E30 with SR20DET

[gareth]

that is really odd. unless the engine setup is a lot heavier on the nearside, i can't see how it's come out that way without effectively having a longer spring

were all tyres up to a sensible pressure? ARB's not sitting strained or anything?

i'm guessing the only assymetrical weight you have in the rear is a battery on the offside?

[Theo]

All tyres @ 28psi and the anti roll bars were disconnected. The battery (which is a big bugger) is on the much lighter side!? The exhaust system is presently only supported on the rear subframe and two backbox mounts, but that's not a great weight.

[gareth]

helium filled battery? dead body in the boot?

i can only see it being a ride height problem across the diagonal. thing is, i can't imagine you'd have a collapsed top mount, broken spring or any of the usual culprits on such a fine chariot

i don't suppose the car's got evidence of a whack at all does it?

[GeoffBob]

Theo, I can't remember whether you have rear coil-overs or not? Actually, looking above I see mention of a spring pad so I assume not. I was going to suggest that you do these measurements with the ARB's disconnected, but I see that was exactly the case.

Which means the discrepancy is most likely (IMO) due to either a difference in your rear spring heights (easily observed uncompressed), or your rear spring stiffness. Are they newish springs from a reputable kit? Are you sure that they have identical stiffness?

I'd suggest taking the rear springs out and measuring up their static height, and then putting them in a vice with a bathroom scale. Record the force indicated on the scale as you close the vice up say 5mm at a time. Plot a graph of force vs displacement. You'd be surprised just how much springs can differ from each other.

If you find that you have springs of similar stiffness, but differing heights, then you can put this right with adjustable spring mounts like on Uwe's car. Alternatively, make a plan to safely space the spring seats in order to augment the height of the shorter spring.

Now here's the other thing you should do. If you plan to go back to the scales, take along a mate, two old broomsticks and about 5m of 8-10mm ID clear flexible plastic tubing. Ahead of time lay the two broomsticks next to each other and place two identical marks (dark ink pen) next to each other so that when you stand the two sticks upright on flat ground the two marks will be at the exact same height. Now, fill the clear plastic tube almost full with water, making sure to remove all air bubbles from the tube. Leave about 20-30cm open either end of the tube to avoid the water overflowing when you move the tube around. Now, with the two broomsticks standing vertical a few meters apart from each other on level ground, it should be possible to place the meniscus of the water in one end of the tube on one ink mark, while the meniscus in the other end will align with the other ink mark. If one end or the other refuses to align, then you are not on a level surface. Do you see where this is going?

Take your mate and your "water level" to the scales and put one broomstick vertically on one scale, with the other stick on each other scale in turn, to make sure that the four scales are exactly level. If the four scales are not exactly level (within a ball-hair of the the water meniscus) then you and your mate take those two broomsticks and beat the living crap out of the bloke hiring you time on those scales, because if each scale isn't exactly level, then this is no different to having your spring-seats at different heights!

It's either one of the above, or as Gareth said, look for evidence of a whack.

[Theo]

Interesting points raised. I think the car has had a bump on the offside front corner, but not a significant one. The car does have rear coilovers (not an adjustable damper and adjustable spring seat in the stock spring location) running Faulkner 325lbs/in rear springs (7", I think) I've also got helper springs fitted to prevent the spring from coming loose at full droop. Top mounts are brand new spherical bearing types, so no problem there.

The rollcage is tensioning the shell a little, it didn't drop straight in for whatever reason but it didn't require drastic measures to get it in place, just a bit of gentle leverage. Perhaps this is having a greater affect than I had anticipated.

I'll take the springs off when I get a chance and test them as you mention, not sure my vice will open sufficiently though.

Pretty sure the floor was level, the guy sets up plenty of race cars on it without issue. Not sure how he'd react if I came in and performed the test you mentioned Geoff!

[Jon_Bmw]

Pretty sure the floor was level, the guy sets up plenty of race cars on it without issue. Not sure how he'd react if I came in and performed the test you mentioned Geoff!

He'd be grabbing the broomstick and shoving it somewhere unpleasent!

[GeoffBob]

Pretty sure the floor was level, the guy sets up plenty of race cars on it without issue. Not sure how he'd react if I came in and performed the test you mentioned Geoff!

He'd be grabbing the broomstick and shoving it somewhere unpleasent!

Sometimes people who have been doing a job for years and years are the first to forget the basics. Anyone who is confident in the tolerances of their equipment shouldn't object to an inspection. It's no different to the old days of carrying around a so-many-ounce weight in your pocket to make sure the green grocer wasn't cheating you. And the more people that did it the less likely he was to ever think of ever cheating anyone.

