I am trying to populate a spreadsheet to help calculate some anti-roll bar settings, roll rates, roll gradient, and other fun facts. If anyone has measured or knows the following values for the E36 M3/325/328 post them here, once I get the sheet filled I will post it for everyone.
Sprung Mass per corner (rear)=?
Sprung Mass per corner (front)=?
Track width (front)=?
Track width (rear)=?
Wheelbase=?
CG height=?
Front Motion Ratio=?
Rear Motion Ratio=?

I think thats it for now. Thanks guys!

What type of roll centre analysis are you using?

Are you looking for stock numbers, or common aftermarket setups, or full race setups?

There are a bunch of companies (rrt, etc) that have this info already and use it for their suspension design.

You might even want to try google researching a bit, I know I have the data but I don't recall if some of it was gathered from other sources than actual measurements.

As an example of whats already on the internet, I just did a quick google and found this (not sure on its precision/accuracy, but there are many sites that have lot so info is the point:)).

http://www.steerbythrottle.com/hccyong/cars/susp_analysis.htm

http://www.steerbythrottle.com/hccyong/cars/suspension_pics/Motion%20Ratios.jpg

I am using "Racing By the Numbers" by Bill Mitchell and OptimumK Software and when all else fails pen and paper.

I am looking for the ideal aftermarket setup. (atleast that is the plan)

I will contact RRT and do some additional searching, Thanks!

Yea, I figured by the title you were using Mitchell (I think I still have Wingeo3 somewhere).

When you say "ideal aftermarket setup" what are you referring to?

Ideal for who, what application?

Have you read Milliken?

If you want a challenge, try setting up Bosch Lapsim. Its a cool piece of [free] software, but you need a lot of (read that as ALL) data to complete any worthy simulations.

Just curious, did purchase OptimumK, or are you using a free trial. $1700 is a bit expensive for personal/small team use. I'd like to try that software out sometime, is the free trial decent? I wonder if interfaces with non-motec data?

I am simply an autox-er so my focus will be different from others, BUT this could help most DIY guys. I look at it like this, the more I know the better, and no offense to anyone but I don't want to run X part on my car because Jon Doe said it was fast.

Sorry about the etc, I initially thought that was an abreviation for a company. I deleted that from my post.

Milliken is probably the Bible when it comes to racing, yes I read it a couple times.

Gotcha.

Any engineering background? Milliken can be a bit tough to grasp properly without some formal education, although, its definitely feasible.

I definitely agree, you should not listen to Mr. Doe :) However, there are companies that actually engineer their products, and do a lot of in depth analysis and track testing to provide a product that is in fact Fast, compliant, reliable, and in most cases affordable :)

What class do you auto-x in? Whats your budget for suspension?

I'd say you are going to have three main factors here. Compliance, setup, and components (i.e. having a good components asopposed to a pos). You have to start off with a quality component (Ohlins, Sachs, Moton). Take care of all the compliance issues so you know the geometry/setup you are designing (assuming its allowed on your class) actually stay within the spec'syou are designing to.

In most cases, besides Mod style classes, you will end up with having setup as the biggest factor, and that really will come down to tire setup and on track testing. Using software to develop suspension is limited when all you are really doing is changing camber and toe, and fixing compliance problems that come from the factory.

Now in a case of many mod cars where you are moving pickup points, changing roll centers, etc...it definitely can be very beneficial.

My copy was purchased for our team at a discount ($395).

Motec or ADL Question: Yes it allows you to import track data from any data acquisition system or CSV file.

You can get a really good idea of how powerfull this software is by downloading the demo:
http://www.optimumg.com/OptimumK%20Demo.zip
http://www.optimumg.com/letterpics/data2.jpg

Oh, FSAE discount for Ohio state?

The interface looks cool, and the migrations are nice, how fast is the software iterations?

(sorry for the OT)

ON topic though, you might want to talk to RRT about working on a project with them to further analyze e36/e46/e90 suspension. I have not seen any other company go into the depth they have with suspension design for these chassis. Even back in 05 when we were just starting the program, we did some interesting testing, like the exact affects each bump stop had on shock/spring force at different displacements. RRT now has a lot more resources available, including an in house shock dyno.

Steve, I can hardly keep up with your post speed! :-)

Yes, I have egr background. It's definately not an "easy read" by any means.

