VCCA Home Forums TECHNICAL - 1945 AND EARLIER 1933-1936 1936 Chevy 3.55 rear gear conversion

The following is an article written by Ray Waldbaum.



Since I completed restoration of my 1936 Chevy pickup in 1976 I’ve wondered what it would take to get a numerically lower rear axle gear ratio than the factory installed 4.11.
In decades of membership in the Vintage Chevrolet Club of America (VCCA) I put that question to every other VCCA member I met. In response I got either a blank stare or “That can’t be done”.

Finally, in 2009 I quit the VCCA after over 40 years of membership because the hands on approach to the car hobby that I enjoy was no longer present there. In looking for another interesting group I found stovebolt.com and 1936 Chevy Owners. On stovbolt.com I found the potential answer to my gear ratio change quest. It turns out that a 3.55 gear set was used in 1950-54 powerglide equipped Chevy cars. The bad news was that the 3.55 gear set is hypoid and Chevy cars through ’36 and Chevy trucks through ’39 use spiral bevel gears. The good news was that Chevy car rear axle internals, exclusive of axle shafts and driveshafts, are interchangeable in the 1937-54 span. I also learned on stovebolt.com that a ’37 Chevy car rear axle is the same width as the ’36 Master car and pickup unit.

Based on that, it looked like a ’37 Chevy car rear axle/torque tube and driveshaft with the 3.55 powerglide gears installed might be the answer. The next step was to get the pieces and see if the ’50-’54 gears were a fit in the ’37 axle housing. The required pieces, a complete ’37 car rear end with spring perches and a complete ’51 powerglide rear end were rounded up from street rod builders for scrap iron prices, because that’s what that stuff is to street rod builders, scrap iron.

Comparison of the ’36 pickup and ’37 car rear ends and disassembly of the ’37 and ”˜51 rear ends disclosed the following:
1. The ’36 and ’37 rear ends are the same width (“track”)
2 The ’36 and ’37 u-joint rear pieces have the same splines. Thus, the ’36 u-joint will fit the ’37 driveshaft.
3. The different distances between the leaf springs require that the ’37 rotating spring perches be moved inboard 1 7/8” per side. This procedure is described below.
4. The ’37 driveshaft and torque tube have to be shortened 6 ½”. This procedure is also described below.
5. The ’37 pinion gear shaft has 10 splines and the ’40-’54 pinion gear shafts have 17 splines. Dealing with this is also described below.

My plan was to keep my ’36 pickup intact and running while the replacement rear end was being prepared. In order to be sure it would actually fit after all the cutting and welding was done, I measured the spring perch locations and the exact driveshaft and torque tube length on the ’36 pickup. That required splitting the u-joint to expose the front of the driveshaft and measuring the distance between the front of the driveshaft and the axis of the axle shafts. I took this extra cautious approach because I could see that there was some variation in the distances between the rear cover plate and third member mounting surfaces on the ’36, ’37 and ’51 rear axle housings, probably due to the crude manufacturing equipment of long ago. Because of that variation, the actual axle shaft axis centerlines seemed to be the most reliable reference points for driveshaft and torque tube length measurements. The perfect fit of the finished product proved that cautious approach to be a wise one.

STEP 1, MOVE SPRING PERCHES INBOARD

The spring perches present 2 challenges. First, their positions on the axle housing is set by a piece of curved flat bar riveted to the axle housing. This prevents inboard-outboard movement of the spring perches and allows only rotation of the spring perches. These pieces of curved flat bar have to be moved inboard 1 7/8” per side. I have no way to buck a rivet head inside the axle housing so I welded the flat bars in the new position. The welds were made at the former rivet locations on the flat bars and the rivet holes in the axle housing were plugged by welding.

The second challenge is creating a machined surface for the spring perch to rotate on. When new, the spring perch has a round bore and a round machined surface on the axle housing is the mating surface. On used parts, the spring perch bore is egg shaped so chucking the whole rear axle housing in a huge lathe to machine a round surface is not the answer because the round housing would not be a good fit in the egg shaped spring perch. Also, a really big lathe would be required to accommodate a rear axle housing.