Theo, are your spring seats on your rear coil-overs adjusted identically at present? Similarly for front coil overs? Did you take it in to the scales simply to have it weighed or to have the bloke adjust the seats to balance the car? If he adjusted one of the front seats to get the front balanced (which is where balance is most important) it would have thrown out the diagonally opposite rear corner. I don't think 60kg is very much to worry about. Your font-rear split is very good as it is.

For the purposes of my model, what was each individual recorded corner-weight, and what springs are you running on the front?

[Theo]

I don't think they're identically adjusted - at least one front and one rear seat was adjusted (nsf and osr) to improve the balance across the front axle.

Front springs are 6" 2.5" ID 600lbs/in, corner weights are as follows:

Without driver but with 40 litres of fuel
FL 279.5
FR 284
RL 295
RR 237.5


With driver and 40 litres of fuel
FL 304.5
FR 305.5
RL 304
RR 243.5

(I dont weigh alot!)

[GeoffBob]

Theo, where is your battery located and roughly how much do you think it weighs?

[Theo]

The battery is located in the offside boot well and weighs approximately 20kg.

[gareth]

Theo, i have a loadcell calibration rig at work that also doubles as a damn fine makeshift spring rate tester if you need any springs checking. will only a take 10 mins during a lunchbreak...

when i needed to test mine one weekend to find out what rate i was running, i found a known 240lb/in spring, got mine (unmarked) and set about testing them back to back in the best fashion i could muster in the garage. this involved marking the uncompressed heights of each on the side of a bucket, then in turn standing on them (mass = 1 fat git ) and marking the compressed height of each on the bucket. it's rather difficult to bend over and mark the bucket with a pen without falling off in a comedy fashion....!

anyway, after doing the maths, simply comparing the deflection of the know spring to my spring, i came up with a value of 500lb/in. upon testing at work against a very accurate loadcell and vernier caliper, they were actually 550lb/in. not bad for heath rominson testing techniques. you may not be fat enough to get a decent measurement resolution though

[GeoffBob]

.... This involved marking the uncompressed heights of each on the side of a bucket, then in turn standing on them (mass = 1 fat git ) and marking the compressed height of each on the bucket. it's rather difficult to bend over and mark the bucket with a pen without falling off in a comedy fashion....!

Great story Gareth. One way to do it I guess. Beats putting the bathroom scale in the vice, lol.

Anyone here old enough to remember the kids TV show "Magic Roundabout". There was a character called Zebedy I think who had a spring instead of legs, and he used to bounce around instead of walk. If you'd fallen off that spring onto your arse then your online name could be Zebedy

Theo, I'll post up some results of calculatons for you shortly. I think I can see where your 60kg rear offset weight is coming from.

[Theo]

Haha, nice technique Gareth! I may well take you up on the offer of testing the coil springs on your rig, thanks for that. I also toying with the idea of getting the LSD refreshed, we must have a chat about that sometime.

Geoff, I look forward to the calculations - I'm very intrigued as to how you can pinpoint the offset weight just from a bunch of numbers.

[GeoffBob]

Geoff, I look forward to the calculations - I'm very intrigued as to how you can pinpoint the offset weight just from a bunch of numbers.

I use a mathematical model that I constructed in MATLAB. The model predicts the change in static weight distribution as a function of the height of the spring seat. I begin by assuming a paricular mass distribution and a zero offset/adjustment to all spring seats. I then start adjusting the spring seats in order to balance the car (just as you would in real life). Then I insert "you" into the model (61.5kg behind the wheel) and see if the before and after numbers match your posted results.

And I am sorry to say that I just can't get your numbers to add up. It looks to me as if there is something very wrong with that data!

For your own benefit, do a quick tally of the left and right numbers with you in and out the car. With you out the car the left tally is 574.5kg and the right tally is 521.5kg. That is to say the left side of the car is 53 kg heavier than the right. Now, with you in the car the left tally is 608.5kg, and the right tally is 549.0kg. That is to say, the left side of the car is now 59.5 kg heavier than the right side.

Or, to put it another way, when you climbed in the front RH seat, the left side of the car increased by 6.5kg more than the right did! I would have expected to see the right-hand weight increase by 16 to 20kg more than the left-hand weight. In other words, your 60kg weight should be roughly distributed 1:2 LH:RH. Your data indicates that your driving position is slightly left of the handbrake ????

