differential rebuild

[seniorchristo]

Thanks Vxah for the crush washer explanation. It is indeed a twist and is obvious when a straightedge is applied.
Here is a better picture.

[Bill]

FYI

In my S1 parts manual the bearing description is "LJ 30 WSRR Ransom and Marles or R.I.V." (Lotus parts no. ALN 30)

My industrial bearing supplier cross referenced, had lots of Timken equivalent - they were cheap, cheap, at the time. I was told that the bearings common use was for the domestic laundry washing machine drum shaft and they were as common and cheap as borscht.

Colin was a frugal (smart) chap!!

Bill

[vxah]

Ah, yes thats a better picture! You could have a problem with that one!

[Chancer]

FYI

In my S1 parts manual the bearing description is "LJ 30 WSRR Ransom and Marles or R.I.V." (Lotus parts no. ALN 30)

My industrial bearing supplier cross referenced, had lots of Timken equivalent - they were cheap, cheap, at the time. I was told that the bearings common use was for the domestic laundry washing machine drum shaft and they were as common and cheap as borscht.

Colin was a frugal (smart) chap!!

Bill

Lotus used the bearings as fitted to the Ford diff housing, "at the time" they were cheap because of the huge volumes of RWD Ford vehicles being produced and were probably the only bearings whose ID and length were correct. They were and remain to this day a standard bearing size.

"At the time" you bought yours from the bearing supplier, several d?cades later they were still cheap because the washing machine manufacturers had adopted the same standard bearing size, in the 60's a washing machine probably cost more than a second hand Elan

If Lotus had been able to fit something non standard but cheaper from a mass produced consumer item at the time you can be sure that it would be as rare as rocking horse **** today and priced accordingly.

[rgh0]

The bearings in the actual original Ford diff - i.e. the pinion bearings and the diff carrier bearings were Timken taper rollers. They are still available today and Timken is still one of the largest taper roller bearing makers. Lots of other bearing manufacturers make equivalents but their assembly height may vary fractionally between manufacturers which would cause a change in the pinion shim potentially on assembly.

The diff output bearings were a very common industrial single row, sealed bearing that Lotus added when they adapted the Ford diff centre to the Elan alloy housing. This is a SKF 6206-2RS or equivalent and almost all bearing makers make an equivalent and its normally in stock in any bearing shop you go to as its probably the most common bearing in the world !! i dont know if Lotus used the SKF one originally but I suspect they would have used something cheaper and probably British maybe RHP ?

cheers
Rohan

[Chancer]

I have learnt something, I had not thought it through, of course the output bearings needed to be seald ones whereas the Ford ones were not.

I recently bought a bearing and seal kit for an American Maytag washing machine I use for the rental flats, usualy story you have to buy the whole drum assembly which costs as much if not more than a new machine and here in europe its at an even higher price.

I ordered the kit for a great price from an American on line seller that turned out to be a car accessory shop

So automotive and washing machine bearings continue to be interchangeable.

[gjz30075]

I spin dry my car as often as I can

[seniorchristo]

I am finally getting around to completing the rebuild of my 3.55 differential.
To review: I purchased a 3.55 to install in my S3 Elan (currently has a 3.9). When I discovered the input hub had to be changed I decided to look inside. I ended up replacing the pinion and axle bearings, all the seals, and of course a new crush washer. Since the diff bearings looked almost new and I have no accurate way to check cap spread, I decided to leave them alone and hope for the best. the new pinion bearings were .004 shorter than original so I assume pinion depth is somewhat off. What I wanted was a good pattern and proper backlash between crown wheel and pinion. Pinion bearing preload was set to 10 inch pounds. After torqueing the differential, I have an excellent pattern, centered between face and flank and a backlash of .008-.009 inches. This is out of spec by a couple thousandths. I can adjust the backlash at risk of altering the pattern. Does anyone know how much backlash is affected for each movement of an adjustment hole? Is .008-.009" excessive play even though it is out of spec? If I remember correctly I think the diff had about .015" when I disassembled it. The 3.9 in my car feels like it has a quarter inch

[Chancer]

I had to change my washing machine bearings once again this week, guess what, the smaller of the two is exactly the same reference number as the Elan!

