Escape The Room  2          4-2017


Machining The Gears

 

The first order of business was to cut off the scale from the outer surface of the aluminum. Now most metals have a thin 'scale' that might need to be removed depending on what you're doing (aluminum or steel). For instance, if you were going to weld on it then the scale isn't something you'd want because it would contaminate the weld. However, in my case I'm going to send all these gears to the plating shop to be anodized. And the type of anodize will be Type III, which is hard anodize or hard coat anodize. The reason for anodizing is two-fold: first a lot of people will be handling these gears and being hard anodized will take the abuse of people bumping them around much better. Second is because hard anodize will let them 'wear' much longer than just being raw aluminum. And hard anodize is somewhat slippery which means the gears will work together much easier.

One more thing about hard anodize, this type of plating goes into the material .001" and builds up on the outside the same amount, .001". In total that equals .002 of plating buildup (approximately the same thickness of a human hair). I'll revisit this process later because this is an important fact.

This machining process is an easy one, but the way I have to do it ends up being a very lengthily one. I'll explain: as you can see I'm holding the gear with two clamps. Once it's held in place I can machine the entire top surface, except the area held by the clamps. Once that's done, I remove both clamps, clean the part and the milling table surface and then re-clamp on the machined surface now. At this point I can cut the area that I was previously clamping on so the entire surface ends up nice and flat. Now I'm only removing a few thousandths from each surface, in other words a minimum cleanup. But even if I were taking a large cut in place of this small cut, I'd still have to do the clamp swap thing. Total time cutting all the gears was just over eight hours. Like I said, very time consuming.
 

 

 

With all the gears looking pretty good after being surfaced, they now had razor sharp edges so they needed to be deburred. I came up with a couple of ideas for this deburring process but using a Dremel was the best. I used a carbide burr with a rounded tip at the end to go around each tooth. The reason for the rounded tip is because it's more forgiving if you bump into an area by accident. I tried using a burr with a square end and this is why I switched to one with the rounded end.

This was another easy process but again a very lengthy one. I would start at the top of the tooth (the small flat area) and I did this from the front of the gear first. Then I'd move to my right a step or two and lean over slightly so I could get to the area that you see below. In this position I could do the long curved part of the tooth and the bottom area that you see me doing in the picture.
 

 

 

Once I had those two areas done I'd move back to the first position again and do the adjacent tooth that you see in the picture below. However what you can't see is 'how' I did this edge break. What I mean is, I didn't just keep my Dremel at the same angle and run across the surface. What I ended up doing was starting at a steep angle for my first pass, then moving to a lesser angle for my second pass, to a lesser angle again for my third pass, to a much lesser angle for the forth pass and then close to where I started for my fifth and final pass.

I can only imagine what you must be thinking after reading about this procedure. The method to my madness is that I wanted the gear to be free of burrs and if I kept the Dremel in one position while doing this, it would push a burr on either side of the angled cut when I was done. But doing it the multi-angled way, there is no burr and it actually has a slight radius now. Think of the cuts as five small facets very close together that produces a curve. That's what I ended up with and they all looked great. The bad news is this process took a very long time. How long you ask? Keep reading....
 

 

 

The numbers that you see on each gear is in minutes it took to deburr each one. Note that the larger one with the holes went quicker than the smaller one next to it that has triangular windows. The one with the windows are a little harder to do because those have a little more surface area than the holes.
 

 

 

As the tooth count increases, so does the time it takes to deburr. The one on the left is the largest gear at 12" pitch diameter. Total time deburring all of the gears was 10 1/2 hours.
 

 

 

Now that all the gears are easy to handle (no burrs), it's time to do more machining on them. When I was in the planning stages, one of the things I decided on was to use .500" diameter (1/2) shafts for all the gears to spin on. With this number in mind I asked Scott at R & I Industries what kind of tolerance they could hold with their water jet machine? The reason I asked him is because I needed to know if he could put the exact size hole in the gears that I needed. He told me that he could hold about .010 thousandths of an inch. Well that wasn't going to work for me so I drew all the gears in my CAD system with the holes undersize, knowing later I'd open them all up to the size I needed.

In this machining operation I'll be reaming all the gears to my final size of .503. I'll explain this odd size later. And to aid me in holding them all in the same place, I'll be making a fixture. This fixture will be made out of a scrap piece of wood, a 2 X 8 Douglas fir to be exact. Wood tooling works great for a job like this, and it's easy and cheap to use. What I'm doing is taking a clean-up cut on the top and bottom surface to make sure it sits flat and parallel before I start machining.

In case you were wondering what all those holes were for, I used this same piece of wood for separating all my valve train components while working on my 33 Coupe awhile back.
 

 

 

I clamped the freshly cut wood tooling down to my mill table and then drilled and bored two different size holes. You're probably wondering why I have two sizes? Keep reading.
 

 

 

The gears that you see below had to be made over again by R & I Industries and the reason is because the holes were off center. I noticed this problem the day Angela and I picked them up because I looked at each and every gear before we went home. And for some reason only these two sizes had this off center problem. Scott was very sorry and didn't know what happened either but he had no problem making them over and then shipped them to me.

However, the holes in these new gears were different than all the others so this is why I have two different sizes in my wood tooling.
 

 

 

Here you can see the different size pins I'll be using along with my fancy marking system to let me know which gear goes where. BTW, the replacement gears have the larger holes.
 

 

 

A keen eye will notice that I've turned my wood tooling 90 degrees. What I'm doing is indicating my gage pin so it will be in center of the spindle.
 

 

 

Here you can see one of the gears clamped down and ready to be machined. Here is how it works: I place the gear on the wood, put the pin through the gear and into the wood tooling, clamp the gear down, remove the pin and then ream the hole. This makes it so much easier to center each part and is well worth the time doing so. Yes I could have floated the reamer through each hole but I wanted to have each part clamped while reaming. Why you ask? Sometimes while reaming parts, they tend to climb the reamer while it's turning and I didn't want that to happen.

Note the paper between the parts and clamps. This prevents any nicks or dings being transferred into the part.
 

 

1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23