Theo, I think you need to go over these numbers with the person that weighed your car. Unless you have springs on the left that are an appaulingly bad match to those on the right (or you drive with the handbrake up your bum) then either there is some aspect of his scales that need calibration or one or more of his scales are plugged into the wrong channel of the display unit.

Any chance you got the numbers the wrong way around?

Did he per chance make any adjustments on the car between the in and out of car weights were taken?

[GeoffBob]

Theo, I don't mean to be personal, but do you really weigh 61.5kg, as the above data indicates? If not, then this also points to the fact that there is something odd going on here.

[Theo]

Oh dear, not the outcome I was hoping for. The figures I gave you are the ones I have in front of me, so that's not the problem. I suppose it's feasible the rear corner scales were plugged in the wrong way round, hadn't thought of that. I really should have checked at the time because it immediately struck me as odd that the offside rear was so much heavier than the nearside. The only adjustments were made whilst I was sitting in the car, pretty sure the handbrake wasn't stuck up my harris either. I think.

My weight is correct, I'm very slender!

[gareth]

each wheelpan could (depending if they're rationalised or not - eg, matched outputs) have a different output signal to each other. with loadcells (which these are essentially) the channel on the disply needs to be scaled to give a kg output against a mV input from the loadcell. it is VERY common at work for a customer to use the wrong loadcell for each meter, giving a useless measurement. the outputs if not rationalised can vary by as much as ±25%. when they're with the correct calibrated channel on the display, this is not an issue and they'll read spot on. if not, i'm sure you can see how much of an error could creep in.

in addition to the water / broomstick / pipe test. a very quick and simple test for the scaling on all channels is to stand on the wheelpans. they should all read the same. if they don't, there's a problem.

regards

zebedy

[GeoffBob]

Theo, here's the thing about corner weighting a car:

If your car was slightly heavy on the front left (as a result of the starter-motor, turbo and cast-iron exhaust manifold, for example) then it would be standard procedure to lower the front left spring-seat a fraction in order to reduce the weight on the front left corner. This will result in a simultaneous increase in the weight on the front right and rear left wheels, and a reduction in the weight on the rear right wheel. The front left spring seat will typically be adjusted to the point where balance is achieved across the fron axle, with the driver in the drivers seat.

Now, the reduction in weight on the rear right will help offset the weight of your battery - but to the extent that your rear left is now 60kg heavier For this to happen the front left of your car would have to weigh roughly 80kg more than the front right.

The other thing to keep in mind while corner weighting a car is that, regardless of how you adjust any given corner (and regardless of spring-rates and/or accident damage), the total weight across each axle, and the total weight down each side, will always add to up to the same value!! In other words your front weight, rear weight, left weight and right weight are static values.

As you make adjustements to the springs seats the distribution of the weight may change, but the combined mass of the front-left and front-right , the combined mass of the rear-left and rear-right, the combined mass of the front-left and rear-left, and the combined mass of the front-right and rear-right will always be constant.

[GeoffBob]

each wheelpan could (depending if they're rationalised or not - eg, matched outputs) have a different output signal to each other. with loadcells (which these are essentially) the channel on the disply needs to be scaled to give a kg output against a mV input from the loadcell. it is VERY common at work for a customer to use the wrong loadcell for each meter, giving a useless measurement. the outputs if not rationalised can vary by as much as ±25%. when they're with the correct calibrated channel on the display, this is not an issue and they'll read spot on. if not, i'm sure you can see how much of an error could creep in.

Good point Gareth! We've had our fair share of load-cells come through where I work that do exactly as you have described. Muddle up the calibrated displays (where each display is calibrated for a particular load-cell) and they give awful errors.

I think you may have knocked this problem on the head.

[Theo]

Back to the garage then. I'll pay more attention this time - not sure why it didn't raise eyebrows at the time really.

[GeoffBob]

Theo, I stared at those numbers yesterday evening trying to get them to make sense, and then it dawned on me, your weights should actually read as follows:

Without driver:
FL 295.0 kg
FR 279.5 kg
RL 237.5 kg
RR 284.0 kg

With driver:
FL 304.0 kg
FR 304.5 kg
RL 243.5 kg
RR 305.5 kg

Can you see what went wrong? If not, draw a plan-view of your car and fill in your original numbers over each wheel. Now turn your car through 90' clockwise. See what I mean?

