Thats a bummer. I wish you could just come drive my car and see how it is. Its hard for me to know if we just have different perceptions of the popping or if yours is just significantly more noticeable. I alos may just be forgetting how noticeable mine was 5 years ago and it has just worn in with time. I really hope this track day wears it in and gets it to a place you're happy with. Its such a pain in the *** to take it out and restack if it is still too harsh.

 

there is one other option. I do know that Kaaz offered a thicker cone spring / belleville spring for their LSDs. This guy goes through the process of installing a thicker spring during his lsd rebuild:
https://pinderwagen.com/build-topics/diff/kaaz/
If memory serves I remember seeing somewhere that depending on the build date of the LSD you might have the thinner or thicker belleville springs. If you have an older LSD with the thinner ones then it would stand to reason your preload setup is lower and therefore you will get less popping. I don't know if its the same case for Tomei, but an interesting topic. I may call up Tomei USA tech support and see if I can find some info on this. Maybe they even have different belleville spring sets for dialing in the LSD.

 

Back from lime rock and lots to unpack:
- on the LSD
The diff worked great, overall it felt like the rear end of the car had a LOT more traction and I was able to put power down much earlier than before. The entire day was essentially an exercise in re-learning how early I can apply power. Where previously my car was always lacking rear end grip and very tail happy, now I was able to apply power earlier and earlier. After only 4 session I don't think I am anywhere close to extracting the full potential of the new diff, it will take a few more sessions to relearn to drive the car and be much more liberal with throttle application so as to utilize all of the usable grip. Definitely a big improvement and if anyone is on the fence about an LSD for track use, I would say definitely do it. It's a noticeable improvement.
Once I got home I inspected by diff cover and overflow tube and found absolutely no signs of spillage or leaks, which suggests the diff was not overheating in any way and seems to be quite fine in terms of heat management. A big plus especially considering a warm day (low 80F weather mid day) as I'm not too crazy about adding diff coolers if not needed.
edit: Forgot to mention that I noticed no abnormal understeer/oversteer behavior during mid-corner power application changes. This suggest the lock up during changes in power application mid corner is not too drastic and I likely do not need to reduce the active surface count. Overall, very smooth and predicable during mid corner power application changes.




- on the car
Despite my significant efforts in correcting the overheating problems I had last year at NJMP by adding a new radiator, improved radiator ducting, and bleeding the system multiple times I ran into overheating problems AGAIN at lime rock....so what gives. My track buddy and I did a few more bleeds between sessions and eventually confirmed what I had feared...found a bit of oil floating in the top of the funnel during one of the bleeds. So it appears that my headgasket is done. Honestly kind of glad I know what the problem is now. The car has been down on power per the calibrated but-dyno and the fact I am getting destroyed on the straights while catching ppl in the twisty bits. Seems now is as good a time as ever to swap or consider engine alternatives. Oh well.


Despite the engine trouble the car actually behaved really well. Great balance, the LSD was excellent on track, no abnormal tire wear, and I made it back home without issues.




So the Tomei LSD will stay, its a great upgrade, and with 6 active surfaces on either side worked very well on track (actually made ZERO noise on track, compared to the parking lot where it popped quite a bit HA!). The only thing I will do once I have the time for it is likely pull it and do something about reducing the preload. This is because even with hardly any load on the diff for example when coasting or adding a tiny bit of power and making turns on the parking lot the diff POPS pretty hard especially when warmed up. This suggests to me the preload is just too high. There is no need for the diff to lock up as hard while coasting or when moving at minimal power applications. I will see if I can get a weaker belleville spring washer set or if I can remove a disc on either side of the LSD stack so there is less pressure on the existing Belleville washers. I will certainly update the thread when that happens and post pictures of what I did.
Also will do one more oil change and see if the track day break-in and an oil change quiet it down a bit more. Good upgrade, a bit loud, and definitely a good idea to re-configure to a lower lock-up setting, but overall a well functioning upgrade

 

My '07 autocross car has had two potentially dangerous spins going onto full throttle through a 4.36 rear gear in a tight turn. Apparently the less loaded axle shim pack is not unloading? Reduced the Tomei to 80% with no effect. I plan to follow your lead and wonder if you found out anything with regard to a thinner preload washer. Thanks

 

