Hello,
I started collecting parts of this project 30 years ago and the last parts have fallen into place.

In this post I hope to clear up some confusion and convey some of what I found trying to complete this project.

First piece
Click to view attachment

purchased form PAR in Yonkers NY 1986 914.332.209.00 casting number 51D

The LSD found its way to me in 2013 from the venerable Cap'n Krusty

Click to view attachment

The diff needed some work as it was missing internal spring plates which are NLA.
Click to view attachment
Item number 4 in this factory illustration.
This part places tension on the clutch pack which is measured as drag.
The factory spec on this is 29 - 58 ft. lbs

This item can be replaced by what is now called a "cup" spring from and early 928
part no. 928.332.555.00
Click to view attachment
seen here as item number 10

As for the internal wear parts, the 914 parts are NLA but, get this they are replaced by mostly 917 parts! WAY COOL.

Just the beginning, stay tuned!
Mike

I didn't realize there was a factory Salisbury diff. Very cool! I designed a custom one for a formula car using BMW plates and it worked incredibly well.

Can't wait to see the rest of your progress!

OK here are a few tidbits about the ZF LSD.

Below is a link to thesamba.com website, lots of good info there too.
ZF LSD Diffs

This image shows all of the LSD diffs that are for 911's '64 - '88 this includes the 914, I thought someone would be interested in the variations.

We are interested in the late 901.
The image shows the external dimensions.

The late 901 ZF LSD which I will call the 914 from here on require a different transmission drive flange from stock. The 914.332.209.00 which has a longer output shaft than the standard flange 914.332.209.10, about 63 mm long, about 5 mm longer than the stock flange.
Click to view attachment

Here is a picture showing that dimension.
Click to view attachment
This image is the 901.332.209.21 it has the same shaft but the flange takes a larger cv joint about 110 mm, this is a good upgrade if power is on your mind.



At this point I have to mention that there are a number of output flanges out there but they don't fit.
There is one for sale on Ebay part number 914.332.209.00/10 it has the correct length shaft but it has 2 flaws.
Click to view attachment
Looks pretty good but looks can be deceiving. The land for the oil seal is 5 mm longer than necessary so this negates the longer shaft. Additionally the shaft itself is almost 2 mm thinner than the 914.332.209.00.

I bought one to see if it was possible to use. I took it to a machine shop to see if it could be modified. The answer is basically no. The flange is hardened steel, so all of the steel lathe tools would bounce off the piece not cutting it, I would have to try a machine shop that specialized in grinding as stones may be able to cut the material, but that still left the 2 mm difference in the shaft, which I am sure, after time would begin to vibrate and shake itself loose. Not a pleasant thought.

I've had one of the drive flanges on my shelf for years and I had to search for another, that was not easy. I searched the web and found a few sites with the part number but each time I ordered I received either a stock flange or some other incorrect number that looked good to someone else ie: 914.332.109.02 This is exactly the same dimensions as the stock unit but I can not find any parts catalog with this number.

So here is the biggest problem with using this LSD the drive flanges are rare as hens teeth. And the /6 guys think they have it tough finding parts.
OK more to come later.
Mike

Maybe you can find a shop that does hard turning?

http://www.mmsonline.com/articles/taking-t...of-hard-turning

I'm lost on the 2mm thinner part of the flange.

Is this something a precision sleeve could take care of?

The 901 332 209 21 flange was used for the early 1969 model year 911/912 with an aluminum case transmission. In the middle of the 1969 model year, Porsche changed to the magnesium case transmission and the axle flange used was 901 332 209 22.
The aluminum case transmissions used an inch size differential carrier bearing with an ID of about 43.xx mm. The magnesium case used a metric size bearing with a 50mm ID. In addition, the magnesium case differential carrier carrier is 11mm wider than for the AL case. (120mm vs 109mm)
Also, the distance between the axle flange faces was the same (260mm) for both the AL and magnesium case transmissions which allowed the transmissions to be switched with no changes to the axles.
The 901 332 209 21 flange has a total length of 110mm. The end that fits into the differential is 57mm long and and 34mm diameter.
The 901 332 209 22 flange is 105mm long. The differential end is 57mm long and 35mm diameter.

This is great information.
I look forward to the following responses.
Thank you

More good info!!!