With the front-right and rear-left spring-seats dropped by 4.5mm and 4mm each, the model calculates the following weights:

FL 295.1 kg
FR 279.4 kg
RL 237.4 kg
RR 284.1 kg

These are close enough to your values in my book to indicate a match. I’d appreciate it if you would phone up your guy and ask him which corners he adjusted. I'll bet you a pint he'll tell you he lowered the front-right and rear-left (or converesely raised the other two)

Now, when I add you into the drivers seat (not on the handbrake this time) I get:

FL 302.9 kg
FR 304.3 kg
RL 246.3 kg
RR 303.9 kg

I'd say that these are a close enough match to the measured values given that I don't know your exact seating position.

Your cars static corner masses (without you in the car), BTW, will be close to:

FL 285.0 kg
FR 289.5 kg
RL 247.5 kg
RR 274.0 kg

(Edit: Last two values above were the wrong way around)

The only way to actually measure these though is by inertial methods, as these are real masses, and not simply the forces exerted by the masses through the frame, springs, wheels etc onto the scales. These values do, however, indicate that the front-right of your car effectively has 4.5kg more mass than the left side (pretty good I think) and the rear-right of your car has 26.5kg more mass than the rear left. These are the point-masses effectively acting upon each wheel.

Now what you need to absorb from all of this is basically two things:
1) The balance across your front axle is already very good
2) The sole purpose for dropping your front-right and rear-left spring seats by almost 5mm each has been to offset the mass of your battery (located behind and to the right of your rear axle) and you (offset slightly to the right of mid-way between the two axles)

Take note that any mass that you locate behind your rear axle will add a weight to the rear wheels greater than its own mass (for the simple reason that it lifts weight of the front axle).

If I were you I would seriously consider relocating that battery to somewhere middle-forwards of the rear axle. Then have your car corner-weighted over again.

And remember, while it may well be the weight over each wheel that defines the traction at each wheel, it is the inertia of your car that defines how you are going to spin out and crash once you lose that traction. A well balanced car is not only about weight distribution (by means of corner-weighting), it’s also about good mass distribution. And in your case you can do a whole lot by relocating your battery. I honestly think it would be a shame to leave it where it is at the moment.

I trust this has been of some help to you

[appletree]

Geoff, you allown make me wish i'd paid more attention in maths at school rather than continualy asking the teacher why it was important and were would i use it in life, know i know LOL

[GeoffBob]

Ta Matt.

BTW, I hated maths so much at school that I was determined that if I absolutely had to learn it I was sodding well going to do something with it that I enjoyed - like cars! And the more I use it for something practical the more I find I enjoy it

[Theo]

Can you see what went wrong? If not, draw a plan-view of your car and fill in your original numbers over each wheel. Now turn your car through 90' clockwise. See what I mean?

Did that, I think you could be on to something- very cunning. It was the FL and RR which were raised during the adjustment process by the way. I'm bloody impressed by your results from using the model, kudos indeed.

Makes much more sense (i.e. more than zero) that the battery is causing an imbalance across the rear. I think the battery could easily be relocated to the floorpan area behind the passenger seat, 'suppose would be worth fitting a smaller one whilst I'm at it.

The certainly has been of great help Geoff, many thanks.

BTW, I hated maths at school too, but on the plus side I did sit next to an extremely attractive girl who was prone to the occasional 'fondle', those were the days.

[GeoffBob]

It was the FL and RR which were raised during the adjustment process by the way. I'm bloody impressed by your results from using the model, kudos indeed.

Ta very much. Raising the FL and RR has the same effect as lowering the FR and RL spring-seats. I would have expected him to follow the latter option though, as the car will sit lower overall, but really not that much difference between the two. Certainly no difference as far as the corner weighting is concerned. If you ever want to adjust your ride-height just be sure to adjust each spring-seat by an identical amount , that way you won't effect your corner weights.

Makes much more sense (i.e. more than zero) that the battery is causing an imbalance across the rear. I think the battery could easily be relocated to the floorpan area behind the passenger seat, 'suppose would be worth fitting a smaller one whilst I'm at it.

I think this would be a great idea. I did the same on my own car. Although my seats are so far back I had to basically mount the battery directly over the differential. I also opted for a smaller, lighter battery.

BTW, the corner masses (not weights) that I calculated include both the sprung and unsprung. I can't tell you your exact sprung mass (over each wheel) unless I know your unsprung mass at each hub. Not really what you wanted to know anyway, but thought I would mention it before someone comes along and makes this point for me.

A question: Any reason why the FR could have a slightly greater mass (4,5kg) than the FL? I would have thought he opposite with a cast-iron manifold, exhaust, turbo and starter-motor on the left. Wait a minute your starters on the right, same as my Toyota? Anything else down there to push up the right? Pedals and brakes I suppose? Your inlet manifold is aluminium so that's quite light. Any ideas?

