Turnable database with TimeLine

01-03-14: Richardkrebs
Yes the Goldmund used a JVC motor.
Like, I suspect most DD designs, it is a synchronous motor. The rotor ( platter) is compelled to follow the rotating field, back slightly in phase.

I'll explain how the Goldmund motor really works.
Firstly Brushless DC motors require an electronic controller to continually switch the phase to the windings to keep the motor turning.
Secondly, the Goldmund Studio uses a coreless DC servo JVC motor. The JVC motor has a frequency generator which generates a frequency when the motor is running. A phase comparator then compares that motor generated frequency to a reference frequency generated separately by the Quartz Crystal. Any difference detected is then fed back to the servo controller to correct the speed. The motor will not run or hold speed without the servo running.

In a nutshell the speed accuracy of the Goldmund will be dependent on the quality of the electronic controller and the quality of the servo.

Note that Goldmund themselves then added lead mass to the platter to increase the inertia. They claimed that this was required to smooth out any speed irregularities. Clearly Goldmund themselves did not believe that properly designed servos are adequate in of themselves for accurate and consistent speed when they designed the Goldmund Studio. The Goldmund Reference TT was belt drive, high inertia.

Brinkmann allude to the issues of servos in their white papers. They have gone for a low torque minimally invasive ( S L O W ) servo action in the interests of sound quality.

It is interesting to note that in listening tests comparing the Goldmund Studio and the Kenwood L07D side by side, the Kenwood has better timing, despite the servos not kicking in until a speed error threshold of +-3% is reached. The Kenwood L07D relies on inertia and back emf within the motor to hold the speed.

01-03-14: Richardkrebs
The rotor lags in phase slightly behind the rotating field. If it didn't it would produce no torque. Increase the load and this phase angle increases and the motor draws more current. But the motor then continues rotating at the same speed. .

If the phase angle increases and the motor draws more current, this means that the motor has lost speed. The servo kicks in and brings the speed back up.

01-03-14: Richardkrebs
This is what I observed on the scope. The motor was responding to stylus drag, literally note by note. It was not showing a problem, it was showing the motor working properly and the relative enormity of stylus drag.

The conclusions from your testing are wrong. What you are measuring on the scope is the solution that Goldmund has provided to record playback. You are measuring the sum total of the quality of their controller, their motor, the inertia built in to the platter and the quality of engineering and design for THAT particular turntable. You have not measured the platter speed.
If you had scoped the power supply of a Garrard idler, Micro Seiki or EMT you would get different results.

01-03-14: Richardkrebs
For a DD drive, it has a reasonably high inertia platter but this was not enough to "push thru" these load changes. .

The Goldmund Studio does not have a high inertia platter. The Goldmund Studio platter weighs about 3kg. Both the Technics SP10mkIII (11kg) and Kenwood L07D (7.7kg with stabiliser) DD’s ( comparable products in terms of market positioning ) have significantly higher inertia - 2-3 times higher.
Of course the Micro’s are 10-15kg, EMT 927 5kg distributed to a 16” platter to achieve an equivalent 50kg, and my Final Audio has 22kg. These are what I would describe as high inertia TT’s.

01-03-14: Richardkrebs
Wrap a properly designed servo around this type of motor and the phase angle change with load is reduced. .

There is no consensus on what is a “properly designed servo” for Direct Drive turntables.
The Technics SP10 servos use algortihms to estimate predicted errors and employ rapid response times (limited by the technology of the day). The servo action includes error and overshoot.
The Victor/JVC decks use an averaging process to provide a smooth transition when servos call for speed correction.
The Kenwood L07D uses no servo, and relies on inertia and back emf unless the speed error is quite large, at which point the servos kick in and additional torque is applied.
The Brinkmann uses very slow servos for soft recovery.

In reality the L07D and Brinkmann DD's are closer in conception in maintaining accurate speed behaviour to a high inertia, non servo AC synchronous motor driven solution like the Micro or Final than they are to the Technics SP10. I also note that Brinkmann claim that 15kg is the minimum platter weight required for adequate speed stability.

Lew
The Beat appears to uses electromagnetic drag to "pre load" the motor. Moving the rotor back in phase relative to the rotating field. I am not sure about the other TTs mentioned. The theory being that this artificial torque demand is significantly higher than any seen due to stylus drag. It seems like an elegant solution. The question is of course how does it sound? I do know that we are very sensitive to micro pitch changes and that stylus drag is large. The choice of open loop or closed loop speed control in ones TT is a personal one and we all make that choice with open ears and hopefully open minds.

Very interesting thread Peter. Thanks for getting it going. I'll volunteer my Grand Prix Monaco to be a test subject, as soon as I can barrow Albert's timeline...

The Grand Prix claims highly accurate speed control (from their web site):

"DSP Signal Processing / Active Feedback Loop Technology:
Our speed control features a CPU utilizing a Digital Signal Processor that interrogates and maintains the speed of the platter thousands of times each revolution and thereby ensures frequency accuracy and distortion free playback that is simply not possible with conventional types of drive systems. A highly accurate test procedure has demonstrated the speed error to be an extremely low .002%! Speed accuracy equals frequency accuracy, which means there is virtually no distortion on playback."

Peter, one thought on your SME, have you ever cleaned the motor pulley (and sub platter) and then flipped the belt over? I used to do this with my SME 20/2 every 9-12 months and it did seem to sound better...(that is unless you used a new belt for the test)...then I'd replace the belt every 18-24 months...

Thanks Jfrech. I used to clean the belt on my SME Model 10 about once a year. I've just ordered two new belts for my 30/12 which was new in March 2012. I'll clean the belt I have in the next day or two and see if I notice a difference. I sent the TimeLine back to Albert. I just can't see spending $400 on a new one.

Your Grand Prix will be a good addition to the database. Thanks.

Addendum to above post :
The big Micro's use DC motors with frequency generated servos built into the motor. So their response to speed issues has the same issues as the direct drive, the main advantage of the micro's is the large inertia. This would explain why some users prefer the Micros set up with a controlled slip by chalking the thread.
The Final Audio has a high torque ( the 22kg platter can achieve full speed in less than 1 turn ) AC synchronous motor with precision oscillators and 80 wpc power amplifier to lock the speed.
As Lewm alluded in his post the AC synchronous motor responds to phase lag by self correction. This is why AC synchronous motors can run accurately without servos. A DC synchronous motor will self correct to some degree, but the self correction is less sinusoidal and less smooth than an AC synchronous motor.
It is wrong to imply that the Technics SP10 relies on self correction of phase lag - its servos are active all the time, as is the case with the Victor/JVC and Goldmund.