Since the overheating issues at lime rock I've parked the car and been focused on a few other items. Actually thank you for reminding me, I need to call up Tomei USA and ask about the thinner preload washer and if its available. That being said 80% seems too aggressive on this LSD from my experience (however brief and limited at this point). I'm running 6 out of 12 contact surfaces so effectively 50% lockup and the rear felt very nice on even the lower speed tighter turns like the T3 lefthander at lime rock. I did not notice a spike in rpm or loss of inner wheel traction as the diff gave up due to lock slippage. The bigger issue however to me seems to be excessive pre-load even at zero or very little power, just coasting through a turn at little to no power the diff seems to chatter especially when warmed up. This is where the shims come in or potentially removing a disc or two out of the entire stack so that the disc stack thickness is reduced and there is less pressure on the bellevile washers and with that less pre-load.

I'll call up Tomei on monday and have a chat with them. Depending on what they said one of these two options is the way to go:
-thinner belleville washer
-removal of discs from the stack

Unfortunately whatever the solution is I probably won't get to test it for a year at minimum since I'm taking a work assignment abroad for a year, so it will be up to people like you to test it out. Overall this is a very nicely built lsd, but what sets it apart from the likes of OS giken is that it really requires a bit of engineering and understanding to configure correctly, while the OS giken comes exceptionally well configured out of the box.

 

Great! I haven't found any correlation of this, but I'll beginning to wonder, if as I improve grip via suspension and tire choice do I even need a LSD. Thanks for posting your experience.

 

You may benefit from a swirl pot setup if youre constantly chasing air in the cooling system. Sucks that the headgasket is done though. I myself, would rather blow on the track while having fun than on the street.

 

Thanks, that's actually a really good suggestion. Granted the engine needs a rebuild, but I could probably manage a bit more abuse out of it this way.

On a different note, I did get a hold of a representative at Tomei USA (Taka, if you read this, thanks again for the nice discussion!)
He will reach out to their engineering contact at Tomei Japan regarding a solution for reducing preload. Apparently there is a thicker "spring disc" (I presume that's the name for the belleville washer inside the diff they use) that is available if one wants to increase the pre-load or breakaway torque, but nothing regarding going the opposite way, which is exactly what I want to do, that is reduce pre-load or initial breakaway torque.
Also here is a great diagram and explanation of preload and break-away torque for those looking for better understanding:

 

Great news! According to Taka @ Tomei and his engineering contacts at Tomei Japan there is a way!



Here's what Tomei said:






In short it lines up with what I said earlier; the easiest way is to reduce the thickness of the entire clutch pack and the simple way to do that is to remove the two spacer discs. Honestly I feel a bit dumb for not noticing earlier that there were two spacers in the pack. As I look back over my pictures from the install its obvious there are two spacers that don't do anything and can be easily removed to reduce thickness of the clutch pack. They sit right next to the spring discs (Bellevilles washer) on both sides of the clutch pack :



Unfortunately because I am taking an abroad work assignment for a few months it is very unlikely I will get a chance to remove these spacers and give the diff another test run for at least the next 6 months, but if any of you guys want to try it and report back on the LSD noise and chatter reduction, it would be greatly appreciated.

 

Can't thank you enough, Row2K. I'll report back. After thinking more about clutch type LSD function, I'm not certain preload had any part in my sudden spins. I'll be looking for something broken as I apply the methods above. Great work!

 

if you take it apart and take a look inside, please chime in, always good to have more information on what's going on inside these units this way we all learn how best to set them up.

 

I did think of one other thing that should be considered when stacking and setting up the Tomei LSD. The clutch disks ARE directional. Let me explain, take a look at these two clutch discs from the stack:



Notice that disc A has the groves moving clockwise in and disc B has them moving counterclockwise in. These discs are identical, so what you are looking at is a directional disc that is simply flipped over. This brings up an interesting point. Depending on how you flip this disc it will draw oil in or it will draw oil out. The LSD does not have much in terms of oil input flow locations along its axis, so really the only way to get the proper lubrication between discs is to draw oil in from the outside.
I'm now quite convinced that if anyone is looking for better lubrication and smoother function of the LSD during operation the discs should be stacked so that as the diff rotates it will pull oil in between the plates. I think we can all agree that metal on metal contact with little to no oil flow is bad and will likely result in rougher operation (likely more chatter) vs when there is an oil film between the discs. So the next time I take this thing apart I will re-orient all the discs so that as the LSD rotates during normal operation (same direction as the wheels) the disc grooves will be oriented to pull the oil in. Below is a quick diagram on the 2 ways you can orient the grooved discs and what I think is the correct way to set it up so oil is drawn inwards:

 

Note taken. Never thought there was a difference. I'll document what I find. Thanks, again.

