

Since there are two front wheels, should it be front sway bar wheel rate time 2?

In some respects, you can go crazy trying to make things better, by observation. NOT that they can't be improved upon.

As for the A-body handling, check out "The Green Brick" which "Mopar Action" built and road raced with great success. Not sure were it might be online, but it's an interesting read! Especially when they went to Road Atlanta!

To get an idea of Chrysler suspension geometry, there are some factory training manuals online at either OnlineImperialClub or MyMopar.com. In that book, you'll understand that Chrysler engineers knew about "controlling the contact patch", even back in the 1950s. Plus, an explanation of camber angles in cornering!

As for spring rates, f/r, there were some excellent articles in the old "Car Life" magazine, circa 1966. About how the f/r stiffness balance affects how the car bounces and comes to "flat" after a road disturbance. This orientation keeps the car more level and comfortable in general driving, such that one end doesn't bounce more than the other, with good shock absorbers.

There was a graphic of a car encountering a dip and how soon the car would be level again. "Number of bounces" at each end. This relates to spring AND shock absorber issues . . . AND can vary with speed.

Back when I was driving the '66 Newport all the time, with a good set of HD shocks (first the Chrysler "MAECO" stamped factory HD shocks and later a pair of Gabriel Strider adjustable shocks on the front), the car was flat and stable "at speed" and pleasantly firm otherwise. The observed natural cruising speed on the highway was the 75-90mph range. Faster, it felt "busy", slower it felt "bored".

Until you spend "road time" with a torsion bar Chrysler product, you cant really feel how good they are, by observation. I mean on a trip of several hours, not just going across town. Keying into the engine response/throttle input, the way you don't have to slow down for corners, the confidence of the 11x3" drum brakes, and other aspects of the driving experience.

Go into www.allpar.com and read the sections of the Chrysler "squads" and LAPD testing. Quite outstanding, even in the later 1950s!

Chrysler didn't like to use a rear sway bar for many years. They chose to use stiffer rear leaf springs to help with that rear roll stiffness issue. But in the "Police and Taxi" section of the 1970 parts book, there is a listing for a rear sway bar with the note "LAPD". Not sure if they co-opted the Helwig rear sway bar (aftermarket).

The first rear sway bars were on the radial tire cars circa 1974, B & C body in particular.

Pontiac used rear bars on their GTOs and such first, then Chevy came out with their F-41 package for many cars, which helped handling a good bit.

In a 1969 issue of "Car Life" magazine, they did an article on "Power Cars", which were full-size cars with larger engines. In that time frame, they had a test corner where they had a camera set up to picture the front suspension and tire/pavement interface. It clearly showed how the Chrysler products had superior front suspension geometry to the other cars, which greatly aided their cornering performance.

As the MasterTech book shows, the outside wheel goes into negative camber such that the outside tire is better braced to take the added load. The inside tire goes into positive camber, to help with the cornering load, too. GM and Ford geometries kept the tire perpendicular to the body, so that as the car leaned into the corner, the tire was not perpendicular to the road surface, resulting in greater outer tire wear and decreased cornering performance. ONE reason the rear bar and less lean made so much difference in their handling!

What has not been mentioned here is tire pressure bias f/r. Factories generally wanted an understeering car, so the front tire pressures are generally lower than the rear (especially in the C-body wagons! and when "full load" is planned for.

What I found, going back to the tire's load and excess load potential at a particular tire pressure . . . one of the "Car Life articles on tires had a chart of tire pressure and tire load by tire size (in 2psi increments between 24psi and 32psi, the pressure ranges back then). On a Chrysler C-body (or others) with a 55/45 f/r weight distribution, here's the strategy I prototyped and found to work well for me . . . proceed at your own risk.

The minimum tire pressure for "high speed driving" (in the owner's manual) was 4psi over "normal". That meant 28psi in the tires. Adding in the correction factor for the front being heavier, by the pressure/weight carrying/tire size chart, adding 2psi to the front balances out the load which each tire sees, proportionally. For a 60/40 split, a 4psi front bias yields the same situation. Key thing is that EACH tire had the same static percentage load it supports. That would mean that each tire should "slip" at the same time in a corner, as a car with a 50/50 f/r weight situation. i.e., "neutral" handling, rather than the front sliding first (understeer.

Understeer, for the masses, is deemed "safer" as you'll hear tire noise before you get into trouble and, as a result, slow down to a safer cornering speed. In a Mercury Marquis I rented in '97, that speed was 25mph in a city setting with Michelin Symmetry radials. Any Valiant would out-handle that car!

One thing on many Chrysler suspensions. When factory a/c was ordered, the front torsion bars were upgraded to the HD bars due to the added extra weight. Just the bars, not the shocks. The full HD suspension added the shocks and heavier rear springs/shocks to the mix. That's the way it was on the C-body cars. Check the respective service manual for specs.

Using the f/r tire pressure bias I mentioned also tends to result in greater tire life. Reason is that EACH tire tread is flat against the road surface. No wearing on the edges from too-low air pressure, no wearing in the center from too much air pressure.

One other thing you can do to check the contact patch is to be on a dry concrete surface and with the car parked, turn the wheels from lock to lock. Then back the car up a foot or so and check the resultant rubber marks left on the concrete. With the factory alignment settings, with the 2+psi pressure bias in the front, it should be a solid patch.

