Beogram 4002 (5513): A New Exciting DC Motor Failure Mode Due to a Shorted Spark Snubber

Another Beogram 4002 (5513) DC motor arrived recently for some TLC. When I did the usual pre-restoration test in one of my Beogram 4002s, this motor only ran very sluggishly and only if I would help it along a bit. This made it stand out from the usual DC motor that I am seeing on my bench. Usually, these motors run vigorously, but they have strong RPM variations and often make shrieking noises due to dry Oilite bearings. This one was very lackluster and weak in comparison.

I opened it up and did my usual oil infusion of the bearings to make sure that the mechanical aspects were up to spec before I would look into the sluggishness issue. Here are a few impressions of the Oilite bearing oil-re-infusion process. This shows the motor as received:

And disassembled:

This are the bearings in motor oil under vacuum:

The bubbles rising from the smaller (bottom) bearing is air being released from the porous Ni "Oilite" material while it is being replenished with oil. The top bearing only released a few bubbles indicating that it still had oil in it.

After 24 hrs the bubbling stopped and I reassembled the motor. When I tested it the sluggishness had not changed, i.e. there was definitely another issue beyond dry bearings.

I opened the motor up again and I remembered that I had a similar issue a while back. In that case I was able to get the motor running again after removing the spark snubbers that are connected across the three coils on the rotor. Spark snubbers short circuit the EMF that is generated due to Faraday's law when the commutator switches between coils. This prevents sparks and potential damage to the driver circuitry. This shows the rotor hooked up to a multimeter in resistance measuring mode:

The three blue units arranged on a ring around the commutator are the snubbers. Normally, they are insulating at the low voltages typically applied during a resistance measurement. The 16.8 Ohm that show up here is the resistance of the rotor coil this particular snubber is connected in parallel. I measured the snubber one over:

22.9 Ohm for the next coil. On to the third snubber:

This measurement shows that it has a short. This effectively takes the associated coil out of business and this explains why this motor did not have any power. It only ran on two cylinders instead of three, so to speak. From the previous motor with this issue I still had the three-spark-snubber assembly in my B&O parts box. This shows it in front of the rotor:

I cut out (they are spot welded together) the shorted snubber from the rotor and implanted one of the still working ones of the extracted assembly:

A test with the multimeter showed that this coil now had again about 20 Ohms.

I reassembled the motor for the second time, and inserted it into the Beogram 4002. I hooked up my BeoloverRPM device (which is available to other enthusiasts - just send me an email or use the contact form to the right) and did a 12 hrs run. This is the curve I measured:

It appears this motor is up to spec again.

I read up a bit about spark-snubbers in motors. There are many different types of these devices. A quick&dirty measurement with my bench power supply yielded that the working ones started conducting at a voltage close to ±10V. This indicated that the devices in these motors are most likely bidirectional transient voltage suppressor (TVS) devices with an onset voltage in the 10V range. This suggests that one could probably rebuild these motors with modern through hole type TVS units with the right form factor. Another option is of course to just remove the devices, but that runs the risk that the life time of the brushes in the motor may be reduced due to sparking. It probably is also not great for the motor control circuit in the Beogram, but my earlier experiments yielded that it apparently can survive with a snubber-free motor for a while. Anyway, replacement with new TVS units is a bridge I will try crossing for the next motor that shows up with this issue.