I am in the process of building the car, ETA one month. I started building for STS but went to some local races and watched some M3 get spanked by EVO/WRX cars, I am not saying we are not competitive, but that is what I saw. Budget is TBD, as I am straight out of college and still settling in.

The car may now end up in either Prepared or Mod, I don't know because I am still thinking about my direction. Either way, even if I don't take home any First Places, I will have the car built my way. And after all this is just a hobby (addiction).

Yes FSAE, not Ohio State, just a little bit north. Software is surprisingly quick, probably depends on your computer hardware too.

I will shoot them an email, Thanks Steve!

Well, back on topic, what info are you looking to gain from this analysis?

Are you trying to find the numerical value for whats happening on the stock suspension/chassis?

Hopefully you see what I am getting at, not trying to avoid the questions, just help get you to your goal.

Basically I would eventually like to find the ideal settings for anything that is adjustable on your typical E36 aftermarket setup, and maybe I will find that the ideal parts are not "Off the Shelf" parts, who knows.

Ok, then back to my previous question, what are you basiing your "ideal" data on? What is it ideal for?

Your "typical" e36 setup will really only be changing ride height, spring rate, toe and camber (and possibly caster).

You have to assume there are no major compliance issues though, which is not realistic for the standard aftermarket application.

You also have to factor in the components you are using, in respect to their operation. For example if they allow for proper droop and compression travel.

Fortunately, for the average street/track application BMW did a fantastic job with the chassis/suspension design. The real trick is getting the setup right for the tires and driver...and thats going to come from 99% track testing. You can also use a skid pad for some preliminary spring design under static conditions.

You also have to determine if the setup will be for autox or track use, as they will each benefit from setups specifically for each application.

Compliances (i.e. bushings) are very important on these cars (and all actually), and so are non-linear stiffnesses inherent in the designs. Good luck getting useful output of models without really doing your homework. We have some secret weapons, but they'll remain secret for now.

I'll chime in and say Milliken is a great start but it's a little weak in some areas. There are other techniques and approaches out there...

My car will will only run Autocross events. On top of what you listed, I would like to play with ackerman, wt distribution, chassis stiffness, and roll centers.

You said it "The right setup will come from track testing," true BUT I have not yet found the money tree, and I believe that I can calculate some values before purchase, for example spring rates.

Chris, I am not understanding your post "without doing your homework?" I am I not doing my homework now? I already stated I have read most popular books, and have access to all the top software.

Secret Weapons- Although I respect your views on keeping your technology secret, I do not believe there are any secret weapons in vehicle dynamics, sorry.

Have fun getting tire Data :)

Regardless of all the fancy calculations and pretty graphics, you will end up needing to do real life physical track testing, no way out of it.

To vary your roll centers you will have to do a bunch of modifications, and to have a setup with variable roll centers (like what I am doing for my e46 rear end) you'll need to decide if you want set roll enters or infinitely adjustable (think panhard bar mounting style).

Unfortunately, even with FSAE resources, you will not have the time/finances to get usable data, and track testing will be your best bet, especially if you have the big variables like different track aggregate types, and tires.

The secret is setup, period.

How much modification are you planning on doing? To do what you are suggesting will require your car to be a fairly modified MOD car.

The Secret weapons in vehicles dynamics are called experience setting up cars and interfacing with drivers.

The car itself is nothing without a fast driver. A fast driver can make slow cars "fast." When dealing with Amatuer/clubracers, 95% of the time the car's limits are way beyond the drivers. You can chose to adapt the car to the driver, or go with experience and a setup that data shows works and train the driver to take advantage of the car's speed.

If your car is only running Autox events, then you should be on the skid pad, and on it often. Go setup a slalom and skidpad and spend a week there testing various setups. Get data, and then start plugging away.

Steve, I noticed your adjustable rear end, I like it. If changing the roll center (above/below ground, or migration) is necessary to make this chassis fast, then so be it.

Tire data is "black magic" but I do know people who have built very simple testing rigs that attach basically to trailer hitches that achieve the data rather effectively.

Your chassis setup will always require fine tuning at the track or tuning to changing track conditions. But there is no secret to this. Fast drivers will always be fast and all things equal a well prepared car will beat a poorly prepared car.

