9 inch Ford differential in 1/3 scale

[gbritnell]

Gentlemen,

Over the years of attending model engineering show I was repeatedly asked "when are you going to make a car for your 302 engine?" At that point I really never gave it any thought. When I finished the T-5 transmission and would show it with the engine I got more of the same questions. This past year saw the building of the final piece of the driveline, the 9 inch Ford differential. It was at this point that the original question was given some serious consideration. Hopefully there will be enough time in the future to complete a full chassis, running gear, brakes, suspension etc.

So with that introduction I will present to you the construction of the 9 inch Ford differential.

The first thing I had to determine was how I was going to make the ring and pinion gears. If I couldn't machine them then the project wouldn't go any farther. From my automotive background I knew that ring and pinion gears are in the class of hypoids, offset, curved tooth gears. But how to machine them. I turned to the internet and mainly Youtube to see how these gears are made. After lengthy investigation I realized that they were almost if not totally impossible to cut in the home shop. At this point an alternative type of gear but with similar characteristics was investigated. A series of build articles for constructing geared locomotives by a very talented engineer/machinist, Kozo Hiroako, for Live Steam magazine showed how to make skew bevel gears. A normal bevel gear set is where both pinion and ring gears have the same centerline but skew bevels have offset centerlines. Using all of the algebraic equations that the author had presented I tried to come up with pitches, tooth counts and offsets that would let me fit them into the dimensions taken from the full sized differential. The first stumbling block was my math knowledge (algebra) was a little rusty but with the help of some great people I got that issue resolved. I was even offered and given a spread sheet so that all I had to do was plug in numbers and the calculations would be generated. Whew! Once I got the numbers I needed the next step was to make some cutters and try out the equations. These were made from drill rod and hardened. I had never made this type of cutter before. The blanks were offset to give clearance to the cutting edge.

- - - Updated - - -


With the cutters and arbor finished the first task was to cut the gears from Corian to prove that everything would work. The dividing head was set up on the mill, the blanks were roughed out and mounted and the tooth cutting started. In the calculations for creating the tooth profile the cutter will only cut one side of the tooth form and then needs to be reversed to cut the other side. Along with this the blank needs to be offset from one position to another and the cutter centerline needs to be adjusted to create the correct profile and width.
As you can see the plastic set came out quite nicely.

 

[gbritnell]

Here's a short video of the finished plastic gear running together.
http://youtu.be/up1-lQm_32c

- - - Updated - - -

Next up was the metal gears. I made the ring gear from 12L14 steel and the pinion from brass with the idea that if one or the other was going to wear it would probably be the pinion and it would be easier to remake.

- - - Updated - - -

Another short video of the metal gears running together. I made up a small right angle fixture with the required offsets to test the gears. I really didn't want to chuck them up in the small lathe as I had done with the plastic gears. There is a little gear noise because I didn't have them shimmed to exactly the right mesh position.
http://youtu.be/fW5wzl9lGaM

 

[gbritnell]

With the gears completed and running together reasonably well it was time to move on to the main case. This was machined from a piece of 6061 aluminum. As with all of my billet creations the inside was cut out first so that there was always some square stock to chuck or fixture to. The part was machined to the overall dimensions and then given layout lines as a rough guide for machining. I always pick up the sides of the block with an edge finder and then use those points and my readouts to work from. The most important thing is doing work like this is to know exactly what your next step will be. It's like the old saying about painting yourself into a corner.
gbritnell

- - - Updated - - -

At this point the housing was clamped to an angle plate to bore for the ring gear bearings. The centerlines were wiggled and the holes were roughed and finish bored. The part was then put back in the vise and the ring gear clearance was stepped out using a sine/cosine step-off chart created in Autocad.
gbritnell

 

[gbritnell]

The inside is finished for now so it's time to move on to the outside. With some layout lines to give me a visual reference to cutting I start out by putting the pinion mounting holes in and boring for the pinion housing. The next steps are to start removing material and hopefully leaving the good bits. As with the transmission project I make up numerous step off charts to cut the radii with ball nosed endmills. There are times when I use my rotary table for such work but usually for the time it takes to remove the vise, set up the rotary table, fixture or clamp the part, pick up centers and cut it, it takes less time to step it off. I have been doing this type of work most of my machining life so to me it seems quite simple.
gbritnell

 

[gbritnell]

The idea at this point is to work around the housing cutting the features, ribs, bosses and such. Although the steps look quite large in the pictures I generally try to make the cusps (height of the crest between the cuts) no more than .003. The steps could be made smaller but it's a trade-off between spending the time cranking the mill or working the cusps down with burrs and stones. To finish the part requires hand work so a couple of thousands here and there doesn't make much difference. I cover everything with marker ink and then start working the machining marks down. When you get them close all you see is a thin blue line. I use a Dremel with a flexible shaft for all my hand grinding. I then work out corners and areas that I can't get into with the burrs with riffler files. The final smoothing is done with varying grades of emery paper, some wrapped around dowels and some glued to wooden sticks with different shapes.
gbritnell

 

[gbritnell]

I continued to work around the housing, getting into corners with files and sanding stick until everything was smoothed out. Aluminum is quite unforgiving once you start to smooth it out. It shows any little nicks or irregularities so you have to take your time it you want a nice smooth surface.
With most of the housing polished I had to make a fixture plate to do some additional machining. On this particular housing the pinion area has a recess underneath it which couldn't be cut when I was doing the other surfaces. It's there to provide access to the 2 lower mounting studs and nuts. With the housing set up on the plate in the vise I put a hardened dowel into the pinion rear bearing hole to that I could pick up the center. There weren't any more flat surfaces to gauge from at this point. The area was stepped out and then finished.
gbritnell

 

[gbritnell]

The following pictures are of the finished main housing. A little bit more work needs to be done to finish it completely but that will be presented in some future postings. I have included a quarter in one of the pictures for a size comparison. I'm guessing that this piece has about 70-85 hours from start to finish.
gbritnell

 

[xalky]

Thats a beautiful job. Thanks for posting the pics and the method. This is not something that i would undertake, mainly because I'm not a model builder, but i certainly appreciate all the hard work and head scratching that went into it. :goodjob2:

 

[gbritnell]

The next part is the pinion housing. This is also made from 6061 aluminum. The part is mainly circular so the diameters were turned and extra stock was left on the outside of the flange to shape the bolt bosses. From the lathe it went to the dividing head to start cutting the shapes of the inside with the ribs and the outside with the bolt pads. I used a small brass plug as my machining guide. As I rotated the part I would bump almost up to the plug to give myself witness cuts to form the circular shape on the ends of the pads. With the usual burrs, stones and files the part was finished and bolted to the main housing.
gbritnell

 

[gbritnell]

The next operation is the backtracking that I had mentioned earlier. The main case was finished except the threading needed for the bearing retainer/adjuster collars. I turned up a bar .001 larger than the bearing caps on the housing, loosened the caps and slid the housing on the bar. The housing was positioned so that I had about .375 of depth to put the threads in and the caps were tightened. The threads are 48 TPI. because there is only .300 of working depth to allow for the bearings. The caps were turned and threaded then moved to a fixture on the dividing head to put the locking slots in.
gbritnell