The Project Pad

I'm adding IMS to spots where I routinely hang out and post project updates. Here's a brief summary of what I'm running:

1990 Coupe Quattro
Drivetrain:
-034EFI IIc, 4bar map sensor, blahblahblah
-034 Stage 2 Wideband/EGT to talk to ECU over CAN-BUS
-3B Head
-3B Pistons
-3B Crank
-3B Intake Manifold
-Scat rods
-034 Coated main and rod bearings
-ARP Head and Main Studs
-Spec Alu flywheel and 3+ clutch
-Currently running a T3/T4 46 trim on an eBay log manifold with custom lines and intake and stuff
-Siemens 830cc Injectors, 45psi base pressure
-28x3? CXRacing Intercooler

Chassis:
-S2 front hubs, B5 A4 Rear Hubs
-D2 S8 rear brakes
-WIP 996 Turbo Front brakes, going to be somewhere between 321mm and 345mm depending on how fast I want to spend money on brakes and wheels
-17x8, 18x8.5 wheels, ET30-35. Have run 205/50r17 snow tires to 24/61r18 Michelin Slicks
-Superflex? poly control arm bushings
-034 Solid subframe mounts
-Forged/stamped arms front/rear respectively
-034 Chro-moly tie rods
-QRSport Upper strut mounts
-V8Q Torsen Rear Differential, FWD 20v 90 Axles

Interior/Exterior
-Cobra Suzuka GT Driver's seat, Racetech brackets, Sparco Sliders, custom mounts
-Corbeau FX1 passenger seat, Planted brackets, custom mounts
-Momo 350mm Steering wheel
-RS2 headlights and B4 hood
-WIP Reiger RS2 Front bumper w/ 993 Turn signals and brake ducts

For previous updates, see my Motorgeek thread: http://motorgeek.com/viewtopic.php?f=6&t=36387

Car is an all-season fuckabout car. Meant for ice racing, winter driving, summer cruising, and track days.

Sam

I read a very good post from Mance dated about two years ago talking about all of the spherical bearing failures he got with his solid mounts. Basically, he was saying that there are two types of spherical bearings he could use in the top mounts--steel races or PTFE lined ones. The steel ones overheated and would gall up, and the PTFE dents and loosens up. While it seems fundamentally wrong to use spherical bearings in this application, it seems not everyone has issues with the bearings pounding out. As far as I know, Petar never replaced them (and these were tight when I got them), and lots of guys on S2 Forums seem to be getting years of use out of them without issue. Mance's solution was to incorporate some type of cushioning, poly or rubber, which seemed to make them last indefinitely. I've contemplated adding a rubber cushion ring between the top of the mount and the strut tower, but that seems like it would create issues with the mount loosening up on the strut tower over time.

Anyways, that stuff is only of concern because I spent another weekend beating the shit out of the Coupe, and it seems to have faired very nicely again. The local Audi club held their last scheduled winter driving event of the year this past weekend. We got a fair amount of fresh snow and cold temperatures the week before, so I actually got off my ass to make sure I was set to go. The only real prep I had to do was to figure out a front bumper solution. I took the pieces from the last event out of the hatch, realized I only really had two more broken aprts than before, and did what I think is unbelieveable and put the damn thing back together again...this time with more duct tape and zip ties!



I mounted it up and went for a drive. It basically fell off three times on that drive, which was kind of a fail. The zip ties I was using to hold it on kept breaking. I replaced those zip ties with safety wire and I was all set! I put about 30 miles on the car on Friday night and Saturday, and really started to get a feel for what the new suspension mods and torsen can do. It's really amazing, the car is super predictable and easy to drive really, really hard (albeit on snow tires). On power it corners better while having a lot more forward bite...this diff made a forward bite improvement in an AWD car!? There are several common handling 'features' that seem to now be absent by having some kind of LSD on the car. There are a few new ones present, but I haven't found the flaw in them yet. The one characteristic of note is that in the lower gears, in power, when you're not REALLY trying to put the car into an outright slide, it feels like the back of the car steps sideways a little bit before it takes a set. It doesn't feel like it would loop the car or anything, but you can feel it and it needs to be accounted for. Still very happy I tried it.

