TTurbine

After about 3 track events ,

The LSD on my 7GT2 and my brothers 7.2 GT3 were shot dead and even the casing was cracked ,

Pictures where posted by Matt on Mikes LSD buster thread!!

MarekN

What Kussmaul says goes together with what I have on my car, when it is jacked up there is hardly any load. The wheels turn in the same direction, though, when you turn one the other goes the same way.

With the paranoia regarding this subject, it almost reads like an explanation made up around the way our diffs behave in the real world.

But of course it would fail the regular jack-up test discussed previously. If you hold one wheel you can turn the other.

In a second gear corner I can definitely feel the diff locking and the rear tightening up on throttle, so it must be working somehow.

Does the diff use the lateral g´s somehow to start working?

The car now has 36000 km. More than half of that on empty twisty roads, second, third and fourth gear stuff.

Nizer

This all sounds perfect for the street....

Nizer

I think all of what was said sounds fair and fits with observations made here in past - static torque tests are of limited value; significant radius change on/off throttle, etc.

The one comment that doesn't seem to hold water based on feedback is durability and Matt has given detailed explanation in other posts as to weaknesses of new diffs.

https://rennlist.com/forums/997-gt2-...sd-buster.html

Comments from Matt on static torque tests from 6speed thread:

Hello,
The answer regarding the wheel test being a fallacy is that it's a bit tricky to explain and I'll try to give you a bit more detail. As I tried to explain in that long technical post on page 2, when the car is on the road you very quickly and easily overcome the pre-load on the differential. This is what I am talking about when I say that it's acting like an open differential. The preload on the differential doesn't put enough load on it to make it lock up. There's just not enough force there.

The reason that the wheel test is useful as a basis for determining the condition of the LSD is that it gives you a rough idea of how much friction the discs are still generating. It's something that you can measure over time and see that the differential is wearing. But when you do this test the differential isn't truly locked. The ramps are still pressed inward, instead outward into their locking position. It's really just an empirical gauge of where the diff sits relative to it's new condition.

It's kind of like running on a dyno. A dyno is a very useful tool but it's all relative. For it to be useful as a tuning tool you first need to run a baseline. Then you can modify your car and see changes over time and how much power they give you. Porsche has published what the LSD should show at rest when it's new. When it starts to drop substantially then you know that it's not going to lock up the way it's supposed to in the field. Does that make more sense?

Also, look at this new OS Giken differential for example. They are claiming it's a zero pre-load LSD so the wheel test shouldn't work on that differential. Not having tested one I cannot tell you if this is the case or not. But the point is that it's not a universal test that you can use in all situations.

Significantly expanded commentary from Matt to be found here:
http://www.6speedonline.com/forums/g...sd-buster.html

FFaust

Matt's comments above and in particular the one posted below begs the question:

If a LSD has some pre-load, it means that it is always slipping to some extent during normal driving (in anything other than a straight line). Does this slipping then cause the plates to wear with time? In the affirmative, would it therefore not be better to have a diff with no pre-load, a diff that would be more an on-off unit?

997gt3north

If you are counter steering in the brake zone (no joke here) it is time for a rebuild. At first I was like WTF, then I just used the opportunity to become a better driver counter steering at 240kph. While my mechanic was happy to keep checking and charging me for new alignments to see if we could find the "problem", he eventually removed the OE LSD checked it out and said this POS is toast - he was actually more than a little stunned - just to be clear, he runs his own shop, raced porsches is his younger years and basically all he does is 80% track prep porsches.

The perfect test for a working LSD is a slight downhill fast funky brake zone. Because in this situation the rear of the car is lighter and more likely to get unsettled. You will know instantaneously if it is working - this is the definition of Mosport turn #8. I now actually almost laugh when a gt3 DE participant comes up to me and says, "WTF my car is crazy wiggly in brake zone 8". I then take them out in my car if they would like, brake way later than they do straight as an arrow. I have even had a well known pro driver take me out while setting up a friends 997 cup at the same event and state "this thing brakes better than the cup car" - this is on back to back testing, same day, same track, out one car in the next.

PCCBs with P50s with new R6s and a Guard 50/80 is almost impossible to beat - that is unless you trigger ABS and the ****ty too slow OE ABS computer has a heart attack - but that is another story.

Since it appears Porsche might actually read these forums occasionally here is my wish list:

1) Buy a Guard 50/80 LSD and daily drive it and NRing drive it and you will see there is no downside - then make the next gt3s more like that
2) Fix the ABS module / enhance it to work better on bumpy tracks
3) make the 90L tank standard
4) GTx cars should be geared for the track with a long 6th to drive to the track. People buy GTx cars for the passion of driving - they are not meant to be practical, their single purpose is to deliver joy. Put the cup gear ratios on the street cars with a longer 6th. Don't give us the option, just do it. If the customer doesn't want it they can buy any other of your great 911s.

