Traction issues when turning left

[pepejesus]

Disconnecting the rear sway will make the car oversteer more, be aware of this!

Disconnecting the rear sway would (should!) have the effect of moving the balance more towards understeer.

[rascal]

Disconnecting the rear sway will make the car oversteer more, be aware of this!

Nope, its the other way around. Disconnecting the rear will give you less oversteer.. And better drive out of corners.

[Roadrunner]

Did a bit of sleuthing. Good news is the Torsen is working.
On advice from a friend, I put the car on the grass, put the scissor jack on a bit of timber and lifted up the drivers rear juuuuust off the ground and started the car. As expected in gear it spun freely like an open diff. I revved to 3000rpm and applied the handbrake to put the raised wheel under load and it drove straight off the jack via the rear left.

I also tried to disconnect the swaybar to check if it swivelled freely in the bushes. Couldn't get the nut off the drop links but while I was swinging on it the bar did move opening a gap in the bushes, so I guess they are ok?
However the bushes looked squeezed out the side and quite a bit of paint has worn off the bar.
Does this look normal? It's the same on both sides.

[NitroDann]

I revved to 3000rpm and applied the handbrake to put the raised wheel under load and it drove straight off the jack via the rear left.

that would work with an open diff.

[Roadrunner]

Really??
Any other suggestions for testing a Torsen?

[sailaholic]

According to miata.net SEs, late nbs a all NCs didn't actually get a TorSen but another brand that were manufactured differently using metal cone clutches. It's still unlikely it's worn out, but it's also not serviceable.

Sent from my SM-P605 using Tapatalk

[The American]

In theory... The trick would work with an open diff IF you applied enough brake force to make the loaded wheel on the ground the easier path for the torque. If your hand brake is balanced, this might never happen. The reason why it works with the Torsen is because you only need to overcome 20% of the amount of torque required to turn the loaded wheel.

Unless you've been giving your diff massive abuse, it's probably fine, and wheel spin in corners will be due to unloading the inside wheel.

Because of the way the torque bias ratio works, the wheel with grip can only ever accept a multiplier of the torque the unloaded wheel can transmit to the road. Think of it this way:

Let's say your car has 100 torques, and you're using all of them. If suddenly one wheel looses traction and can deliver less than half your torques to the road - let's say 40. Happy times! The diff will do it's job and transmit upto 4 times that amount to the other wheel according to the bias ratio: 80% or 160 torques to the loaded wheel. Given that you only had 100 to start with, all of the remaining torque will be available to the wheel with grip = no wheel spin.

In the next corner, your inside wheel unloads even further and can now handle only 20 torques. Still all is good, the other 80 are sent to the loaded wheel according to the 20/80 torque bias ratio, but you're at the limit of grip with the Torsen.

The third corner has more camber, is tighter and slower, and the inside wheel unloads to the point where only 10 torques can pass through it to the road. According to the torque bias ratio, only 40 torques can be delivered to the loaded wheel for a total of 50 torques. That's only half the torque your trusty 5 has at the wheels. Unless you lift off the the torque pedal, the other 50 simply becomes wheel spin of the inside wheel.

Any time you're lowest grip tyre has only enough grip to handle less than 1/5 of your 'at the wheels' available torque, you'll start to get wheel spin with a healthy Torsen,

That, or my Torsen is also stuffed.

[Magpie]

Yes I got it wrong on what happens if you disconnect the rear sway, blonde moment

[Roadrunner]

So I have come to the conclusion that the diff is working fine.

But, I think the issue has presented itself in the suspension components.
Firstly, ride height is the same left to right.

When I jacked the car up by the diff, I noticed the passenger tyre lifted off the ground much sooner than the drivers side.

With the car in the air I started measuring everything I could think of not knowing what it is I am meant to be looking for.

Jacked up, wheels off the ground, I measured the distance from the top of the spring (top plate?) to the spring base ring and both sides measured exactly the same (160mm) but could easily tell the right wheel dropped lower than the left.
However when on the ground, the same distance measured 126mm on the left side and 104mm on the right. The right side spring is compressing around 20mm more than the left side.
Clearly the lack of drop on the passenger side wheel is why I'm lifting a wheel/braking traction on the left wheel in left hand turns only.
In fact when jacked up, the gap from the fender to the top lip of the rim was 20mm bigger on the drivers side.