With the wider tires and rim width of more modern times, such a solid patch might not be possible, but with the 5.5 and 6.0 inch wide wheels of the 1960s, it worked fine for me.

From the luck which "Mopar Action" had with The Green Brick Valiant 2-dr sedan on the road course, it proves that a stock Chrysler suspension can be easily upgraded to handle very well with few modifications. But then, this was in the 1980s, well before the hyper-touring orientation came along.

KEY thing is chassis/body stiffness. Chrysler's UniBody was a better "unit body" than GM knew how to build, back then. Ford too. This stiffness helped give the suspension a stable place to work.

The assymetrical rear leaf spring configuration helps in acceleration performance, as did the r/l spring stiffness bias on the HP cars of the '60s.

And then there are shock absorbers! Of which we have fewer options than in the 1970s! And then there are some new options, too.

One of the things which, to me, made Chrysler products such great cars was the "Torsion Bar Feel" (which encompasses the whole gamut of their suspension design orientations). It was the greater high speed handling which made Chrysler products the desired vehicles of so many state police and other law enforcement agencies. Check out the "It's a Wild, Wild, Wild World" video and watch the several Chrysler products in the "racing" scenes. Watch the suspension dynamics at work on those cars, even a '62 Imperial convertible.

Also, find the "On The Test Track with the 1957s" video on YouTube where 1957 Chrysler cars were compared to "the competition" from GM. It's easy to see how much better the Chryslers were! And how much flex were in some of the GM body/frame cars in an unexpected encounter with a train track crossing in the country.

To me, when you start putting new k-frames and even frames under a Chrysler product, with the lowered ride height and 8-10" wide wheels, to me the great chassis feel of the stock Chrysler chassis is lost. Be that as it may, but I suspect lots of money can be spent for a very specific gain in a specific application. You can spend money, get "the look" and performance increases, which leaves the only reason to have a Chrysler product to be its styling rather than otherwise . . . for which a Camaro would do the same thing.

Sorry for the length, but these are the things I've researched and pondered over for a good while, but I still believe that what the factory provided is better than some might suspect or realize. Others might not agree, which I respect.

CBODY67

THanks for the detailed post.I appreciated it.

Yes, thanks for sharing all that info!

Front Sway Bar wheel rate Opt 2 = sway bar rate * (Motion Ratio^2)
Leaf spring wheel rate = (listed rate) * (Motion Ratio^2) * 2


These two formulas are both supposed to determine the wheel rate.
They both have spring rate times motion ratio squared.
Yet one of the formulas has a x2, and the other doesn't.
Why aren't these formulas the same if they're doing the same thing?

It seems like the *2 at the end of the Leaf Spring Wheel Rate Formula shouldn't be there.

Using 70 B body as an example, distance between leaf spring perches is 44 inches, divided by track width of 59.2 inches = .7432432 motion ratio.
that motion ratio squared is .5524104
multiplied by a 160 pound spring rate = 88.38566
then when you multiply that by 2, you get 176.77132
It seems odd that the wheel rate would be higher than the spring rate.

Couple things going on here.

First quote; No, you would not double the sway bar rate calculations because the sway bar works over its whole length. The sway bar interacts between both wheels so its motion ratio calculated rate will link the two wheels. Its not isolated to a single wheel.

Second quote; Yes, they are both calculating wheel rate, but they are two different components, so the rate is achieved in different calculations. In this case, the spring must be doubled, because the original formula was to determine total roll couple, not isolated spring rate. Since you clipped it out of context of the rest of the formulas, the *2 appears to double the spring rate for a single spring. Rather it is doubling the spring rate in the overall picture of the whole rear roll rate of two springs and a single anti-roll bar. So, you have 176.77# for the two leaf spring rates. You now need motion ratio adjusted sway bar rates to know the total rear roll couple rate.

Thanks, that makes perfect sense.
I wasn't sure, and thought I'd ask.

Your comments here and in other threads have been extremely helpful towards my understanding of this stuff.
I really appreciate it, THANK YOU!

I'm pretty sure that I understand what you're saying.


Just to be clear though (which may also help anyone else reading along), Here is the Rear Roll Couple formula as it was presented above.



It makes perfect sense. There are two leaf springs, so you multiply the spring rate by 2.
I'm not using a rear sway bar on the car (yet), so I don't have to worry about that part of the formula.
I do need to calculate a leaf spring wheel rate though, so I look for the formula, and it's presented as...



I recognize that the middle portion "(leaf spring width / track width)" is the motion ratio of the leaf springs.

Using 70 B body as an example, distance between leaf spring perches is 44 inches, divided by track width of 59.2 inches = .7432432 motion ratio.
that motion ratio squared is .5524104
multiplied by a 160 pound spring rate = 88.38566

I believe you're saying that the "Leaf Spring Wheel Rate" formula should Not include the "*2" on the end.

Also, that the leaf spring wheel rate does get multiplied by 2 in the "Rear Roll Couple" formula.

And that multiplying the leaf spring wheel rate by 2 in Both formulas (as it's shown) would result in an incorrect answer.

You've got it!