KeithB
Resident Half Fast Machinist
One of my customers ordered some bearing housings. It was fairly conventional machine work but had a little twist at the end that I thought I would share.
The part is made in four steps:
1. The material is 2011 aluminum round stock 1-3/4" diameter. It was shipped in three 4' lengths so all we had to do was put the outboard support bushing and the three jaw chuck on the CNC lathe. The first operation was to face the end flat, turn the largest diameter to 1.625" and the hub on one side to .875", drill a 5/8" hole, and cut it off with a cutoff tool. With the gang tool holder setup I've written about here we were able to do all the machining without changing tools.
2. We switched out the three jaw and put on the air collet and a 3" face collet. We bored out the collet so the part would be grabbed by the largest diameter and the face of the flange is up against a solid surface. The flange is only .100" thick so we only bored it .09" deep. We were cutting within .010" of the collet face. We also bored out the center to clear the hub and the boring tool that we needed to use to bore the ID. This only took two tools and again they were mounted rigidly and not changed out for the entire production run. By doing this we were able to maintain the bore diameters to within .0006" for the entire run.
3. Over to the CNC mill, where the face collet was pulled from the lathe and used in a manual collet closer mounted to the machine table. We were able to use a #4 center drill to drill the 12 hole bolt circle; the tip was long enough to go all the way through the flange and by controlling the depth we were able to chamfer one side of the hole with the 60* taper of the center drill. (We used a #3 center drill in a cordless drill to chamfer the other side of the holes between operations.)
It was basic machine work to this point. The next step required crimping in the two roller bearings, which are .75" OD, .25" ID, and .281" thick. Our first try was a die that looked a lot like a top punch in a sizing press, shaped with a ring that would roll the edge over. The results were not satisfactory. Our next attempt was with a home-made crimping roller.
4. Using the manual collet chuck on the manual lathe, the part was mounted in a 7/8" collet so that the flange was up against the collet face. The crimping tool was mounted in a dovetail tool block and adjusted for height and aligned straight in the Z (long) axis. With the lathe running at 60 RPM the tool was fed into the part and VIOLA! after about 10 revs and relatively light pressure the edge was rolled over very nicely.
The apprentice did fifty pieces in a little over an hour.
I made the toolholder wider than necessary because I wanted to be able to install a spring loaded plunger to push the bearings back into the housing while crimping, but it wasn't needed. The good bearing fit retained the bearings with no problems.
If we get another batch I plan to make the wheel from some hardenable material and polish the working surface better. But I know what works now and can afford to invest the time improving things.
The part is made in four steps:
1. The material is 2011 aluminum round stock 1-3/4" diameter. It was shipped in three 4' lengths so all we had to do was put the outboard support bushing and the three jaw chuck on the CNC lathe. The first operation was to face the end flat, turn the largest diameter to 1.625" and the hub on one side to .875", drill a 5/8" hole, and cut it off with a cutoff tool. With the gang tool holder setup I've written about here we were able to do all the machining without changing tools.
2. We switched out the three jaw and put on the air collet and a 3" face collet. We bored out the collet so the part would be grabbed by the largest diameter and the face of the flange is up against a solid surface. The flange is only .100" thick so we only bored it .09" deep. We were cutting within .010" of the collet face. We also bored out the center to clear the hub and the boring tool that we needed to use to bore the ID. This only took two tools and again they were mounted rigidly and not changed out for the entire production run. By doing this we were able to maintain the bore diameters to within .0006" for the entire run.
3. Over to the CNC mill, where the face collet was pulled from the lathe and used in a manual collet closer mounted to the machine table. We were able to use a #4 center drill to drill the 12 hole bolt circle; the tip was long enough to go all the way through the flange and by controlling the depth we were able to chamfer one side of the hole with the 60* taper of the center drill. (We used a #3 center drill in a cordless drill to chamfer the other side of the holes between operations.)
It was basic machine work to this point. The next step required crimping in the two roller bearings, which are .75" OD, .25" ID, and .281" thick. Our first try was a die that looked a lot like a top punch in a sizing press, shaped with a ring that would roll the edge over. The results were not satisfactory. Our next attempt was with a home-made crimping roller.
4. Using the manual collet chuck on the manual lathe, the part was mounted in a 7/8" collet so that the flange was up against the collet face. The crimping tool was mounted in a dovetail tool block and adjusted for height and aligned straight in the Z (long) axis. With the lathe running at 60 RPM the tool was fed into the part and VIOLA! after about 10 revs and relatively light pressure the edge was rolled over very nicely.
The apprentice did fifty pieces in a little over an hour.
I made the toolholder wider than necessary because I wanted to be able to install a spring loaded plunger to push the bearings back into the housing while crimping, but it wasn't needed. The good bearing fit retained the bearings with no problems.
If we get another batch I plan to make the wheel from some hardenable material and polish the working surface better. But I know what works now and can afford to invest the time improving things.
