This was how I started this blog post. I started writing it when I was about 80% done with the project. Guess I jinxed myself because after I was finished, it turned out to be a heartbreaking failure! I knew it was too good to be true. Figure I'll post about it anyway just to keep me humble.
A few weeks ago I posted about ultrasonic case cleaning for reloading. It worked great for dirty cases but left a little to be desired for really old, grungy, tarnished cases. Then I found this video for cleaning brass with stainless steel pins in a rotary tumbler. After reading some more about it online, I really wanted to try it. Unfortunately, it requires a rotary tumbler that I didn't have and didn't feel like shelling out about $180 to purchase. I studied the design and decided it was probably something I could make myself.
Thumler Model B rotary tumbler |
Designed for rock polishing, the Thumler Model B consists of a water-tight drum spun on some rollers at 40 rpm by an electric motor. Looked simple enough. I had a 1.5 gallon plastic cookie jar I could use as the drum, some feed rollers from an old printer, and plenty of electric motors. With the parts on hand, I set to work.
steel equipment case |
I had an old steel equipment case that I thought would make a good base for the tumbler. It was way too heavy to use as an actual case for anything anyway. I took it all apart and used one half as the base.
tumbler parts |
I found some paper rollers in a pile of old parts from some broken printers. Also found 4 matching ball bearings that I could use to support the rollers. The motor was a very old gear reduction motor I had been hanging on to for years. Can't remember where that came from.
boring out aluminum tube for bearing bushing |
parting off bushing flush with bearing |
bearings and bushings |
bearing mounted in base |
Next I mounted the bearings in the base. I eyeballed the spacing and figured 4" between the shaft centers should be good. The bearings had a lip so I could mount them by simply drilling a hole in the base; no bearing mounts were needed. I finally had an excuse to purchase a step drill bit to drill the sheet steel for the bearing. Drilling the holes to 11/16" gave me a perfect press fit for the bearings. After the bearings were installed, I reused the rubber feet from the case and mounted them on the bottom to dampen any vibration and noise.
The roller shafts already had e-clips at each end. I just needed to move the clip on one end closer so the spacing was right for the width of the base. I placed the shaft in my lathe and with the shaft spinning, I used a cut-off wheel in my Dremel to cut a groove in the proper place for the e-clip.
The next step was figuring out the gearing. Here's how I did it. My drum is about 6 27/32" in diameter. I want it to spin at 40 rpm. My roller shafts are 0.62" in diameter (at the rubber bushings).
6.84375" (40 rpm) = 0.62" (X rpm)
Solving for X tells me I need my rollers to spin at 441 rpm. My motor is rated at 1600 rpm. If I divide my roller rpm by my motor rpm I find I need a reduction ration of 0.28. Luckily I had a box of gears that I had saved from various pieces of equipment. I found a nice set of metal gears with a toothed drive belt. Looked perfect if I could find the right sizes. Turns out I had a 25 and a72 tooth gear. Dividing those gives a reduction ratio of 0.35, close to the 0.28 I was looking for. If we run the numbers we can determine what the drum speed should be with those gears.
25 teeth (1600 rpm) = 72 teeth (X rpm)
= 555 rpm shaft speed
0.62" (555 rpm) = 6.84375" (X rpm)
= 50 rpm drum speed
Off by about 10 rpm. A little over is ok because the motor will slow down some (this turned out to be a slight understatement) when a load is applied. Of course the inside diameter of the gears didn't match the shafts. So I made two more reducing bushings on my lathe. The motor gear was press-fit on the shaft. The roller gear was held in place with set screws.
The next step was mounting the motor. The 72 tooth gear was too large to clear the drum if I put it inside the base. So I attached it to the drive shaft outside of the base. I drilled a hole through the side of the base to pass the motor shaft. Then I positioned the motor with the drive belt installed, marked the spots for the holes, and drilled for 1/4" bolts. The larger gear also needed a longer drive belt. I got one from McMaster-Carr for $3.
Next up was the wiring. I used a switch and a C14 inlet from an old printer to connect power. I carefully measured for the holes, drilled a pilot, and used a nibbler to cut the openings. The old fabric covered wires on the motor were in bad shape so I covered them with heat shrink. After soldering the wires, I held my breath as I flipped the switch. I never actually checked the motor to see if it worked! Luckily it did.
power switch and inlet |
Unfortunately the drum was not water tight. There were two small indentations on the lip of the drum. I placed the lip on the disc sander to get everything even. Then I placed a piece of neoprene from an old mouse pad inside the lid. After that, the drum held water just fine.
ready to go! |
The last step was checking the speed of the drum. I placed it on the tumbler empty and flipped the power switch. I counted the drum revolutions while counting down one minute on the timer. Wow, 50 rpm just as predicted. Now for the moment of truth. I filled the drum with 1 gallon of water and 5 lbs of stainless pins, placed it on the base, and flipped the switch...
the sweet sounds of failure |
pretty, but useless |
Nothing! Turns out the motor was completely under powered for the roughly 13 pounds I had loaded on it. I have larger motors but the 25 tooth gear will never fit the motor shaft. And the motor would be too large for the base. Now I'm pissed. I'm thinking I need a complete redesign, ridiculously over engineered and over powered. But it will have to wait while I lick my wounds.
No comments:
Post a Comment