Anyways, Sunday morning I headed out to Lake Cynthia with the Audi Club Glacier Lakes. It was a gorgeous day and the conditions on the lake were great. Fairly low snow banks, but lots of packed snow on the track and lots of grip. There were some pretty good sized bumps where the track crossed the layouts from before; I wish I would have raised the car up to get a bit more suspension travel for traversing them. I was definitely on the bumpstops and then some at a few places on each lap--there were some major POPs going on over the bigger ones, it sounded horrible inside the car. I also lost my boost control halfway through the day--I had to unplug the boost control valve to avoid the overboost, and now somehow my 13psi wastegate has me at a fairly consistent 20psi. I also sat on the rev limiter too long and an epic backfire blew the rear section of the exhaust out of its slip joint. The exhaust note after was loud and awesome, but the drive home was brutal...if I could tame the drone I would love to run an exhaust that loud. Other than that, it drove awesome and was very fast all day long.



I need to realign it since the balljoints seem to have slipped a bit, but now it's just a matter of fortifying my bank account and installing things to get ready for track season this year. I need to finish the seat project (all set to go except for some steel), buy a helmet, do the brakes, and do something legit for the bumper. There are some smaller projects, but they're pointless improvements if I can't really use them like they're supposed to be used.

Sam

I am so amped on your review of the torsen diff Sam. I am excited to get my URQ running with one out back at about the same power levels as you. Any annoyances with the open diff in 0% traction apps?

I haven't had any problems with my GC mounts up front, but my comp mounts in back have been noisy since day one. I am about to go back to track density mounts on the struts back there.

Hank

I went racing on a frozen lake and never once had an issue with 0% traction. Figure at that point you're no worse off than you were with the stock diff, and you still have the center to do something.

One thing I did notice, while on a relatively slippery surface TRYING to get the car sideways to go around corners faster, is that that line between understeer and power oversteer is much finer with the torsen. I haven't actually gotten it to do anything really unfavorable like this on pavement yet, but it did trip me up a few times once it started getting slipperier.

Do the GC top mounts use a spherical bearing for load and rotation, or do they do it the smart way and put the spring perch loads through a roller or ball bearing and then use the spherical for the strut shaft? I was really thinking about going with the 2B 'Revolution' stock placement camber mounts, but it seems like that type of mount is just doomed for failure with this type of use.

Edit: They do

The Ground Control camber plates feature two sets of bearings, a proprietary bearing, made to our specs by Aurora in the USA, which allows for articulation of the shock shaft. This spherical bearing is very understressed, as the Ground Control design does NOT support the weight of the car on the spherical bearing. Instead, the weight of the car is supported by a completely separate articulating needle bearing, which is also manufactured for Ground Control in the USA.

Everything said though, after beating the SHIT out of my QRS mounts, the damn things were quiet and don't seem to have any play in them even after this weekend's event.

Sam

I'm also a little freaked out about rear diff fluid temps with the torsen, on the track, with this much power. The trans and diff temps even with the stock diffs are probably fucked with this much power, but with only 1.5L fluid in the back and no real organized cooling, tracking temps and setting up some kind of basic cooler this summer is on that big list of mine. 75w140 lube is supposed to help as well in these types of situations.

Sam

Yeah, I was thinking a diff cooler in my setup. I'll log it on the AIM dash and see.

Some pictures from last weekend, courtesy Peter Lapinski.









To quote Ice Cube, it was a good day.

Sam

Looks like a lot of fun out there on the ice. One thing winter is good for.

Nice project! I need me a coupe someday.

Those are some awesome shots, especially that first one!

nice! some cool shots...looks like fun. snow bank took out the bumper?

I pulled the bumper off of the car before I did any competing. It's in really sad shape, I didn't need to tempt fate obliterating it for the fourth time this winter LOL.

Sam

So, when I switched over to the A1 racing upper spring perches in front, I gained about .5-.75" of ride height due to the additional height of the perch versus the Ground Control ones I was using prior. I'm using what I'm fairly sure are 8" springs up front (they're at about 6.25" compressed length at ride height, which seems about right for an 8" spring), and the perches are basically bottomed out on the collars because of the steering arms. I only have about a 1.5"-2.0" of suspension travel (depending on how much the bumpstop ends up compressing), which looks like it would be ok for a 6" spring without dealing with coilbind.

The problem is that no one makes a 6", 400lb/in spring that I see--they're all 450lb/in+. To be honest, I wouldn't mind going 500lb/in+, but I'm running Konis and they don't seem to do well beyond 400lb/in. I don't know of any good options between these and some KW V3s, but $2k isn't exactly in the budget. Anyways, that leaves me with 7", which should help this issue but I would like to run the car lower than I think this would get me. I may be ok if I end up going nuts with the GC camber plates and custom strut towers, I guess.

Sam

LOL I missed this before, EPIC bumper obliteration!!

i went ice racing this past sunday for the first time in my coupe.

i obliterated my s2 bumper :(

did you ice race before you had the torsen diff installed? i am curious if locking the rear diff would of helped with breaking traction around the corners. I had issues making the car break into a slide, how much e-braking did you do? wish i had a go-pro to show how badly i did lol.

Holy crap, you put that /\ back together?! ahaha nice! I've only ice raced a dirt bike with 'ice screws', great fun! Did you two (cory and sam) have studded tires? It doesn't look it (hard to tell though), and you'll have so much more fun with the traction gain

I run General Altimax Arctics; 205/50r17 without studs. They're great snow tires; very cheap and good quality. Fantastic grip and they last a long time. Any snow tire is not going to have great grip on a glare ice surface, but often the course they plow out has a good base of very compacted/dense snow that gives quite a lot of grip, all things considered. Studs would be ideal of course, but I haven't come across a good set of tires to put screws in yet. I tried really hard for a week to acquire some Michelin studded rally tires, but couldn't find any on this continent. Those would be the titties.

The torsen feels similar to the locker on power in terms of forward bite, but it doesn't have the AWFUL turn-in the locker has on moderately grippy surfaces.

Sam

April fool's update!

For the past month or two I've been putting a lot of thought into suspension setup of the car, both to gain some experience at analyzing vehicle dynamics and to try and make sure that what I'm doing has a chance of actually working when put through its paces at the track when using sticky tires. I realized last winter after spending all this money and effort getting various solid/performance pieces into the suspension that all of that precision handling is useless if the car does weird things in the middle of corners. Thankfully, drivetrain effects aside, you can't ask for a car much simpler than a B3 Audi, with McPherson struts front and rear and a lot of the geometry the same between the two. Armed with a full shop including an alignment rack and corner scales, I've made a lot of measurements and am starting to get some interesting observations.

My first move was to go to shorter front springs. This was kind of a half-baked idea, but a lot of good has come from it. When I installed my QRS front top mounts, I had to move to A1 Racing upper spring perches, which added .75" of ride height to the front of the car when my front spring perches were already basically bottomed out. I had 8" 400lb/in springs installed, and wanted to go down to 6". Noone makes 6" springs that soft because they're liable to coilbind, so I settled on 7" which turned out to be a good idea. I got all set up to install them and take a bunch of measurements. Before I started, I weighed it:


With the car still on the scales, I also raised it up until all four springs were unseated from their perches, leaving only the unsprung assembly resting on the scales. The results, including G60s in front, D2 S8 rear brakes, 17x7 cast wheels with 205/50r17 snow tires, 034 Chro-mo tie rods, forged front arms, and sold rears was 110lb/corner front weight, and 95lb/corner rear weight.

I also measured/verified motion ratio at about .98. This means that for every 1 unit the wheel moves, the suspension moves .98 units at the motion range I measured it at (basically my static ride height). This is typical for McPherson strut cars.

I started running numbers. With those springs installed, there is about 2.25" preload on each spring at ride height (averaged). The coilbind height of these springs (made by Hyperco) was 2.1", leaving me 2.65" of useable spring travel, which seemed sufficient for the rates I was running. Looking closer though indicated that would not be possible. With the A1 racing perches and the ride height I was running (about a 1-finger gap from tire to fender, or 6" from pinch weld to ground), there was about 1.5" of compression travel available between the strut cap nut and the bottom of the spring perch, where the two would hit (the stock strut nut is larger than the ID of the perch). That moment went my dreams of running a minimal bumpstop too. I was able to remove about 30% of the one I was running but it is still there, which is better than metal-on-metal. 1.5" travel isn't a whole lot, especially when if you consider an example 50% load transfer side-to-side would give a corner weight of about 1290lb, or a total spring compression of about 3.23"...leaving about .5" of travel from there to absorb bumps or for further load transfer. Basically, nowhere near enough to make this work very well. I confirmed my theory this weekend once I switched over to some 235 Michelin PS2s while driving hard over undulating roads. It works ok until you hit a bump or a dip, and then the tire skitters and car porposes due to hitting the bumpstop.