Jacked the back end up again and measured the whole length of the strut from top plate to the bottom locking ring and they are the same length both sides.
Am I right in assuming that the passenger side spring has 20mm less pre-loaded/compression than the left, and total ride height adjusted via the bottom collar to match the drivers side? Hence the equal total strut length but the wheel isn't dropping as much as the right when uploaded?
But if that is the case, why is the spring distance equal when the wheels are off the ground?
I have only ever adjusted ride height by the bottom collar, as the manual says do not adjust the spring perch as this is set from the factory, so am hesitant to adjust it if this isn't the issue.

Do I just need to adjust the preload to match the drivers side (when the car is on the ground) and then adjust overall height, or is there an issue with the coilover/spring on that side?

I assume the sway bar has zero effect when both wheels are on the ground or both raised in the air and so isn't contributing at all to this issue? Or should I persevere and disconnect it just to rule it out?

[Locutus]

From what you've described, you have the same amount of spring preload on both sides and the same shock length on both sides so the problem is elsewhere.

Disconnect the sway to rule it out since it is quick and easy to do.

If the problem still exists I would check for bent wishbones, worn bushings and retorque all suspension bolts, in that order. Note that suspension bolts should be torqued with the weight of the car on the hubs.

How even is the front?

[Roadrunner]

Haven't measured the front, I will do that tomorrow and take some photos too

[Magpie]

You would expect to see different amounts of preload on each corner due to corner weights.

For example
FL 293kg and 100mm FR 331.5kg and 100mm

LR 263kg and 108mm RR 293kg and 112mm

Irrespective of spring perch height the amount of suspension travel should be the same for each side. If you do not have the same amount of travel then the damper length is incorrect and this needs to be rectified.

To do this.
1. remove damper and disconnect the sway bars
2. remove spring and bumpstop
3. turn damper to softest setting and compress damper. Take measurement when damper is fully pushed in.
3. install damper
4. jack up with wheel installed and check measurement on damper. You want the damper to be fully compressed, but not contacting anything. For the front wheel make sure you turn the wheel and see if it hits anything.
5. Adjust the damper length so that you have the full amount of travel but not rubbing on anything. You use the lower collar for this (adjust damper length).
6. Once you have the correct length, make sure the lower collar is very tight, uninstall the damper, take measurements of the damper, put it back together and reinstall. Make sure the spring perch is low enough for the spring to be uncaptive.
7. Take out the other side and duplicate the measurements.
8. Repeat for the other end of the car.

9. Once damper length is set then you can do ride heights, sway bars disconnected.
10. Adjust ride heights by the spring perch. Wind up the spring perch collar so that the spring just touches the top of the damper. Lower the car down and take measurements of the ride height. Jack the car up and make adjustments. Personally I do 2 full turns, lower the car and measure the ride heights again, note the change and calculate how much the ride height changed for the 2 turns and use this calc to make more adjustments. DO NOT TOUCH THE LOWER COLLAR.
11. If you are doing this with no driver allow about 5mm of rake in the car.
12. When finished use liquid paper to put marks on the collars so that you can check for any movement or as reference points for any changes.
13. Connect the sway bars, again liquid paper on the bolts as a movement indicator.
14. Take for a drive and fine tune ride heights

Personally I prefer to use the pinch welds to measure heights. I use liquid paper to put measure points on the rails so that I can keep things consistent.

[pcmx5]

I had a similiar ssue when one of my rear Tein SS's was stuffed. New Coilovers= no issue.Would lose traction quite suddenly when trying hard.Torsen fitted.

For others information,in my NA the space saver sat almost on the RR adjuster for the Tien's;now they tell you to avoid this by mounting the Space Saver through one of the wheel stud holes not the centre bore.

Peter.

[Locutus]

You do not need correct preload or the car corner weighted to accelerate out of left turns.

[Roadrunner]

So I did some more measuring in better light. (also can not for the life of me get the rear swaybar end links undone)

This photos show exactly whats happening:


Double checked ride heights are equal left to right, but have a look at the spring compression on the right compared to the left (and subsequence difference from spring perch to bottom collar to make the heights the same).

Left:


Right:


Heres some measurements:

At ride height:
Top plate to spring perch:
Left: 130mm
Right 100mm

Total strut length:
Left: 202mm
Right: 205mm

Raised:
Top plate to spring perch:
Left 160mm
Right 155mm

Total length:
Left:232mm
Right: 260mm

Conveniently I have another NB here also with the same suspension, ride height etc.
Took the same measurements on the ground and raised and it measured equal changes side to side.

Soo....why the difference in my rear shocks? Why does one spring compress more than the other when lowered when the other car stays the same side to side?
In fact, to get equal ride height the left side body is wound to the very bottom of the bottom collar, which in turn is raising the wheel up.

Stuffed shock/spring or just not set properly when installed?
For what it's worth, it rides and handles perfectly fine on the road.