 

After disassembly I found one spacer missing(as from the oem) and one "T" disc with broken/bent tabs(my fault from a previous teardown?).on opposite sides of the pressure ring. Also found evidence of past sway bar-to-toe control arm binding on the left. Everything else looks like new so I presume these things, combined, led to sudden spins under full throttle exiting a left hand turn. I've attached stacking diagrams of two breakaway torque tests. I'm going to try the 66% version. Thoughts, anyone?

 

EDIT: I used the wrong conversion values, you look like you are on point with the breakaway values






As for the broken tooth, if it was on the outside I would just suggest be very careful not to bind any of the teeth between the bottom and top parts of the housing during assembly. I had a particularly difficult time getting the clutch stack into the housing and not binding any of the discs between the bottom and top housing shells during installation. I would imagine that if any of the teeth get snagged or bound between the housing top and bottom you would instantly see a much much higher breakaway torque. So that may be something that is going on in your case, although I imagine rather unlikely.

 

Here's the conversion table I'm using. https://www.convertunits.com/from/kg-m/to/lb-ft.

1 kg-m to lb-ft = 7.23301 lb-ft

83% is with no shims, lightly lubricated plates and the stackup above. 66 lb-ft / 7.23301lb-ft/kg-m = 9.12kg-m. So I think I'm pretty close to what Taka predicted.

 

Yup, my mistake, I edited the earlier post with the correct values. I used the wrong conversion. You are 100% on point, looks good !
Im very curious what your feedback will be with 83% locking especially on things like long sweepers when you go on/off throttle. Will the car have a sharp understeer push when going on partial throttle mid-corner?

The last time I drove my car was about a week ago driving it to storage where it will sit for about the next 6 months and as I was taking a few sweepers going onto and off highway ramps I did some on/off throttle tests and noticed that even at only 50% lockup (6 contact surfaces) it has a distinctive and noticeable push to understeer when throttle is applied. Granted this is at low speed and no-where near the limit of grip compared to when I did this on track with a fully heated up diff and heated up tires this behavior was simply not there, but still worth noting for those that only drive on the street at well below limits of adhesion that this on/off throttle behavior is noticeable even at only 50% lockup and both spacers installed. I will test this again once I remove the spacers to reduce the preload, but I'm definitely sticking to 50% lockup for now and really looking forward to your feedback regarding 83% lockup.

 

I'm trying the 66% option (lowest described by Tomei as applicable). Since this is for autocross I won't have the high speed issues that you want to correct. Using a 4.36 rear gear I have tons of torque in 2nd gear of a 5AT and a wide torque curve to throttle steer. Goals are to get rid of the harsh popping in the paddock/grid areas and get both axles to have LIMITED slip in tight turns. Have a new thought about the sudden, full throttle spins to check out.

 

That makes sense.
Hopefully by the time we test a few different options it'll be possible to put together a general setup guide for this LSD. Like I mentioned before I think this is a very nice and versatile unit, but it is most certainly not a plug and play solution out of the box unless one is looking for a hard locking track only application (drifting maybe?). Otherwise to get it to behave just the right way it most certainly requires some clutch pack restacking, preload adjustment, and an appropriate lubricant choice.

 

I just finished repacking my tomei 1.5 lsd and learned a few things. First is that the above is not correct sadly. The disks are stamped with concentric circles that are not aligned to the center. It took me quite a while to figure it out tracing the grooves with picks. They do not spiral, so you do not need to worry about getting them the correct orientation.

I had been running 8 contacts faces for 40k miles and was releaved to find very very minimal wear to the disks. I did want to go significantly softer so I restacked it with 5 friction faces and removed the spacer shim(I only had one). I left all 24 friction plates in. I do wish someone had a suggestion on bolting the lsd case back together with less frustration. Getting the two halves together without pinching one of the little clutch plate ears is a struggle.

I also found that my stub axle seals had a bit of wear and the lip was getting thin (214k miles) So I went ahead and put new seals in. I should have it back in in the next week or so and will update on how it feels.