or the best way to make it turn

About 2 years ago I started work on a turntable project - basically a large-platter design by Thom Mackris (Galibier Design) using parts procured from Thom by Chris Boettcher for a Randall Museum group project. I hadn’t really given much thought to turntable design since I bought a SOTA Star Saphire turnable with an ET-2 linear-tracking arm in 1986. This combination has served me well over the years, but I wanted to upgrade the arm and turntable, so signed-up with Chris for the turntable parts.

Along with the kit of parts, Chris supplied a copy of “Turn Your Table” by Dr. Götz Wilimzig (Sound Practices issue #10, pp. 10-14, 43, Winter 1995) which seemed to encapsulate the turntable drive philosophy. As I read the article, I started noticing strange statements and ideas interwoven with logical ideas. By the time I finished the article, I was scratching my head and ended up pulling out the old engineering books, and after I went through them, bought a few new ones and went through those. After a few months of research, my conclusion was that Dr. Wilimzig’s article had the right goal - control turntable speed due to variations in drag caused by the needle in the groove, but that his method was confused and his recommendations flawed. I think I’ve come up with a better solution, and plan to try it out.


This posting will be the first in a series reviewing my research on turntable motors - a series because there is a lot to say, and more research will happen as I write these.

Part 1 - the article

Dr. Wilimzig’s article starts out with a surprise:

“Please stop reading if you insist on feeding a synchronous engine with AC. They’re always vibrating like hell.”

Yes, you want to keep motor vibration away from the platter and arm, but this is true for any kind of motor. Some of the old turntables really did have rumble problems, but the problem seems to be more of an isolation probblem than one of motor technology. Also, I thought that good synchronous motors were pretty quiet. I remember the main noise the capstan motors in the old Ampex tape recorders put out was a “whoosh” from the ball bearings in the motor - not AC vibration. Oh well, I guess AC motors are out.

The article then explains the fact that the drag of the needle in the groove slows down the rotation of the platter, and that the motor must make up this loss. At this point the logic and English get confusing. The gist seems to be that the instantaneous forces from the needle must pass from the platter through the belt to the motor, where the motor can make up this lost speed, ideally instantaneously.

This results in the first condition:

“(1) The rotor’s moment of inertia should be low low down [sic].

That’s why all engines with high mass rotors like synchronous engines don’t fit. They can’t follow the platter’s motion.”

Well this is wrong in several areas. You don’t want the speed of the motor to change, so wouldn’t infinite mass be best? Also, if the motor is tightly coupled to the platter (as recommended), then the inertia of the motor becomes irrelevant compared to the inertia of the platter. A quick calculation showed that the inherent inertia of the Maxon motor used by Chris Boettcher (#xxx) gets increased by a factor of about 85 when coupled to the 17 lb platter.

A quick description of the torque vs speed curve for a DC motor is given - the fact that a shunt-wound DC motor fed from a constant-voltage supply will tend to counteract a decrease in speed by increased torque. Keep in mind, this is not perfect control - hence the slope of the curve. This somehow leads to the second condition:

“(2) The DC engine is a real must. Without any exception.”

Well AC motors have similar curves, too. But they are to be excluded. The third condition makes more sense:

“(3) Look for a strong engine, specified by a low value of delta n/delta M.”

This leads to the inevitable conflict with condition 1, since strong engines are usually large engines with high inertia.

Now comes one of the stranger paragraphs in the whole article:

“To diminish the errors of speed the engine has to pull back the platter to 33 or 45/min. It should do that like a rocket. We are looking for a sprinter reaching his speed in the shortest time, Carl Lewis would be right. Some manufacturers publish data regarding how quickly the engine runs up to 63% of its maximum speed of rotation without load. So:

(4) Run up time has to be short. Imagine Carl Lewis is leaving the starting block.”

Again, the added inertia of the platter will swamp out the effect of weak motors. Also, initial start-up time really isn’t linked to rotational speed accuracy. A motor that needs to quickly change speed, like a servo motor moving a magnetic head assembly in a disc drive, needs the ability to quickly change, but a turntable is essence of speed constancy!

Next comes a discussion of the choice of DC motor, and here we are on firmer ground. Bronze sleeve bearings are preferred over ball bearings - good. Don’t exceed the maximum radial force rating - good. Precious metal brushes are preferred over graphite brushes for consistency - good. Use a good voltage regulator - good. Several motors are evaluated (although names are omitted) - but is a decent exercise in comparing motor specs.

My initial analysis of the article led me to analyze the motor-to-platter mechanical situation, as well as look carefully at how well motors keep to a specific speed. I’ll give my overall evaluation of Dr. Wilimzig’s article at the end of this series.

Next posting: My ears are opened in Zurich.