[Jon_Bmw]

As I am an inherrent pikey and have no real budget for the 205-and a lack of space under the bonnet, I put the battery behind the passenger seat. It seemed the obvious place really, nice and low down and offset a small amount of my puny weight. Logical, if not mathmatical.

Steering column is over that way ish Geoff. Not sure if that is far enough forwards, but affectively it is quite high up??

[Theo]

Ta very much. Raising the FL and RR has the same effect as lowering the FR and RL spring-seats. I would have expected him to follow the latter option though, as the car will sit lower overall, but really not that much difference between the two. Certainly no difference as far as the corner weighting is concerned. If you ever want to adjust your ride-height just be sure to adjust each spring-seat by an identical amount , that way you won't effect your corner weights.

I did say to him that I didn't want the car any lower as it is already very borderline going over the smallest of speedbumps. If it was a track only car which I transported on a trailer it would have been the opposite corners which were adjusted. I'll remember the tip on adjusting each coilover by the same amount - good tip.

A question: Any reason why the FR could have a slightly greater mass (4,5kg) than the FL? I would have thought he opposite with a cast-iron manifold, exhaust, turbo and starter-motor on the left. Wait a minute your starters on the right, same as my Toyota? Anything else down there to push up the right? Pedals and brakes I suppose? Your inlet manifold is aluminium so that's quite light. Any ideas?

Hmm, as well as the starter there is the alternator and also the oil lines and remote filter housing are on that side. As you said the pedals are of course on that side along with instruments etc so perhaps that accounts for it?

[Theo]

Well, had a little play at Brands Hatch today after booking it cheaply last night.

A little vid of a flyby:

[appletree]

Nice vid, any incar footage???

did it perform as you had hoped? looks pritty quick in the clip and spitting flames to!!

[DRIFTBOY]

Very nice mate!

You were lucky with the weather too!

As appletree said, what are your thoughts on it...?

[Jon_Bmw]

I'm retracting the use of my wideband! You'll burn the wiring

Looks good. I am so sad, that for about 2 minutes, I would love my car to pop a flame or two.

[Theo]

No incar footage as I don't have a camera mount, got a few more clips which I might upload but they're all quite similar. The flames are awesome, but when I get my dump valve working properly I think they'll be a thing of the past.

I was really pleased with the car, only problem is a slightly dodgy gearbox which occasionally rejects my attempts at changing gear. Brakes were excellent after I knocked a bit of rear bias off, although Brands probably isn't the best place to judge this as all the braking zones are on cambered or changing surfaces. Power is great, wasn't overtaken all day (except for in-laps)although it was a very slow bunch of cars as it was predominantly an MX5 trackday.

Balance of the car is very nice and it had excellent grip. Nice still coilover suspension really does make a massive difference to the wobbly e30, it's so nice to drive a car which has hardly any roll.

Oil temp and water temp were perfect, boost was a little higher than I wanted but I just cannot reduce it any further with the manual boost controller I have. An electronic boost controller is on the christmas list, along with a bigger baffled sump.

[gareth]

looking good theo i do love a bit of flame spitting
mine spits flames when running on LPG, but being gas, they're blue ones nowhere near as impressive as that though

flamer kit, wikid bruv innit

[GeoffBob]

Excellent work Theo . Damn, I suddenly have the urge to get mine finished. What a pity we aren't in the same country, we could have a little race. I'd probably be just ahead of you, up until the point where you lap me

I trust that you are adjusting your brake bias via the bias bar? Remember our earlier brake discussion: Get you prop valve set to, ummm, whatever that value was I worked out (which caps the rate of increase in rear pressure), and from there on adjust via the bias bar. If the prop valve isn't set correctly, you'll battle to get her setup spot-on with the bias bar alone.

Once again, well done. Those flames are very Group-B rally .

[Theo]

There's fat chance I'd lap you, but it would be great fun to have them on track together.

I was using the balance bar adjuster to tweak the bias, the proportioning valve is left alone. You advised I set the rear line pressure to 500psi - I think at some point I'd like to test the rear line pressure with a gauge to check I've got the valve set correctly (I never did find a graph for it). Perhaps I could even swap it for a screw type for finer adjustment and then lock it in place somehow.

On a seperate note, I wa getting intake temperatures (post intercooler) of around 25 degrees C, when the ambient temperature was something like 6/7 degrees. Is this good/bad/average?