I decided to inventory my flange collection and provide part numbers and some better dimensions. The attached picture shows 6 flange pairs. From left to right they are:
1. 901 332 209 15/21 used in the early 1969 911 with Aluminum case. Total length 109mm. Length of extension into differential is 56.5mm. Diameter 34mm.
2. 901 332 209 22 used in 911 with magnesium case from mid 1969 to 1971. Total length 105mm. Extension 56.5mm. Diameter 35mm.
3. 914 Flange. Total length 99.5mm. Extension 56.5mm. Diameter 35mm.
4. 901 332 220 915/21 used for mag case 911/915 with LSD. Total length 109mm. extension 61mm. Diameter 35mm.
5. 923 332 209 02 small flange used in 911SC after about 1981. Total length 109mm. Extension 56.5mm. Diameter 35mm.
6. 923 332 209 04/05 used in early 915 transmission. Same as 901 332 209 22 except total length is 106mm.
Then I stuck the flanges into the matching differential carriers and measured the flange to flange spacing. The 901 and 911 were 260mm. The 915 was 262mm and the 914 was 250mm.
The last picture compares the LSD flange to 901 332 209 22. Looks like there is enough metal there to allow the flange to be ground down to fit a LSD differential. Total length of the modified flange would be 4mm less than the original but that should not make a difference.Click to view attachment Click to view attachmentClick to view attachmentClick to view attachment

Hey oldie914 thanks, I was hoping you could help with a question.
I would like to know the correct procedure to prepare a used drive flange to accept a new oil seal?
For that matter is there a prep procedure for installing a new flange?

I have heard mixed comments on this subject, you seem to know more about these flanges than anyone I've meet.


Mike

Thanks for this detailed information / education.
I'm sure you already know, but for those that have not experienced a car with say 60 to 80 % locking factor. It will bring a BIG Smile to your face when you exit a corner with the gas on. Just point and shoot.

Are the 928 and 917 (!!!!!!!) replacement parts available ?

When installing a new seal, clean the rust off the flange with a wire wheel. Check the seal location on the flange with your fingernail to see if there is a groove. You can live with a little wear but not much.
Make sure the new seal is at full depth in the case so the seal is running on clean metal. You can check the seal contact location by smearing a light coat of grease on the flange and putting in in place. Pull it back out and you can see where the seal lip makes contact.
It is amazing the number of different axle flanges Porsche used. There were 4 or 5 additional types used during the 1965-68 model years.
They also tried 3 different methods for securing the differential gear shaft from 1965-70. Finally settled on the block and retaining pin system for the 914 and the 1970 model year 911.

For those desiring a custom flange,I'm sure there are others available.

http://www.bugat5speed.de/en/beetle-co/tra...44-hole-circles

A patient acquiring of the right stuff.
Those are the parts for an original -6 Tx,and as all limited slips,can make wet handling a little twitchy.
A careful setting of the preload can make it work gently in the snow or wet.
Many a 911 has backed into a rail for the same reason.
In dry conditions on race gummies,heck yea.

btw the seal's surface must be polished to a fine finish of < 32 rms,shiny,installed with a lick of grease,any roughness will fail the seal.

911 owners do this more often than not when backing into a gaurdrail

https://en.wikipedia.org/wiki/Lift-off_oversteer

Yes. many times the new 911 owners think they are better drivers than they actually are.... these days, new GT3 and the like owners have the nanny devices to save their rear.

Not to hijack, but I thought this was pertinent:


https://www.youtube.com/watch?v=CvDD8V9Liq8

Thank you !

OK so at this point I though I should get some refernence material on the ZF LSD, web searches for the 914 version brought up little material. I found some references mostly pointing back to the factory repair manual, which is a little behind the times as for replacement parts.

So what I looked for were similar diffs ZF 904 and 915

Here is a rebuild and modification done by a fellow who calls himself Gears on the Samba board.

915 mod

Additionally I used a 904 rebuild as reference also from Gears on the Samba board.

904 rebuild

So with this info in hand I now had a procedure to approach my rebuild.

What to do next?

Every good project should start with an inventory of the parts which complete the job.

Next up I will list the internal parts and their mesurements.
Mike

Yes Porsche Classics have the parts but the are getting VERY pricey.

OK here is the list of internal parts and my measurements.
First I labeled the consumable parts, those consisted of 4 fiber coated disks (#6) and 8 metal disks with external splines (#5 and #4). #4 is the cup spring which is NLA and replaced with the newer style belleville washer 928.332.555.20.
Additionally there are 2 differential shafts (#11) with 4 spider gears (#10) with 2 thrust rings (#7) and 2 bevel shaft gears (#8) with 2 treaded pieces (#9) and 2 circlips (not seen in this diagram), these make up the center of the diff. These items with the cover (#2), LSD body (#12) and 2 thrust washers (#3) complete the unit. Additionally #1 in this diagram is the cover attachment screws (NLA) for the 914 part, substitute the 65-89 911 pan head screw part 900.119.002.00, they are yellow colored not black, so don't let the concourse guys look into your 901 box!

