todinlaw

Paul I did the test I spelled out in my post, jacked up one tire, out of gear, put a torque wrench on a lug nut it started turning at 8.5 ft lbs. if I had it on the center axle nut, I bet I would have been around five. this is not much torque or load. way off of the 20 to 40 lbs that Joel discusses in his email to me. Paul correct me if I am wrong, did you put a guard in, if so, put a torque wrench on a lug nut tell us how much it takes to turn it just to compare, or anyone who has an updated LSD. Thanks for the help.

Just when I thought i was in good shape for the track season. I am going to have to ride it out for awhile, my track budget needs some stimulus money.

997gt3north

I will post my Guard tq numbers asap
Likely take a few days (business travel) and crazy cold weather here

The good thing is you clearly have a working OE unit - unfortunately for all gt3 buyers the OE unit has low tq

996FLT6

With fried LSD with one wheel up could turn freely cw or ccw by hand. With my guard it turn then lock- could not move at all by hand. Mike

TT Surgeon

Did the test, not a gt3, lsd appears to have failed. I take it this test is applicable to awd tt's as well.

997gt3north

- i would think so (lift only 1 rear wheel, other 3 wheels on ground)
- i have to go try it on my S4 with a Guard unit in it (which i know has a working unit - i feel it on every tight slow corner - if i don't disable Stability Control (ESP), the computer kills the gas when the LSD does its stuff every time

997gt3north

- had some time this morning to very quickly test the Guard Unit (50/80) in the garage on my 997gt3
- rear jack point only left rear wheel off ground
- using my hand on spokes of rims (volks), i can with effort slowly turn the wheel - it does take effort
- my tq wrench has been sitting over the winter at my pre-set 90#s so i moved it up and down a bit to re-set
- drop my car back to ground tq the wheel nut i wanted to use to 90#s
- re-jacked the car up
- somewhere between 60 & 70 ft/#s is what starts the wheel turning
- my garage isn't exactly flat and i didn't have enough room to move around so it didn't test if the wheel will keep turning with this force or if it will eventually lock
- using the leverage principle (i can't remember physics class) my tq bar is 50cm and the lug nut is 8cm further out than the center point of the wheel so my 60/70 estimate would have to be grossed up by about 15% so 70/80 is my crude estimate

todinlaw

WOW that is a bunch more than my 8.5 ft lbs in my OE unit. Paul one other question. when we discuss the LSD unit is 50/80 for example are we saying that the preload on the LSD is 50 ft lbs. and the 80 represents the number of ft lbs it takes to release or unlock the unit so the tire rotates again. If I am confuses, what does it mean exactly??

TT Surgeon

Mine took about 40# to turn the wheel, then it got harder to turn, that make sense?

997gt3north

- i'm not an expert at this, that is why i asked for people to comment about exactly what happens when you do this test

- im going to call guard and ask

- i would have thought that it would start to spin and then get harder to spin as the pre-load and lockup come into play and then it would stop until enough tq was applied to make it Slip as in a Limited Slip Diff

- clearly if it just spins and spins easily it is not working

- mine appears to need a constant amount of force (60/70#s at the studds) to turn at all - but i didn't spin it more than 3/4ths of a single rotation in total (some by hand some by tq bar)

viperbob

The 50/80 is the amount of lockup on acceleration and then deceleration. More % on acceleration leads to more understeer in a turn. The higher HP cars generally will do 50/80 or even 80/80 (and more clutch packs). GT3s are happy in the 40/60 and maybe the 50/80.

997gt3north

As Bob said, higher number, the higher the lockup rate - I'm stupid in this stuff so can't convert 50 or 80 to actual meaning

I will say that with the gt2 rear bar on the middle hole to help rotation I can still get on the throttle fully at the apex and the rear isn't kicking out even with my alignment

But also as important if not more important to me (my driving style) is that the "80 decel" lockup holds the rear of the car behind me even with heavy trail braking

Brian S

I tested my 08 gt3 with mostly track miles and here are the results

one wheel up, little friction and the rear wheel turn about the same as the fronts do from brake disc drag. certainly nowhere near half of what it would take me to hit 80lb/torque on a bolt with a torque wrench.

both wheels up, they both go forward with the unturned wheel slightly lagging to start and not rotating quite as much

This may really clue me in as to why my lap times have been raising lately, I have attributed it to air or just being lazy but perhaps the LSD does have something to do with it and I do drive with TC off.

997gt3north

Frank

As I understand it, the pre-load is none of these numbers but is something else entirely and really effects how quickly the LSD reacts (but other stuff comes into play here as well)

The 50/80 or 28/40 (stock OE #s, but Guard tells me the OE unit is actually a 28/55) represents the ratio (if that is the word) that the clutchs in the clutch stack force the diff to lock into - ie inside wheel is turning at 50% rate of the outside wheel under acceleration once slip has been detected)

In the case of deceleration (ie trail braking when slip is detected) the clutch stack forces the wheels to lock into the 80% ratio (ie inside wheel turns at 80% of the rate of the outside wheel - ie it is preventing the car from turning at more than a certain amount - ie even if the rear end wants to come around it is trying to prevent it)

So, even if my technical description it is wrong it is close to laymans term I understand to be correct and now you can compare the numbers

28/40 versus 50/80
- Let's compare 28 to 50 and 40 to 80
- once slip has been detected (this is key) and you have working clutches (this is key as you need something to push against)
- under acceleration our OE LSD will lock the inside versus outside wheel movements into a 28% vs 50% ratio - ie the OE unit at 28 will allow the outside wheel to turn more than a 50% unit
- under decel a 40 versus 80 unit will allow much more rotaion as the outside wheel is turn much more than the inside

So, higher numbers effectively prevent rotation (after slip has been detected)

If "my mother" did not know this was going to happen she may hit the guardrail in an emergency

If I knew this was going to happen I could get on the gas earlier as the LSD is holding the rear end behind me and thus helping me get the power down

If I knew this was going to happen then I could likely brake and turn at the same time (brake later) and I wouldn't have to worry as much about becoming a helicopter but my mother may freak out that the car isn't turning

If you have below 20 pounds (meassured properly - the one tire lift test is close) then the clutches are slipping too easily and may (not for certain) not be able to provide enough resistance to lock your diff and thus when you press the gas in the corner the power will go to the point of least resistance (the inside lifted wheel)


Paul

jrgordonsenior

I had an interesting experience this past weekend with a new LSD that I thought I'd share. I put what I thought was a 45/90 Kaaz diff into my Boxster that I'm developing into a GT4S car with a new 2.7 motor, about 2500 lbs. w/driver, using 245/285 Hoosiers.

I ran it at Cal Speedway last weeekend and couldn't dial out the oversteer exiting high speed corners, the car was sluggish thru the banking and then jumping all over the place under heavy breaking. We adjusted the bars, played with the new JRZ's, all to no avail. I had a pro driver take it for a spin and he came in and said it's the diff. He said it doesn't matter what you do to the suspension, that diff is locking up on the high speed turns scrubbing off speed when you're full throttle. We pulled it Monday morning and sure enough, it was installed as a 90/45 diff by the manufacturer. When I asked why, they said they sell mostly to AXr's and drift cars which prefer the higher accel lockup ratio. I said that's great for them, but this is a road race car and they agreed that flipping the diff (easily done) would be better for a road car. Now I can't wait to try it again. That was really frustrating...

997gt3north

After speaking with the guys at Guard today, I know enough to know that I don't know LSDs -but, I do know there are lots of other stuff besides the quoted numbers that affects how the LSD reacts / feels and it sounds like in addition to being 90 accel there is lots of stuff going on in the diff that is affecting things

Good luck