The Korf Blog

The inside story: our research,
development and opinions

20 July 2022
Turntable Drives. Part I, Belt Drive
We haven't been blogging for quite a while! In the last 2 months, we were only building and shipping the tonearms. No part of the process was outsourced—doing it all ourselves is the only way to iron out small irregularities before the series production begins.

If you want our tonearm, or a headshell, or a ceramic spacer—please subscribe to this blog. We will open a new web shop in September, and you will be the first to know.
~
Now that the tonearms are dispatched to their owners, there is time to share some light summer reading with you. Absorbed with the tonearms, we have neglected other key elements of analogue reproduction—pickups, drives, amplification etc. Over the next couple of months, I want to talk about the drives. They are fun, and there's a lot to discuss.

Also, most information available online consists of monosyllabic platitudes like "belt drive bad, direct drive good" (or vice versa). I think this conversation calls for a bit more nuance.

Why are the turntable drives complicated?
Page from a Thorens 124 user manual shows its complexity
At a first glance, rotating a dynamically balanced platter (that's neither loaded nor going anywhere) at a gentle 33 1/3 RPM is trivial. A washing machine drive feels a lot more challenging. And even the simplest electric moped seems starship-complicated in comparison.

It only starts being hard once we specify the requirements precisely. Ideally, both the average speed and the momentary speed should not drift at all. The bearing has to be completely silent. The whole machine should serve as a "mechanical ground" to the considerable vibrations generated by the stylus. Last but not least, while the braking force exerted by the pickup's needle is quite small, it also varies a lot.

This "fine print" hints at why turntable drives are so diverse. There's a need to balance speed precision, low noise, ample torque, size, weight, and the obvious cost concerns. We all know belt drive and direct drive, but there are also idler drives, hybrid drives, gear drives, magnetic drives, hydraulic drives and so on and so forth. Our focus will be on the three most common types: belt, idler and direct.

Belt Drive






A. Motor

B. Motor Pulley

C. Belt

D. Driven (Platter) Pulley

E. Platter

F. Main Bearing
These days, most turntables use a belt drive. Inexpensive commercially available motors rotate fast, and their rotation is quite uneven (they exhibit torque ripple—you will be seeing this term a lot). To drive a platter smoothly and slowly, they need a mechanical transmission that would bring the RPM down and get rid of the ripple.
Two pulleys and an elastic belt are just the thing: the belt acts as a low pass filter, rejecting the high frequency torque ripple of a fast rotating motor.


It's perhaps easier to understand if we draw its electrical equivalent. The compliant elastic belt can be approximated by an inductance, and the load of the bearing and the platter is shown as a resistor.

This circuit only works as a low pass filter if there's a matching load R. The filter's time constant τ is defined by L and R. Or, in our original mechanical circuit, by the belt elasticity and the bearing resistance. The latter is usually provided through high viscosity lubrication.

This is why the type of the bearing oil matters so much. Changing the oil for a supposedly "better" thinner one can thoroughly ruin the wow and flutter performance of a turntable.
A nice side effect of having a belt drive is a physical distance between the motor A and the platter bearing F. This allows for a degree of isolation, both vibrational and magnetic. A separate motor (or platter) suspension can be designed to effectively cancel motor's acoustic noise.

But that's not all that we need from a drive; remember the load fluctuation? Now our model stands on its head. The platter is the source of the disturbance, and the changes in loading need to be quickly and faithfully transmitted to the motor. Any lag between the energy loss at the platter and the energy compensated by the motor torque will worsen the drive's performance.


To put it another way, the last thing we need here is a lowpass filter. This is at odds with the need to smooth out the motor's rotation.

Belt drive designs often sacrifice transient response for better measured performance
The skill of a turntable belt drive designer lies in solving this contradiction by choosing the motor, the lowpass filter time constant, the belt elasticity, the bearing resistance and the platter mass in such a way that the energy losses at the platter are compensated quickly while the motor's torque ripple is still reliably rejected.

This is hard, and belt drive designs often sacrifice transient response (ability to replenish energy losses quickly) for better measured performance and quietness. Now you know why most belt drives do well in wow, flutter, and rumble, but have disappointing dynamics.

But this is just the first challenge facing an engineer who wants to design a competent turntable belt drive. In the next blog post, we will go through a few more critical problems the belt drives have.

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