So, that problem of insufficient travel is a connundrum. The easiest option is to raise the car up. That doesn't look cool, and is really difficult right now because my coilover perches are all extremely stiff. It took me about two hours of reefing on them to make up for the difference in spring length. It's kind of a pithy point, but I am working on trying to determine if the increase in (estimate) CG height is better/worse than the other possible handling side effects (again, this is an educational project). I should also mention that I am not concerned about camber issues at my current ride height, and I am waiting to make a judgement on roll center height until I get more measurements of strut length and SAI. Not concerned about camber because this suspension won't lose camber until the control arm goes perpendicular with the strut angle (which appears to mean it would need to point up at an 11* angle). Basically, there isn't enough suspension travel from where I sit to make that an issue, any compression will cause the suspension to maintain or slightly gain camber (from the pathetic amount of -1.5*).

The next best solution is to go to shorter strut housings, giving physically more distance between the housing and the upper perch. That costs money for new shocks, though. That said, based on Timmmy's experiences, a nice set of Koni 8611s are on my list someday. I would also be very curious to know what the KW coilovers do about this as far as shorter strut bodies and travel at these ride heights. But those are also spending money.

Another option would be to raise spring rates, and decrease the amount of travel used per load. Two issues with this. One is that after measuring unsprung weight, I went back to my suspension frequency calculator, and put in much more correct data for the unsprung weight portion of the equation. With the corner weights listed above and 400lb/in springs front and rear, I'm running about 2.23hz front, and 2.58hz rear. This is typical/aggressive for a track car with no aero, and based on what I know now should be sufficient for the tires and use I'm looking to run on the track (streetable r-compounds). The rear is about 15% higher than the front, which is sufficient for level ride over bumps at speed and helps it be a bit more neutral. So, based on that I don't really want to be running higher spring rates in terms of handling balance, and I can't on the regular Koni SAs because they seem to start blowing out over 400lb/in based on Jim's experiences.

So the next solution I'm going to revisit is sway bars. I'm not running any right now, partly because of laziness and partly because of the beliefs out there that they aren't needed on these cars. Which is bogus. Based on wheelbase and CG height, suspension travel in a straight line is just fine due to the lower amount of weight transfer. But these cars are much more narrow than they have wheelbase, so more roll rate is needed than ride rate to keep load transfer in check--the same as any car. I'm working on determining the rate of the two different OEM options (sway bar or control arm mounted), and seeing what is available for the back of the car since I foolishly sent my sway bar rear subframe to the scrapyard with the 80. I can only see an improvement in handling by adding these back, but I'd like to get something that allows some adjustment as needed.

So that's that. I've got pages of notes and data and computations written down, and cannot wait to get the car ready to actually do something with it. I'm acquiring a full set of Porsche 996 Turbo brakes next weekend, and tomorrow I might just buy a bumper and a huge list of stuff to get my 30R installed. If all of my maths and textbooks and theory verify, maybe I can move on to setting up something cooler...


Sam

Love your posts Sam. Nice work on the chassis setup.

Good post, ill be watching this! I've been toying with the idea of modifying the front strut housings on the 200.

Baby steps...



Not sure if anyone can see that, but I wrote something that will give you a roll center height based on a couple of strut dimensons. Next step will be to determine load transfer from that, and then see how it is affected by sway bars--which will give me some practical information I can apply to the car right now. From there I'm going to try brake torque and bias since I'm putting together a BBK, and from there I'm going to try to automate the strut calculation so you don't have to manually enter new lengths and angles for different situations. I've spent a bunch of time writing this and a bunch of time trying to make accurate measurements of the strut and its installation in the car. There's a lot of wrongish/incomplete data out there from the other guys I've found publishing results online.

Sam

I'l be interested to see some of the popular setups run through this when you're done.

Quick update, not too much going on this month.

I installed my summer tires, which for now are a set of half-worn 235/40r18 PS2s I bought off of a customer. They're awesome tires as Pilot Sports always are. I put them on only to get dumped on with snow for the next two weeks. The next chance I had to drive the car, I found the fronts rub on the fenders quite badly (odd, since the 240mm wide slicks didn't), and then I somehow managed to strip second gear running only 15psi. It was rather unexpected, as previous failures have all occurred running >23psi and also seemed to always happen on a relatively cold transmission. I only got something like 700 miles out of this one too, so it must've been bad to begin with. Unfortunately I don't have the budget together to put an 01E in right now, so I've got a couple options lined up for 01As that will have to get my by for a little while longer.