The only solution to this problem that I could think of was to use a hand file to duplicate the shape of the worn axle housing 1 7/8” inboard of the original location. Doing that precisely required a lot of trial fitting to find and remove the high spots. This was the only labor intensive part of the whole project. Here is a photo of the inboard-relocated, welded on flat bars and the hand filed bearing surfaces on the axle housing. That took a lot of hand filing and trial fitting, but what was the alternative?



STEP 2, TORQUE TUBE SEAL AND BUSHINGS

Before doing any cutting to shorten the driveshaft and torque tube, I pulled the seal and front and rear torque tube bushings. Aftermarket suppliers would like for you to believe that a special tool is required to do that job and they would also like for you to believe that a correctly sized seal is available from them. Neither is true. I removed the seal and both bushings with a piece of allthread and some large washers fed into the torque tube from the third member end with a piece of ½” PVC pipe. The set up is shown in the photo below. A short piece of 2 ½” exhaust pipe was placed on the front of the torque tube to create a void for the bushings and seals to be pulled into. This would have been even easier after I cut off the front of the torque in order to remove a piece to shorten it.





With regard to the seal, the ID must be the shaft size, 1.031” (1 1/32”). The seal offered by all the aftermarket suppliers has an ID of 1.062” (1 1/16”). I checked with all of them, including Chevies of the 40s and The Filling Station. They were very resistant to me asking for seal dimensions but when they finally, reluctantly answered the question it was obvious why. They are selling the wrong seal for the application! A 1 1/16” seal is simply wrong for a 1 1/32” shaft.

The OEM seal was a special size made for GM only and any that might remain this many decades later would be useless because the rubber would be so hardened as to be completely incapable of sealing. To find the solution to this problem I called tech support at Timken Bearing. There, a representative asked for the shaft and bore size then recommended a seal to solve the problem. It is Timken seal number 471076. It is a double lip seal with the correct ID and an undersize OD. I made a sleeve to increase the seal OD, see photos below of the machining process and the completed sleeve alongside the original seal, and that took care of the seal challenge.





The front and rear bushings were OEM pieces found on ebay, not from professional vendors with high prices but from hobbyists, like us, who had them left over from their own projects. That took care of the seal and bushing search. The seal and bushings were installed using shop manual instructions.


STEP 3, SHORTEN DRIVESHAFT AND TORQUE TUBE

The ’37 pinion gear was easily removed from the ’37 driveshaft. It slid right out. The ’51 driveshaft that had the 3.55 pinion gear was another matter. The date stamped on the ring gear and pinion indicated that it was made on a Monday. That can be bad on an automotive assembly line and it was bad. The pinion gear would not come out of the driveshaft, despite a lot of persuasion. Oh no!

The solution I thought of was to cut the ’51 driveshaft right in front of the splined socket and then use a large puller to pull the socket and driveshaft stub off the pinion gear. Photos of that procedure are below and, as the pictures show, it worked.

Examination of the splines on the pinion gear disclosed the source of the sticking problem. The splines were crudely cut and had a lot of burrs. Those burrs were sticking in the female splines in the socket. Whoever did that job on that Monday in 1951 probably had a gnarly hangover from a rough weekend.

The solution was to very carefully remove the burrs with a die grinder using mini abrasive discs. After the burrs were removed, the pinion gear was a snug slip fit like is should have been in the first place and the project moved forward.

The torque tube and driveshaft cutting was done using an abrasive disc that I was able to find at an industrial supply store that would fit my wood cutting miter saw, same arbor hole size and same diameter. Pretty lucky, huh?

Careful measurement and comparison of the ’36 and ’37 driveshaft and torque tube lengths showed a difference of 6 ½”. Because the ’37 pinion gear has a different spline count from the ’40-’54 pinion gear (10 vs. 17) the 17 spline female socket was removed from the ’51 driveshaft by machining out the weld. The 10 spline female socket was then removed from the ’37 driveshaft, also by machining out the weld. The OD of the ’51 socket was then machined down to the exact same OD as the ’37 socket for insertion and welding into the shortened ’37 driveshaft.