Well, "fast" is relative. I'd bet you my car what Joe schmoe club racer says is slow, a pro driver can make much faster simply by adapting to the car and applying his skill.

Tire Data these days is a lot less black magic, and a lot more knowing how to deal with tire testing. Many of the teams that are top 10 in FSAE do their own tire testing, to an extent. In 2005 (my last year on fsae) we were in the top 15 at FSAE MI, and we had zero tire data, it was mostly a very driveable powerband and a lot of setup time, with some decent drivers.

My adjustable rear end has 3 available roll centers other than the stock. I did two lower roll centers, and one higher. The data I used actually came direct from BMW Motorsports tuning guide for WTCC E46 cars, just adjusted slightly for my exact application. I actually decided against doing a full tube frame rear end, instead i'm going to modify the subframe...this was due to the risk factor of damaging the tube frame structure. Much easier to swap in a new rear subframe then totally refab the rear of the car.

I think you'll probably find you'll be fastest by using a fairly stock geometry, sticky tires and a lot of work on bushings/bearings. Getting the balance will be key, and thats something the computer can not usually determine without actual driver/test data.

Thanks for the advice, but this discussion has gone off topic.

Maybe a "fairly stock geometry, sticky tires and a lot of work on bushings/bearings" will be fast, but I guess I will find out for my self. This thread was simply intented to gather some facts with hopes of returning some usable info to the group.

Yep, sorry for the OT, but again, just trying to help get you to your goal :)

The links I sent should give you most of the data. What you are trying to do is nothing new, and the people who are interested pretty much have the info already. The companies who develop suspension have the info and use it to develop their components. Its not cheap to do the testing and development, and this is why the info is not publicly available.

Edit: Since we only talked amongst ourselves, i'd be happy to delete all my replies to clean up your thread.

Its cool Steve, leave the replies. I will get as much info as I can compiled into one document or thread.

I am in the process of building the car, ETA one month. I started building for STS but went to some local races and watched some M3 get spanked by EVO/WRX cars, I am not saying we are not competitive, but that is what I saw. Budget is TBD, as I am straight out of college and still settling in.


I'm probably not the norm here, but I think both the M3 and 325/328 can be quite competitive in ST if they're properly developed and well driven.

Even if you don't settle in to ST, it's probably a good "intermediate" zone on your way to a higher class since most of the standard suspension mods, bushing updates and other significant mods (headers / exhaust / tuning) are part of the rule set.

Mike

As an example of whats already on the internet, I just did a quick google and found this (not sure on its precision/accuracy, but there are many sites that have lot so info is the point:)).

http://www.steerbythrottle.com/hccyong/cars/susp_analysis.htm

http://www.steerbythrottle.com/hccyong/cars/suspension_pics/Motion%20Ratios.jpg

that's the website of a bf.c memmber (hccyong)

that's the website of a bf.c memmber (hccyong)


Yep. Harris Yong did quite a bit of suspension work/analysis on his '95 318is. Some of which I think went into his graduate work?

I've done motion ratio measurments on my '95 M3 as well. They differ slightly from the motion ratios Harris determined on his 318is.
I created a bunch of spreadsheets and combined some of Harris' data as well as Ron Katona's data on motion ratios.

PM me with your e-mail and I'll send them to you. Use at your own risk, sheets were not exactly built for mass consumption, only for me to play with. There is component mass (as published by BMW in the ETK) for each piece, motion ratio, corner weights and measured OE spring rates, calculated anti-roll bar stiffness changes with diameter, etc.

You can learn a ton from Gustave's site:
http://e30m3performance.com/index.html

It is based on the e30 M3, but much is still applicable to the e36.

Cheers,
Chet

Cheers,
Chet

Thanks Chet! PM sent.

Going back to your first post. What is the best way to measure your cg height? Put scales on a hill?

http://www.longacreracing.com/articles/art.asp?ARTID=22

Subscribed

Before I start machining... I have some plans plans for new tie rods for the front steering. I believe I can make adjustable (positive) ackerman without too much effort. My plan is to use a large spherical rod end for the outter joint and machine the rest. The adjustment will come from the distance that each rod end is threaded then double nutted or locked in some fashion. The amount of appropriate ackerman will be specific to the tires I test on, but I believe I could find some good ballpark numbers.