Click to view attachment

So 32 parts total make up this fascinating little gizmo!

I decided to make my measurements in SAE and metric.

Measurements of used parts

Outer spline A .0789 - 2.00 mm
Outer spline B .0785 - 1.99 mm
Outer spline C .0785 - 1.99 mm
Outer spline D .0780 - 1.98 mm
Outer spline E .0784 - 1.98 mm
Outer spline F .0785 - 1.99 mm
Outer spline G .0790 - 2.00 mm
Outer spline H .0795 - 2.02 mm

Fiber disk A .0765 - 1.94 mm
Fiber disk B .0750 - 1.91 mm
Fiber disk C .0775 - 1.96 mm
Fiber disk D .0775 - 1.96 mm


From here I will use SAE measurements, I will introduce metric measurements for illustration.

Than I set about measuring the body, cover and the thrust ring stack
Body internal 3.884 and the cover is .2440, the cover needs to be subtracted from the body because it fits inside so 3.640.
When assembled the thrust ring stack measures 2.682.

So we take inside of the LSD 3.640 and subtract the thrust ring stack 2.682 this leaves us with .958 space for the consumable parts.
There is a total of 12 disks, each group of 6 needs to take up .4790 or 12.17 mm

So there we have the used parts, next post I will measure the new parts.
Mike

Now for the measurements of some new parts.

Belleville washers new part # 928.332.555.00
A .078 - 1.98 mm
B .078 - 1.98 mm

These cup washers have a convex and concave side, we measure the thickness of the material not the thickness of it expanded.

Differential inner disc Fibre new part # 928.332.551.00
A .081 - 2.05 mm
B .081 - 2.05 mm
C .0805 - 2.05 mm
D .081 - 2.05 mm

Out spline new part # 917.332.552.12 Differential inner disc 2.1 mm
A .083 - 2.11 mm
B .0825 - 2.09 mm


At this point I seem to have enough parts for a number of different lock variations.



Now a closer look.
Where do these go wrong? Broken internals, severely wore parts and worn internal channels.



Click to view attachment


Here is a close up.

Click to view attachment

The ridges seen here need a bit of clean up with a small file

Click to view attachment

Click to view attachment

With no major problems time to proceed with figuring out a set up.
Stay tuned.
Mike

What sort of locking factor are you going to set the LSD up at?
Original 40 or pump it up to 60 or 80% ?

The diff looks so similar to the BMW e30 LSDs. IIRC they also use 917 clutch plates which makes me wonder if they are similarly dimensioned. If they are it's be a much cheaper way to build an LSD since you can generally get them for ~$300.

If it's not too much trouble, would you mind getting rough measurements of the ring gear mounting flange OD and overall length?

All of the dimension are seen within this image at the samba.


ZF Diffs


ZF must have designed around what they already had.
Hopefully the need for these parts will keep supply up.

Thank you! Now I just have to find my BMW model and compare...

Any thoughts on my question?

You gave me hope.
I don't feel like such a procrastinator anymore.

Hey 0396
I did decide on 40% lock, going to use her on the street, no track time.
I used a bunch of index cards marked with the internal parts to mix and match to get the correct dimensions for my clutch packs.
I will update but I need to find the P-259 tool or fashion a replacement tool.
Stay tuned.
Mike

Another small step for me, one giant leap for my transmission!



Click to view attachment
P 259 by way of Meaney Racing, possibly used on the Daytona cars # 18 and #19.

A small ring, a few shims and the patience of a saint.

Mike

Is it possible to test a LSD on the bench ?
Can you describe how it is done ?
I want to determine the condition my unit.

Yes factory repair manual has the procedure, with pictures and self fabricated tools.

The ZF LSD unit assembled with drive flanges and two cv bolts are screwed into a flange and locked in a vice, the other flange is turned with a torque wrench and the breakaway point or torque is measured.
There is a specific torque that the factory uses but all bets are off as the orignal parts are NLA and I believe the "cup washers" have more pressure than the orignal "spring plates".
What is being measured is the "locking force" of the clutch packs, lower the breakaway point or torque, the lower the locking force.
Mike

Clamp one side in a vice.

Take a torque wrench, you will get a break away. 10-35 Nmeters is the specs that gets tossed about.

Not sure that will truly give you a full assessment as to the health of the unit. In a "correctly" preloaded unit it would give some indication of clutch plate health, but you can shim it (or change springs) to increase preload. The extra clamping force would mask any issues with the friction surfaces.

Since I am using new friction plates, the unit will be as good as it can be.

Sorry, didn't mean to imply yours won't be good as new. That was in reference to the above question about checking a used unit.