The last ones only lasted a year and 257 wash cycles, it might have been bad luck but for an Elan I would only use branded SKF, Timken, NSK or INA, the saving on cheaper ones might be a false economy, there are plenty of other thngs like donuts to occupy your time on an Elan.

[promotor]

Does anyone know how much backlash is affected for each movement of an adjustment hole? Is .008-.009" excessive play even though it is out of spec? If I remember correctly I think the diff had about .015" when I disassembled it. The 3.9 in my car feels like it has a quarter inch

How much the backlash is affected by one hole movement depends on the resistance of the carrier to be moved over by the current cap-spread (ie side bearing pre-load) - if you already have a lot of preload one hole of movement will make less of a difference than if you have minimal preload due to the caps' increasing resistance to being spread apart. There is no set published figure due to it not being a straighforward calculation.

Also it will be affected by the shape of the teeth profile and how worn they are and where that wear is : there may be a low spot in the teeth just at the point you are trying to move the crownwheel over.

Although the 0.008-0.009" isn't excessive at all I would attempt to lower the backlash figure and see what you get - you can always go back to the slightly bigger figure. Don't forget to check the backlash at different points (I check at least 10).

[rgh0]

Measuring differential bearing cap spread is not that hard.
I reverse the bearing adjuster locking tabs and use a micrometer across them. You can do the same with a good caliper. You could also setup a dial gauge with a couple of purpose built brackets using the same bolts that secure the locking tabs.

cheers
Rohan

[seniorchristo]

All very good advice, thanks! Searching the archives, someone suggested each movement of an adjustment hole equals about .002" in crown wheel and pinion backlash although I'm sure this is dependent on factors mentioned by promotor.

[seniorchristo]

Ok, so I adjusted the side adjusters one hole to move the crown wheel closer to the pinion. The backlash went from about .008' to .0025. This being outside the .005 to .007" specs, I moved the crown wheel back to its original position and this time got an average of .007" measured at eight different locations on the crown wheel. I found this acceptable so went to button everything up. Since the pinion preload was set without the pinion oil seal, I removed the pinion nut, installed the seal and installed a new distorted thread lock nut. I torqued it to about 35 ft. lbs., and noticed all the preload was gone. I could spin the pinion flange with one finger. I had to impact the nut down again to achieve the required preload (10 inch lbs.). Could the crush washer have sprung back a bit after the nut was backed off? Also I notice there is almost no resistance in turning just the crown wheel. Shouldn't the differential bearing preload offer similar resistance as do the pinion bearings?

[seniorchristo]

I finished the diff assembly today. I bought some 8" calipers from Harbor Freight to measure the cap spread. Not the most accurate method but after about 30 measurements I got a pretty good ballpark average. Checking crown wheel and pinion backlash was also a challenge because of the angularity my dial indicator setup gave. Once everything was pretty well set, I found that the diff adjusting nut positions had to be compromised in order to secure the locktabs. I have spent so much time fiddling with this diff that I think I can guestimate preload and backlash almost by feel Since the pattern again looked ok I will be buttoning things up soon and then swapping it out with my noisy 3.9. I'm looking forward to the relaxed RPM's of the 3.55 and no whining. Thanks to Rohan and others who guided me through this process. I'm ready to rebuild another one!

[seniorchristo]

I installed the rebuilt 3.55 differential this past weekend. It was a difficult job, mostly because I was lying under the car, rolling around on my creeper losing tools and getting really greasy. Getting the diff in and out was a snap. After the horror stories I figured it would be a tough task. However after raising the flange end up it just rolled out After getting the car back together I drove over to visit Glen and Drew in Princeton (about 70 miles) It was a beautiful thing Smooth, quiet and much more relaxed cruising. An added benefit is that now speedometer accuracy is nearly spot on. I tested it against a speedo app which uses gps. Now that i'm a diff pro my intent was to rebuild the old noisy 3.9, just for grins However after seeing how butchered up the aluminum housing is I am having second thoughts. The mounting holes are elongated and the ears look like they had broke off and were rewelded. Thank you Rohan for walking me through the rebuild process! Difficult but rewarding. I also added a little soundproofing thanks to QMM and this excellent thread.lotus-gearbox-f37/sound-proofing-the-differential-t26887.html
Chris