I would test this by simply setting up a skid pad with a radius close to the average radius of an autocross (?). Complete this left and right, then change the ackerman setting and repeat. Hopefully leaving everything else the same, I will see some sort of pattern through the lap times.

So back to my question, before I start machining, has anyone made their own steering tie rods to adjust ackerman?

And how are you changing the Ackerman %? Its a dynamic adjustment, adjusting that one static constraint will not control your Ackerman.

You will also want to look into relocating the inner tie rod mount, like BMW Motorsport does to properly correct the bump steer.

You should get an AIM system and get a bunch of sensors, rather than just laptime, as thats dependant on other factors.

My tie rods will move the outboard joint therefore adjusting the ackerman percent as a function of steer input.

I Thought about adjusting the rack for and aft which is the same affect, but that would be much more difficult. Bump steer in a vertical correction that is much more easily adjusted, (or corrected).

I will run the data acq when the time is right, need to acquire and mount more sensors.

Anyone already modified their tie rods?

Here is my E46 MS setup, inboard is raised to correct for lowered ride height geometry, and the outer height is adjustable via shims.

http://jaffster.com/E46M3/12-5-06/main/pc030001.jpg

http://jaffster.com/E46M3/12-5-06/main/pc030002.jpg

http://jaffster.com/E46M3/8-10-06/main/tierodmount.jpg

http://jaffster.com/E46M3/8-10-06/main/motorsportclevis.jpg

Great work steve, the inboard steer link (the bracket in the last picture) does that thread into the rack? It appears that your bracket relocates the link up vertically. Now if we could also adjust it fore and aft, hmmm I am going to think about this for a while.

Not a terribly tough design, but I can't take the credit. Its a factory BMW Motorsport part (thanks again Turner, he gave me a smokin deal on them).

Yes, there is a special plug thats threaded and mounts that adapter into the steering rack, it has a very large allen head.

I would not want to adjust it fore/aft, the geometry otherwise is very good.

The only thing I am going to do is reinforce the tie rod a little, as the BMW motorsport design is a bit on the "soft" side, and I know its been know to fail/buckle under light wheel contact...which is ok, but it a lot less likely to have wheel contact in CR than Pro racing, so I'll slightly strengthen it so I don't have to worry about it failing under any other conditions. Big picture though, i'd rather have the tie rod link fail then bust up everything else :)

Only problem with the components is they use VERY funky thread types, and are all different...prob was some intern engineer at BMW motorsport lol

Not a terribly tough design, but I can't take the credit. Its a factory BMW Motorsport part (thanks again Turner, he gave me a smokin deal on them).

Yes, there is a special plug thats threaded and mounts that adapter into the steering rack, it has a very large allen head.

I would not want to adjust it fore/aft, the geometry otherwise is very good.

The only thing I am going to do is reinforce the tie rod a little, as the BMW motorsport design is a bit on the "soft" side, and I know its been know to fail/buckle under light wheel contact...which is ok, but it a lot less likely to have wheel contact in CR than Pro racing, so I'll slightly strengthen it so I don't have to worry about it failing under any other conditions. Big picture though, i'd rather have the tie rod link fail then bust up everything else :)

Only problem with the components is they use VERY funky thread types, and are all different...prob was some intern engineer at BMW motorsport lol


Please put some safety washers on the outer tie rod ends! Fail a heim joint and you're toast.

Otherwise it looks great.

I did this non-adjustable mod to my '95 M3 to help with bump steer at my lowered ride height:
Swap a 97+ M3 front cross member into the '95 chassis. The rack location is moved vertical about 3/8" and it has factory installed x-brace mounting points in the forward location.
Cheers,
Chet

I guess it might be hard to tell in those pics, but those pictures are of the mockup while it rolls around...the car is a bare shell basically right now :)

I have the retaining washers in one of the fastener boxes. Thanks for the reminder though :)

If that ball/race came out of the rod end, it would still cause some havoc with the amount of play it causes, but definitely not as bad as if the wheel became totally free to move.

At about $50-60 a rod end though, they better last me atleast a couple weekends before they produce any play lol

The inner bracket has a locking tab so it positively locks into place on the steering rack.

http://jaffster.com/E46M3/8-9-06/main/p8080009.jpg

Here are the rest of the kits/spares, you can see how the stud is adjustable with shims:

http://jaffster.com/E46M3/8-9-06/main/p8080005.jpg