Please note that no specifications are included for the finished lengths of the axle shafts. There may have been many minor changes over the years in the axle housing, differential, spindles, wheel hubs and locking hubs. A particular length that worked here may not be correct for other applications. The Vern-O-Lock certainly changed the equation, too. Determing the finished length needed is not difficult. Assemble the spindles, hubs, etc. for a test fit and use a broomstick to mock up the axleshafts. The spacer between the axle shafts is not needed with a full-floater and may be removed, although this is optional. Slide the broomstick until it hits the spacer, or the pinion cross shaft if the spacer has been removed. Slide the broomstick back out approximately 1/4" for clearance at the inboard end. Now the finished length can be calculated to properly fit the locking hubs. The overall length is not critical with full-floater axleshafts, as long as there is adequate engagement with the side gears without hitting in the middle. If keeping the spacer, remember that it can shift side to side slightly, so unless both axleshafts are mocked up at the time, each one will have to be slightly shorter to compensate.

Here the stock long-side CJ axleshaft is being cut down for reuse on the full-floater short side. It is being cut approximately two inches extra long at this point. That will leave enough for the lathe chuck to grip for the rework process. The long wagon axleshaft didn't have much extra length to spare, so it was not cut down ahead of time like this:

One problem encountered was that each axleshaft is longer than the lathe bed. The lathe's drive spindle is hollow, but not big enough to let the axleshaft slip inside. A new mongo lathe was out of the question, so I used a steady rest to support the end of the axleshaft. The axleshaft is a forging, and it had never been trued where the steady rest must ride. I made a collar to fit over the raw out-of-round and non-concentric middle section of the axleshaft. This collar is known as a Cat Head. ( I'd love to know the story behind that name.) The collar has a smooth outer surface to ride inside the steady rest. The collar has three screws around each end, allowing adjustments so the unsupported end of the shaft runs true. As the axleshaft is turned down, it will be necessary to fine tune the collar to keep the end true. Some stresses are relieved during the cutting process, and the end of the axleshaft follows accordingly:

This next image shows the original long side CJ shaft on the lathe, only slightly longer than the bed. This is the axle which will become the short side full floating axleshaft. The dial indicator will show how much to adjust the screws on the rotating collar for minimal runout. The lathe chuck is turned by hand for these readings. Lacking a dial indicator, you could hold a grease pencil against the same area with the shaft spinning. Marks will be left at the area of greatest runout:

Turning down these axleshafts has been taking a long, long time. If you try to hurry, you will probably get excessive chatter. It is tricky to get a good job turning down a long, slender shaft. It was necessary to move the steady rest and work the shaft in sections. Note how the previous picture shows the steady rest moved to the middle of the bed to help reduce chatter. I must also go on record to state I am not a machinist by any stretch of the imagination. I'm a guy who owns a lathe, and there is a big difference. Since starting this project, a machinist friend has had pity on me. He specially ground some carbide cutters for me. Using these cutters, with various angles and whatnot ground just right, has made a lot of difference in the quality of the cut. I've also learned that a shallow cut can be counterproductive when working on hardened metals. The cutting action slightly work hardens the surface even more. A subsequent shallow cut will be right in the middle of this thin, extra-hard layer. A deeper cut gets below this thin layer where the metal is not so hard. It also amazed me when I learned not to use oil on hardened metals, as this helps the cutter slip past instead of cutting.

Part way through the job I discovered a problem with the rotating collar. When the screws are tightened, the whole thing is distorted slightly into a triangular cross section instead of perfectly round. It showed up on the wear marks on the collar. It may only be a few thousandths out of round, but it was enough to prevent a consistent running clearance within the three jaws of the steady rest. In this picture, I have the adjusting screws securing the collar to a piece of pipe held in the lathe chuck. The cutter is lightly cleaning up the middle section where the collar will ride inside the steady rest. In theory, the collar will return to this round shape when the adjusting screws are tightened on the work piece:

The hardness of the shaft surface varies. The depth of the hardening also varies in proportion to the hardness. The axleshaft with the thicker neck is also harder overall. Without a Rockwell hardness tester, I cannot tell how hard each section is. But judging by the reaction of the cutter on the lathe, the axleshaft appears to be hardened in three distinct ranges. The outer tapered section is of medium hardness and is fairly easy to cut. This includes the area where the stock tapered bearing is pressed on. The raw forging area in the middle has the least additional hardening. The splines that fit inside the differential side gears also appear to have little or no additional hardening, but no rework of them is necessary. The seal land section just inboard of the bearing retainer lip is the hardest of all and presents the most challenges for lathe work. Being furthest from the lathe chuck or steady rest also aggravates the tendency to chatter. This picture shows an uncut axleshaft for comparison with the axleshaft being reworked on the lathe. This axleshaft will become the long axleshaft in the full-floater conversion.

Cutting the splines is an interesting proposition. Look at the following cross section view to get an idea of how the splines were originally cut at the factory. The cutter is tapered 18 degrees on both sides. The depth of this tapered cutter determines the width of the uncut material between each groove. Since a spline cutter like that is out of my budget, a pair of abrasive cut-off wheels came to the rescue. If you look carefully, you will see that the sides are parallel on each uncut section. Note that the shaft has been rotated slightly so the uncut material between each groove is now oriented towards the cutting arbor. The depth of the cut is not critical with this setup. The critical dimension, the width of the uncut sections, is controlled by spacers between the two abrasive discs. Very precise control can be obtained with thin shims added as needed. The material left at the base of each groove will be cleaned up later with a pass of a single abrasive disc. I had originally experimented using High Speed Steel (HSS) jeweler's slotting saws instead of abrasive discs. Unfortunately, the shorter axleshaft is hardened enough overall that it destroyed the cutters almost immediately. Carbide slotting saws are available but are very spendy, and of course I'd need two. It would also be necessary to round the corners of the saw teeth to avoid leaving sharp corners in the grooves. This is not a problem using the abrasive wheels.

Indexing the shaft for each of ten passes was accomplished with a home made index wheel secured to the axleshaft. If you are a professional machinist, I am not responsible if you hurt yourself laughing. It has ten carefully drawn radial lines for rotating the shaft. The image has been touched up to make the lines darker. The index board is a thin piece of plywood clamped upright on the workbench. The alignment of the wheel cannot be disturbed relative to the axleshaft, and the board must remain fixed to the workbench:

Here is the backside of the index wheel. The pipe flange is a common plumbing item with the inside bored to fit the axleshaft. A setscrew locks the flange to the axleshaft. The index wheel is a piece of cardboard secured to the pipe flange with contact cement. The cardboard came from a Cheerios box. You would probably have equal success with a Lucky Charms box, too. I was not willing to experiment with Bran Flakes or anything healthy. If you look closely, there is only a single cut-off wheel in this picture. It was being used to clean up the bottom of each groove:

Here is the Driving Flange from the Warn hub installed on the completed axleshaft. Click here for more details of the Warn hubs. Normally the Driving Flange would not need to be removed from the inner section of the Warn hub. This was done to show how the Driving Flange rides on a roller bearing. Note how the axleshaft splines extend inboard of the Driving Flange: