Corner weight adjustment, I'm way out!

[chrisgtx]

I managed to get my hands on some scales to weigh my sprint/hillclimb car just to see how it sits.
Even with my limited knowledge,It's way out!

The floor was level, anti roll bars still connected, and me sitting in it.
It's on coil overs and all I did originally was wind the spring just to touch the load ring and then set the heights equally from hub to strut top.
Tomorrow I'm going to increase the LR height first to see how that changes things, any tips, this is only a chance to mess about getting it a bit closer as there is some weight to come out the car and move a small amount of ballast.
Any tips?

[stinkspanner]

First thing to do is disconnect the anti roll bars, then in order to get the rf and Lr numbers up a bit you need to wind up the platforms on those respective springs. Or lower the lf and rr to acheive the same thing, or probably do both. Basically fiddle about till you get them a bit closer, and try to get the cross weight as close to 50% as you can which should be acheivable

Edit, cross weight might not be what you think it is. It's the ratio of lf/rr to rf/Lr. It's not the side to side balance

Edited by stinkspanner on Wednesday 7th December 20:56

[GreenV8S]

chrisgtx said:

The floor was level

How did you measure it to determine that?

[chrisgtx]

I should of re phrased that and said the plates were level, had to pack up the fronts.
As a matter of interest I tried them without, albeit not by much and there wasn't a massive difference.
I've been reading up and the cross weights are added up and divided the the total weight, according to the article I read I should be getting oversteer but understeer was mine.

[GreenV8S]

chrisgtx said:

I should of re phrased that and said the plates were level, had to pack up the fronts.

How did you measure to establish that they were level? Did you measure side to side at both ends too?

This is especially significant in your case because the cross-weighting error you've got indicates that either the ground is not flat, or the chasses is twisted, or your suspension is bent, or your springs are mismatched. Of these, the ground not being as flat as you think is by far the most likely explanation and also the one that saves you doing a lot of work to fix a problem you don't have.

[stinkspanner]

Presumably this is a tin top of some description? I agree about levelling the plates, it doesn't make a huge difference on such a car. Also the readings are quite far out at the moment so you should be able to improve that considerably without worrying about the last kilo or so, or the plates not being dead level.

[chrisgtx]

Yes it's a tin top, mazda 323 GTX, i will cross check the levels again and disconnect the roll bars.
I've owned it for 16 years and the shell is straight, I put a new set of racing logic coil overs on it years ago with hard springs. And it's stripped out, no interior.althought I've got my 17 inch road wheels on it not my 15 inch competition wheels on which would save 25kg.
Considering how it looks on the scales it drove well and I've cracked some class records, albeit with a bit of 4WD understeer.
What I was surprised by is the differences in the front, as the engine block is on the right (and me) and the left front is heavier, yet on track the front left would lock up first.
And the rear without me in it is LR=150.5 and RR 254.5 even though the fuel tank is on the left and everything else physically is equal.

[Evoluzione]

It's surprising how much old cars are out. When we started fitting coilovers (just to road cars, nothing fancy) we would measure each coilover ride height to make them all equal thinking this would be the best way. When we dropped the car back down onto a level floor we would measure from cill or wheel arch to floor to check and they would be 10 - 20 mm out from side to side, a rethink was then needed.....

[Mignon]

A simple suggestion to see if the floor is flat is just turn the car round and weigh it facing the other way with the pads in the same places. If all the weights stay the same you then know where you are.

[GreenV8S]

Mignon said:

just turn the car round and weigh it facing the other way with the pads in the same places.

If you mean turn the vehicle through 180 degrees so the front and rear wheels swap places - that doesn't achieve anything. Twist in the ground would result in diagonally opposed wheels being high, and turning the car around still puts the same pair of wheels on that pair of pads. To do what you're describing you'd need to turn the car through 90 degrees - which is not going to be possible without moving the pads.

You need to actually measure the height of the pads to ensure they're level front to back and side so side. It's not particularly hard to do, but until you've done it you can't assume the ground is flat.

[Mignon]

GreenV8S said:

If you mean turn the vehicle through 180 degrees so the front and rear wheels swap places - that doesn't achieve anything. Twist in the ground would result in diagonally opposed wheels being high, and turning the car around still puts the same pair of wheels on that pair of pads. To do what you're describing you'd need to turn the car through 90 degrees - which is not going to be possible without moving the pads.