FFaust

Paul, absolutely spot on with the wish list, especially #2, 3, which we can't do ourselves

Crazy Canuck

Pretty much my experience braking at T2 (Jacques) at Calabogie which is a fast moderately downhill braking zone. Rear end wagging like a dog.

P.J.S.

Spot on

Nizer

What he said.

malmasri

"Since it appears Porsche might actually read these forums occasionally here is my wish list:

1) Buy a Guard 50/80 LSD and daily drive it and NRing drive it and you will see there is no downside - then make the next gt3s more like that
2) Fix the ABS module / enhance it to work better on bumpy tracks
3) make the 90L tank standard
4) GTx cars should be geared for the track with a long 6th to drive to the track. People buy GTx cars for the passion of driving - they are not meant to be practical, their single purpose is to deliver joy. Put the cup gear ratios on the street cars with a longer 6th. Don't give us the option, just do it. If the customer doesn't want it they can buy any other of your great 911s.[/QUOTE]


Since we maybe talking to Porsche I do concur with what he said and would also add a reminder to:
1-BEEF UP the Pressure plate.
2- Go back to the 5 lug nuts if you are unable to use the cup cars center locks on the GTx.

We love these cars and hope that Porsche makes them better

911SLOW

How do you feel about these observations per Roland Kussmaul and Porsche Powertrain Engineers?

I don’t feel anything.

"The mechanical LSD in the GTx streetcars has no preload on the discs. That means, you cannot find out if the LSD works correctly by just jacking the car up and turning the wheels manually! Nothing will happen in that case because the diff isnt working.

It has preload which is rather low even when new.
The LSD is preloaded by the diaphragm springs in the disk sets so that it can be "ready" to work without torque having to be applied. When driving yes the expansion force acts on the disc sets and the static friction moment produced by preloading them is considerably increased.

Correct about the jacking up the car to determine if it’s working, is not a proper test with the exception that when new the oem LSD shows resistance even if that is low. So the wheel test maybe gives a hint..
Proper test is LSD out and measure in the bench.


The reason for having no preload on the system is because the car would be a lot less manouverable in park-in situations and would therefore lose some of its "everyday driver" characteristics.

Correct it has very low preload but who cares about parking. Buy a city car.


The LSD of a Cupcar has preload on the discs and therefore reacts differently on the jack, although the hardware is very similar.


Up to a certain degree that is desirable. It gets trickier in wet and tight turns when preload exceeds 80Nm. Also the car understeers more.
The old cup LSD works great on the street. Newer with more disks and more preload is almost undrivable.


To find out if the LSD is in good working condition choose a long 3rd gear curve (or a big enough space to drive a neutral 3rd gear circle), go with partial throttle and play with the accelerator (more load/less load). The car will change its radius noticeably because of the diff working. (On overrun its 40% and on throttle only 27% lock).

This again is a vague test and not very practical. Workshop manual states to measure in the bench.
And they haven’t changed the procedure yet as far as I can see.
Also the change to 7GT to a different locking percentage probably was made so that the cooperation between the LSD and the electronic aids is harmonic.

On all GTx projects of the past the LSDs where all still within spec after 30.000 km of hard racetrack use. On dissassembly there was no critical wear and tear on the discs or on the ramps."

With all due respect and knowing that AP and Porsche are probably reading all Savvy posts this is not true. Brass plates are shot long before the 30K mark.
Besides Mr. Kussmaul can call his friend Olaf and ask him why he never lets his GT customers out without a cup plates upgrade on their LSDs..


//////////////////////////////
Now what I never understood is why they didn’t do the obvious.
You want low preload ok use 15Nm but put disks in it that would hold. They have the steel cup parts ready.

PS I love Porsche : )

996FLT6

An 80/80 guard LSD is real scary in the rain. Mike

TRAKCAR

I have newer CUP LSD, I'd love it, but they could not sell it std on the car.
It could be part of the hardcore CS, 5 lug, 90L, CUP aero, data/camera, full wrap package..

Porsche or die.

997gt3north

Again, it depends what specs the Guard or any unit is built to. I have no problem whatsoever in wet conditions having a perfectly 'predictable' car. Tire limits and an OE suspension too stiff for a fast 'wet setup' are the limiting factors - but I am speaking to a 50/80 not an 80/80.

As a question of interest, what about the car or your driving style made you want to try 80/80 versus the more standard choices of 40/60 or 50/80. My choice for setting it at 50/80 was that increased stability under braking seemed like a good trade off and that I could dial out the increase accel push with camber, toe, sways, front tire size, etc.