The Korf Blog

The inside story: our research, development and opinions

30 October 2018
Turntable Main Bearing Vibration, Part I
Bearings and vibration measurement belong together, and I have learned a lot about vibrometry from my tribologist colleagues. They really liked their ball bearings, because of the impressive noise produced during rotation (that translated into easy-to-measure acceleration), and because ball bearings announce their imminent departure with even more pronounced vibration. Most also intensely disliked plain bearings: "They are too quiet, and when they finally start vibrating it's too late."

"Too quiet" is just what we need for turntables, but are plain bearings really that quiet? And if they are, why is there such a pronounced difference in subjective performance between various designs?

In this series of posts, we will measure broad-spectrum and low frequency noise profile of a typical bearing. We will also measure how the bearing acts as a conduit for vibrations originating in playback.

Here's a typical turntable bearing, as fitted to our long-suffering Telefunken S 600 test rig. It's not the best of its kind by far, but neither is it the worst. The stainless steel shaft is rotating in bronze bushings, and a thrust plate (held with a circlip) bears the weight of the platter. The bearing housing is a nice oversized casting, somewhat compensating for a thin shaft.

To measure vertical acceleration, we've attached the accelerometer to a thrust plate. For horizontal axis, the similar Endevco Picomin 22 accelerometer is glued to the side of bearing body with cyanoacrylate.
First, let's see if the vibration generated at playback actually goes through the bearing. We'll play a 1 kHz noise and see if it's visible in the vibration spectrum.
These peaks look large on the chart, but in reality they are an order of magnitude smaller than tonearm resonances. 0.01G here vs 0.1-0.2G typical for the tonearm. Of course the bearing itself is a lot more rigid than a typical tonearm and, consequently, flexes a lot less.

For the second measurements we used a clamp shown to the right, a Micro Seiki ST-10 weighing 1 kilogram. Interestingly and somewhat counter-intuitevely, putting a clamp on a disc reduces vibration travelling through a bearing. I guess that damping provided by the clamp plays a bigger role than a better interface between the disc and the platter (and, consequently, the bearing).

Vertical vibration (not shown) was about half of the horizontal one.


So yes, the bearing is significant when it comes to vibration. We've got a clear positive result. Now let's go explore bearing's own noise.

In the next blog post, we will measure very low frequency (0 to tens of Hz) acceleration, and compare the result recorded with the stationary platter, the platter being driven by a motor, and the platter rotating by inertia with a motor switched off and the belt removed.

Please subscribe to receive blog updates in your inbox!
comments powered by HyperComments