Rebuilding an Induction Motor

Part of the Making Stuff collection
by Douglas W. Jones
THE UNIVERSITY OF IOWA Department of Computer Science

Copyright © 2020. This work may be transmitted or stored in electronic form on any computer attached to the Internet or World Wide Web so long as this notice is included in the copy. Individuals may make single copies for their own use. All other rights are reserved.

The Motor

The motor, as found
photo of a motor with pulley
photo of the motor nameplate
I picked up a small 1950 or 1951 vintage table saw at a garage sale that came with this motor. The motor worked, but the wiring was unsafe, and when I opened the junction box cast into one end bell of the motor, I found that the phenolic terminal board was badly burnt, probably the result of a loose connection.

The nameplate on the motor says that it is a 1/3 horsepower 1725 RPM motor, model 5048301, made by the Delco Appliance Division of General Motors in Rochester, NY. This matches the requirements stated in the manual for the table saw. The Delco Rigidframe trademark is cute, but there's no hint on the motor of a date of manufacture. The nameplate says "Pat. Pend.," but I have not been able to find a patent that this might refer to.

The nameplate says 115V, 60Hz, 6.5A, but it does not say that this is a capacitor start motor, although that is obvious from the large capacitor mounted on top of the motor and the click made by the centrifugal switch as is spins up when turned on.

The construction of the motor suggests that it predates World War II. The bearings on the motor are not sintered bronze, but rather, classic bronze journal bearings with rather large oil reservoirs with felt wicks to bring oil to the rotating shaft, and the end bells are cast iron. Overall, the motor looks very much like the capacitor-start induction motors pictured in Electric Motors for the Farm, Farmers' Bulletin No. 1858, U.S. Department of Agriculture, Nov. 1940. See particularly the cross section in Figure 1 and the overall view of a "Capacitor motor" in Figure 5.


Disassembled and very dirty
Photo of a very dirty motor
The motor worked when I got it, but the unsafe condition of the terminal board led me to to open up the motor for repair. The hardest part of opening the motor was removing the pulley from the shaft. The shaft was badly scored inside the pulley, and to remove it, I needed to heat it with a torch while pinching metal wedges into the space between the pulley and the flat end of the end bell. Each heating and wedging cycle only moved the pulley a fraction of an inch, and I always let everything cool between cycles.

4 long bolts connect the two end bells of the motor, clamping them to the central stator assembly. I unscrewed these bolts and set them aside, along with the nuts and unused rubber shock mounts at their ends. From these shock mounts, I conclude that this motor probably had some former use before it was repurposed to power a table saw.

The pulley end of the motor rotor had a thrust washer between a step in the shaft diameter and the bearing in the end bell. This washer took the force while wedging off the pulley. At the other end, there was no thrust washer, and I was confused until I realized that there was a single round ball bearing ball sitting in a hollow in the end of the motor shaft and held in by oil. This ball works against a flat bronze plate that blocks the outside end of the bearing on that end of the motor.

Removing the end bell with the motor wiring in it was difficult because the wires were very short. Rather than unsoldering the wires from their anchors on the inside of the end bell, I opted to unscrew all the components inside the end bell so I could study the wiring before disconnecting anything.

Once the motor was open, I found ample amounts of dust, some of it oily, but mostly dry. So much dust that my first order of business was to take the motor parts outside and blast them with compressed air. With that done, I wiped down the end bells with degreaser (Goo Gone). The result converted parts that were unpleasant to handle into clean parts I could work with.


Before I began intensive work on the motor, I pulled the plugs on the oil reservoirs so I could top up the oil and check the felts. These plugs are large-diameter, held in by a press fit. To remove them, I had to drill them, tap the hole, and use a screw to get a grip for pulling. Once I inspected the felts and concluded that they were in good shape, I added oil and then used a soft hammer to re-seat the plugs, with the screw holes plugged by screws. From here on, those screws will suffice as oil holes.

Motor parts after cleaning
Photo of the motor clean
The first order of business after cleaning things up was to examine the motor for evidence of abuse. After blowing the dust off the motor windings, there was no evidence of discolored varnish on the running windings, evidence that they have never been run to the point of overheating. Evidently, the black paint on the outside of the motor was sprayed on after assembly, because those parts of the windings exposed through the vent openings in the motor were black, but paint is not damage.

The starting windings are finer wire, varnished and cotton insulated. The cotton insulation is more evidence of the pre-war origins of the motor. The cotton holds dirt more aggressively than varnish, but there was no visible evidence of overheating. While overheating in the main windings would be evidence of running overloaded for extended time or of short circuits, the starting windings could overheat from a bad centrifugal switch or operating under sufficient overload that the centrifugal switch never opened.

End bell wiring
Photo of the wiring
The next order of business was to inspect the centrifugal switch and thermal protector. There was no evidence of problems there.

The centrifugal switch consists of three parts, a spring-loaded fly-ball governor mounted on the end of the motor rotor, a phenolic plastic plunger that the governor pushes out until the motor rotates at a sufficient speed, and the switch itself mounted on the end bell. The plunger rotates with the shaft and has a flat polished face that slides against raised areas on the switch lever. The switch is very simple, mounted on a phenolic board that goes inside the motor bell.