A week after I grendaded the transmission, I got a long awaited package from Fedex. So long, ghettobumper!




Still not sure what to think about having $1000 in bumper, paint, and headlights hanging off the front of a ratty old car, but it is a nice piece and I've wanted it for a long, long time. It's an ABS plastic RS2 replica bumper made by Reiger. A dead ringer for the real deal and a really nicely made product. It fits up great right out of the box and should be nice and strong in the long run. The only tweak I need to make is to fix the opening radius on the wheel wells. I'm not too excited to do that honestly, as it seems like something that would be easy to screw up. We'll see, I'm not in a big hurry to get that perfect.



I made some brackets to attach it to the car in a similar fashion to the stock bumper. I may add some kind of metal reinforcement to the back as well to prevent damage from low speed incidents. I've also heard putting expanding foam in the narrow bits greatly increases strength, especially when it's cold out. I've got two pieces left to improve the turn signal mounting and then I need to finalize how my knockoff fog light brake ducts are attached.



Away from the shop, I put about two weeks of solid work into my suspension analyzer spreadsheet. I've got all the data entered that I wanted to begin with, and appear to have all the quirks in my math sorted out. It all kinda boils down to being able to see this:



All of the sudden, changes in vehicle weight, unsprung weight, weight balance, wheelbase, track width, center of gravity height, suspension geometry, roll center heights, spring rates, spring lengths, sway bar dimensions, and installation ratios can all be tested to show how they affect the amount of weight on each corner of the car in a given handling condition. The information you get back shows you how to fundamentally set up a car from scratch depending on the type of use you're going for with it. You see why suspensions are designed how they are and why cars behave like they do. You can objectively mock people doing things to their cars that physics will disagree with!

It isn't perfect yet. The old Audi B2/B3 chassis is extremely easy to model since it uses basically identical McPherson struts and control arms front and rear, so it's half the work for me and is an easy starting example. Hardly anything else uses struts in the rear, and hardly anything else uses a simple control arm setup anymore for either the front or rear of the car--there is a lot of stuff using anti-dive, anti-squat, non-linear spring installations, etc. In addition, all of these calculations are only as good as the quality of their inputs. Because my car was so easy to take apart, it took me a couple of hours to take all of the suspension measurements I needed. Something like a 996/997 Porsche is a bitch to get the springs in and out of, and would probably take days of work to get everything I want measured up--similar story for the Bimmer bangers, and B5/B6/B7 Audi guys. My next adaptation will probably be my GTI, since the front suspension is pretty simple and I own a car I can do whatever I want with. Time to test the mettle of a larger rear sway bar and see if my shitty lowering spring ride is down to insufficient suspension travel or just too much compression damping!

Sam

Great work on the spreadsheet/calculator. Great to see someone doing in-depth suspension analysis.

Soooo...new stuff.



Two things, one is the bumper is now trimmed. I marked it out literally by hand, and spent an hour alternatively trimming and sanding the edges. The bumper now matches the radius of the fenders on the car.



The other is pretty cool. I finally got my 30R and tube header installed today. Had to build new oil and coolant lines and modify the exhaust and intake piping to match. Have some various odds and ends to finish up, but most of the hard stuff is done. I shorted myself on exhaust manifold hardware, I'm missing a clamp for the wastegate tube, I need to figure out an air filter setup, and I need to build supports for the header and downpipe to make sure those don't crack or come loose.



The turbo is a modified 3076R. The wheel is custom made, and is along the same lines as the GTX/Forced Performance billet wheels. Lighter and better aerodynamics for more flow and better efficiency. The compressor wheel was designed and built by Iroz Motorsport and then installed and balanced on the turbo. The header and downpipe are used pieces I bought for cheap from Hank Iroz himself. The header is a modified eBay tube header. The tubes and collector are actually pretty good, but the ching chong flanges are horrible, so they were replaced with high quality Garrett flanges.

Need to replace the transmission before I can drive it, as well as the water pump as it seems to be leaking. Also have a BBK in the works and I need to build some seat parts. I signed up for the SCCA PDX on July 7th, so I'm trying to get this all done by then.

Sam