The ’37 driveshaft was shortened 6 ½” so that it would match the length the ’36 driveshaft once the splined female socket was welded in. A piece of the ’37 torque tube was cut out of the front part of the torque tube to make it the same length as the ’36 torque tube once the cut ends were welded together.

This shortening and welding required keeping the parts in alignment before and during the welding process. Don’t underestimate how much welding warps metal! I made simple jigs to align the torque tube and to check the alignment of the driveshaft before tack welding the splined socket into the back of the driveshaft..

The torque tube aligning “jig” was 3 pieces of angle iron and a collar I made to temporarily increase the necked down front of the torque tube to the larger size of the rest of the tube. This not so fancy “jig” was held in place by good old low tech wire. This is shown in the photo below.




The “jig to check alignment of the driveshaft to make sure the splined socket was inserted squarely was a length of aluminum channel (actually my off road race motorcycle loading ramp), a couple of pieces of 2 x 4 lumber, 4 16 penny nails and a dial indicator. This Rube Goldberg arrangement is shown in the photo below.



The 17-spline socket removed from the ’51 Chevy driveshaft and reduced in diameter is shown below welded into the rear of the ’37 Chevy driveshaft. I made this weld by rotating the workpiece while holding the welding electrode stationary. That did not work out as I hoped and caused some warping. I was able to straighten the part but if I had the job to do again I would weld on alternate sides to cancel the heat distortion. Oh well, I’m a geologist, not a welder.




STEP 4, RING GEAR AND PINION INSTALLATION

This is not a magic trick. There are plenty of articles available on how to do it and it is easily doable since the Chevy set up does not use a crush sleeve. Two adjustments have to be made; the fore-aft depth of engagement of the pinion into the ring gear that is controlled by shims in front of the front pinion bearing and the gear clearance (“backlash”) that is controlled by side to side adjustment of the carrier bearing adjustment nuts.

The fore-aft adjustment is made by varying the shim pack thickness and checking the gear tooth contact pattern with oil paint from an artist supply store. I used a color called “titanium white”. The shims vary in thickness from about 0.016” to 0.022” and 2 shims are used. I got an assortment of these shims from the 2 rear ends I disassembled and from other hobbyists who had them collecting dust in their garages. The side to side ring gear adjustment is made by simply turning the adjustment nuts and maintaining the shop manual specified preload. Again, this is not magic.

STEP 5, INSTALL WHEEL BEARING SEALS AND BRAKES

Wheel bearing seals are another item that does not require an antique car aftermarket supplier. The correct seal is a readily available SKF 15557.

Next was transferring the brakes from the ’36 rear axle to the ’37 rear axle, a direct bolt on. Because the ’37 spring perches were moved inboard, the original brake hard lines do not fit properly because they interfere with the relocated spring perches. New brake lines were made from auto parts store brake lines using a simple, inexpensive lever bender and an equally simple and inexpensive double flaring tool. The job was then complete except for actually installing the modified ’37 rear end into the ’36 pickup.

The speedometer/odometer then ran 17% slow and that was problem handled by a 17% overdrive adaptor from Mr. Speedometer in Canada. His adaptor goes on the back of the speedometer head. Here is what it looks like.



That’s it! Any questions?

Great details and very well written!

Thanks, Dean

Makes me wish mine was a 36

53 Belair rear and torque tube listed in "Parts for Sale" 3:70 gears.

good job

Good morning Ring.

I'm glad you liked my 3.55 rear gear story. Where in Nevada are you? Reno by any chance? If so I'm just west of you along Hwy 80 in Sonoma County CA.

Since doing that 3.55 job I've hidden a tandem master cylinder in the stock location on my '36 PU as a safety upgrade. The only thing that gives away its presence is 2 brake lines exiting the MC area under the floor.

Ray W