Turntable speed accuracy

when it comes to bearings/ motors and platters I see a difference between the vintage designs and most of the modern turnables. The  service principles say I need to lubricate the EMT' s bearing every year, oiling the motor once in years. Introduced in 1951, the EMT 927 is a massive turntable with an aluminum chassis,  meassuring  6 7. 5 x 52 x 21 .5 cm with a weight of 80 pounds. The platter has a diameter of 44 cm and sits in an amazing precision ball bearing. The bearing shaft is 16.6 cm long and has a diameter of 2 cm! The motor itself is massive. It looks like an industrial motor, 13.5 cm in diameter and 20 cm long! It is a 3 phase synchronous type with phase shifter.

The big MS designs have an oiled, capsuled bearing which should keep a man's life. The massive platters have a weight of 40 to 50 pounds, the full table up to 170 pounds. The Micro Seiki motors are very precise (with a competent controller) as the Continuums motor is as well, a linear 3 phase AC-24V DC Brushless design with integral optical decoder. You need looking at these parts to get an understanding why the usual small motors used for todays TTs are a joke! The Continuums platter has a weight of 60 pounds. I have put an image of the self lubricating bearing on my page to show the difference.

I do understand when Dertonearm questions the quality of new developments against the Micro Seiki, EMT or Continuum standards. In all the mentioned designs the mass does its job with inertia and resonance control. All that makes for a very good turntable in my opinion. And in all designs precision is of importance. 

best @ fun only

Very interesting thread!

Obviously the question how important speed accuracy actually is and how to achieve best results challenges TT designers and puzzles music lovers like us for … as long as vinyl records exist!

Maintaining constant AVERAGE speed of 33.1/3 or 45 rpm has been accomplished, but I don’t think that AVERAGE speed accuracy is extremely critical. The real engineering challenge is to maintain constant speed at any given time in any infinitesimal small time interval. Not everyone might agree, but I don’t think this engineering challenge has yet been accomplished. Maintaining absolute constant speed against an ever changing load is … probably impossible!?

To illustrate my thoughts I would like to use an analogy.
Imagine a car rolling on uneven beams of a railroad track being pulled by an engine which in turn rides on the smooth train tracks. The car in pull is our platter, the bumpy beams represent the ever changing load caused by stylus friction, and the pulling engine is our electric TT motor. Our goal is to keep the car in pull at absolute constant speed (measured in the horizontal plane). Forces caused by the tires hitting a beam instantly try to slow down the car in pull and the pulling engine tries to work against this load change to maintain desired speed.

It seems obvious, that a strong engine is beneficial to minimize speed changes caused by load changes. It also seems obvious, that using a flexible rubber band to connect the pulling engine and the car in pull will make it even more difficult to maintain constant speed of the car in pull. No matter how strong the engine, the flexibility of the rubber band will allow the car in pull to vary the distance between itself and the pulling engine caused by ever changing forces from climbing up each wooden beam.

In this analogy the rubber band represents BD designs, while a stiff beam connecting the car in pull with the engine would represent idler drive designs. A DD turntable places the engine into the car in pull itself.

Quartz controlled speed regulation of any TT drive system requires a speed change to actually happen first, before it can be detected and in return corrected.

In contrast, unregulated A/C motors (Garrard, Lenco etc.) actually don’t try to maintain constant speed. If we ignore that our rotating system has a mass, the resulting platter speed of an unregulated A/C motor drive system becomes a direct function of the applied friction. The absolute speed change resulting from a load change depends on the steepness of the torque-speed curve of the electric motor in the area of operation.

The link below shows several characteristic torque-speed curves of induction A/C motors (Figure 9).

http://inductionmotor.co.in/shaded-pole-ac-induction-motor.html

The famous vintage idler drives (Garrard 301/401 and Lencos L75/78 for example) use shaded-pole induction A/C motors (curve on bottom).

I did some measurements on my own Garrard 301. When working against friction caused by the platter bearing as well as needle drag, the motor operates at about 99.6 % of no-load speed (very close to the right end of the torque-speed curve on the chart). Only the steepness of the curve in this area determines the resulting speed change from a certain load change (increased bearing friction from additional weight on platter for example).

Accepting the fact, that achieving constant speed is literarily impossible, the real question to me becomes, which design approach results in sonically more acceptable speed changes caused by the constantly fluctuating load from needle drag.

The regulated drive system requires a speed change to occur first to detect and correct the speed by applying more torque. In an unregulated A/C system speed changes also occur, but no mechanism tries to correct the speed back to the desired nominal speed. Both systems accept small speed changes as a result of load fluctuations.

Even if we were able to precisely measure these extremely small effects, IMHO only our ears can judge which type and characteristic of speed change is more acceptable to our hearing. The mix of compromises that each design comes with and the effect of all of these compromises to our hearing determines what type of drive system is able to produce most pleasing results.

I went thru my own little try and error TT experience and am by far no expert on TT designs. What I do know though is (even though IMHO none of the existing designs achieves it) that absolute speed accuracy is of utmost importance to the musical presentation. The sonic difference between low and high torque drives needs to be experienced to understand the impact to the musical presentation.

I’m currently using a Garrard 301 with Loricraft’s large PSU providing 220 Volt and 50 Hz to the Garrard, which – compared to driving the Garrard with 110 Volt at 60 Hz, using a smaller pulley – “steepens” the motor’s speed-torque curve in the critical area close to no-load speed.

No matter what drive system is used, I’m convinced that a high effective torque drive is a must, and that any flexible element able to store energy between motor and platter will work against the goal to minimize speed changes from load fluctuation.

I found TT information on Arthur Salvatore’s webpage very interesting - especially his idler drive section.

http://www.high-endaudio.com/RC-Tables.html

Tonywinsc,

For records with holes out of centre my engineer mate simply uses a thread file ( very small round file ) to open the hole up only in the direction you need to move the record to centre it. Then puts a marker on the label to indicate which side of the elongated hole butts up to the spindle. This seems to work reasonably well.

Dover, good suggestion. I have a small, round bastard that should do the trick. I might give that a try.

The Allnic SpeedNic works pretty well. I have put two images on my page (last two images when scrolling down)

best @ fun only