The thermal protector is a classic Klixon device. Unlike modern motors, where the thermal protector is usually embedded in the motor windings, this is mounted in a form-fitting well cast into the end bell, so it only responds to overheat when the end bell gets hot. Thus, it is excellent for protecting against long operation at mild overload, but not able to respond to rapid temperature rises caused hard stalls. Note, however, that Klixon protectors also include resistive heaters that offer some protection against overload even when the local environment is not hot.

Note that when I got the motor and found that one binding post on the terminal board was loose because of the burnt board, I moved that terminal to a spare unused hole in the board.

Burnt terminal board
Photo of the burnt board
After removing the binding posts from the terminal board and cleaing it, it was clear that the holes in it had been labeled 1, 2, 3, 4 and 5 with stamped letters. Holes 1, 3 and 5 were originally unused, and the word LINE was stamped between holes 2 and 4. Hole 2 is the one which was thoroughly burnt out, and the label 2 can only be inferred because it was a casualty of the burning.


The end bell, rebuilt
Photo of the end bell with new wiring
I found a piece of green fiberglass-epoxy board the same thickness as the phenolic terminal board. I traced the shape of the old board onto the replacement, hacksawed it to rough size, and then clamped the old board to the new with two clamps.

I used a bench grinder to grind the new board as closely to the exact dimensions of the old as I could, removing only one clamp at a time so that the two boards remained in alignment, and then I used the square holes in the old board as templates to drill holes in the new board. Of course, with one hole in the old board badly burnt, I could only do 4 of the 5 holes this way. For the final hole, I flipped the old board, aligned the three holes carefully, and reclamped it to drill the final hole.

I do not have a drill that makes square holes (such things do exist) so I used a square needle file to square up all of the holes. The binding post screws do not have slotted heads, they have square shanks instead.

Before reassembly, I carefully taped all of the screw holes and bearing openings in the end bells, and then I spary-painted the end bells with the paint I had on hand, which was brown.

I unsoldered all the old wires from the centrifugal switch and from the Klixon thermal protector. The contacts on the centrifugal switch are fairly heavy, but after an unknown number of starts, they may have been seriously pitted, so I clamped a patch of 400-grit sandpaper between them, held the contacts closed with light finger pressure and pulled the paper out (a distance of under half an inch), then flipped the paper over and repeated the process. A single short sandpaper stroke like that should be just enough to clean the contact surface while removing a minimum of contact material.

After I screwed all the parts into place on the end bell, I used 18 AWG high temperature (PFTE) wire (rated at 16 amps for chassis wiring) to rewire the parts. I added faston connectors for connecting the end bell wiring to the motor windings.

I did not replace the wiring from the end bells to the motor windings for two reasons: First, it did not seem in terrible condition, and second, the way it was connected was unlike more modern motors where I have replaced such wires. In these modern motors, crimp connectors to the motor windings, with the free ends of the windings bound with lacing tape and the connecting wires laced into the bundle to take the strain incurred during assembly. In this older motor, the free ends of the windings are not laced at all, but are restrained and protected by giant slabs of fish paper that protect them from the governor on the motor shaft. Stain on the connections is taken by weaving the wire through the windings. Replacing the connections would have required undoing this weaving, potentially damaging the motor windings.

The rebuilt motor
Photo of the end bell with new wiring
Finally, I crimped faston connectors to the wires coming from the stator and connected them to the rewired end bell. With that end of the motor assembled, I balanced the partly assembled motor on its end, carefully dropped the ball bearing ball into the end cap, lowered the rotor into place and checked that it spun freely.

Then I lowered the other end bell over the motor, put the long screws in place and tightened the nuts on the end. I could not turn the motor shaft with my fingers, but I could turn it with pliers. Not good.

So I took out the long screws, pulled off the one end bell (leaving the motor balanced on its end again), pulled out the rotor, and repeated the assembly process. With the motor assembled but the screws loose, the shaft spun freely under finger power, so I guessed that perhaps I had made a mistake in the way I tightened the long screws. Instead of tightening one screw before starting on the next, I screwed all of the nuts finger tight and then finished the tightening by turning all the nuts just a quarter turn each before moving on. This kept the tension even, preventing one nut from pulling the end bell off center, which must be what I did wrong.

The motor ran well after assembly, smoothly, quietly, and with hardly any vibration. I checked the operation of the centrifugal switch by noting that there was a spark visible through the vent openings in the end bell a fraction of a second after the motor was turned on. Once I was confident it was running well, I used a hand-held file braced against a block very close to the shaft to turn the shaft, removing the burrs that had made it so difficult to remove the pulley. The use of a block to brace the file was important because the shaft has a flat. Without a block to support the file, attempting to true up the shaft with a file would just make the file bounce against the flat. With the block giving tight control, I could hold the file tangent to the shaft and just take off the burrs without falling into the flat.

With the pulley in place, I also used the file (still bracing it against a block) to polish up the faces that contact the V-belt and remove dings from the face of the pulley, mostly dings that I'd created while struggling to remove the pulley. Its worth noting that the motor barely got warm during the time it was running while cleaning up the shaft and pulley.

I described some of this work in a posting on the Vintage Machinery web site.