You need to actually measure the height of the pads to ensure they're level front to back and side so side. It's not particularly hard to do, but until you've done it you can't assume the ground is flat.

The diagonals stay on the same pair of wheels but with positions reversed. You'll still get a change in weights from either side to side or front to rear depending on which way the ground slopes.

[GreenV8S]

Corner weight adjustment is essentially about taking twist/torsional preload out of the suspension and flatness (having all four wheels in the same plane) is critical to achieve that. Just getting it level (same weight readings when you turn the car through 180 degrees) doesn't ensure it's flat. It's very easy to check for level, but it needs to be measured and not assumed.

[chrisgtx]

Sadly I've not had a chance to do anything at all today.
But, after a beer and trying to think logically, the first thing I'll try is winding down the RR and LF? As them two are pressing down the most, although its pretty low ground clearance anyway maybe wind up the LR and RF?

[GreenV8S]

chrisgtx said:

the first thing I'll try is winding down the RR and LF?

Make sure you're clear what you're trying to achieve before you change anything. The ride height figures you reported imply that something is twisted. Since you're measuring the height from the shell to the ground, either the shell is twisted, or the ground is. Which do you think it is?

Edited by GreenV8S on Thursday 8th December 21:40

[chrisgtx]

As I said before, I weighed the car with the scales flat on my garage floor, then I checked and it dipped at the front by 10mm. Side to side was ok, measured it again after getting it spirit level correct .only minor differences noted. So, at this point in time im just messing about trying to get it al little more balanced.
The difference of 120kg between the rears is a little concerning.
There was a comment on here which seems to be removed about someone buying coil overs.setting them all the the same height and it was way off too, that is essentially what I did, the 10mm drop I did last was at angelsey at an event purely done to see what difference it made, I just wound them all down by 10mm, not by measuring cill or wheel arch height.

[GreenV8S]

chrisgtx said:

I checked and it dipped at the front by 10mm. Side to side was ok, measured it again after getting it spirit level correct

I don't want to labour the point, but the weights you reported suggest you have a massive twist somewhere and I would want to understand what was causing that before trying to correct it. You still haven't said exactly how you're measuring the floor for flatness and I hope that's where the problem is. You mention using a spirit level. Have you got a massive spirit level? Have you swapped it end-for-end to ensure it's reading accurately?

[chrisgtx]

I have a straight length of 4x4 long enough for front and back measurements, I understand what you getting at about being perfectly level, but this isn't an F1 set up, even + or - 5kg isn't a problem at this point.
I also know I'm not going to get 50/50 either way, once I've made some changes to the inside of the car over the winter I'll measure again and try to fine tune it a bit, hopefully by then with a bit more knowledge.

[GreenV8S]

By my reckoning your corner weights are out by about 48Kgs so you're miles away from the +/- a few Kgs you're suggesting. I don't know what spring rates you're using but presumably you do - you could calculate how much suspension deflection that represented to get an idea of the degree of out-of-flatness you'd looking for in the floor to account for it. It's probably only a fraction of an inch, which could easily be accounted for by a slight twist in the floor which your spirit level technique wasn't sensitive enough to pick up.

If (as I suspect) the error is caused by the floor not being flat, you would also expect to see the spring compressed slightly more on the 'heavy' corners, and you could easily calculate how much of a difference you're looking for. Just half an hour carefully measuring the spring / damper lengths would go a long way to telling you what's causing the unequal weights. You really do need to know what's causing the error before you try to correct it, because there are very few things that will throw the cross-weighting out, and offsetting the spring seats isn't a good solution for any of them.

Have you swapped a pair of scales over to confirm they're reading consistently?

[chrisgtx]

I'm going outside now to have a go now, I'll measure and check spring heights before I change anything. Swap the plates left to right to see if they read the same. Write everything it down as I go along.
Someone explained to me it's like having a chair with some legs longer than the others.
Also, the car has so many changes to it there's no real guideline to go on

[Mignon]

I really don't see the problem here. In total the car is 77 kg heavier on the right hand side than the left. I'll wager you weigh about 77 kg. There's nothing you can do about that other than shifting actual weight or running with a passenger. Altering ride heights can move the diagonals but it can't change the actual weight distribution. To start with just do everything without you sitting in it. Adjust suspension heights until you get the side to side balanced. You must disconnect the anti roll bars first. Now sit in it and see what happens. Then move actual weight if you have that option.