Turntable Evaluation - what do you need to know and how do you know it?

[andromedaaudio]

Bearing wear in the pre purchase period , you got to be kidding .
These high end turntables are likely the most simple strains on a bearing that exists on the entire planet .
They are well designed / machined lubricated machines rotating at very slow speeds indoors at room temperature

 

[the sound of Tao]

Measurements have value and tend to give us the simplest of understanding.


Just remember Tao and to put things in perspective , without measurements one wouldnt have a low noise turntable at all

And yes, that is the start of the journey of evaluating. Simplest doesn’t mean unimportant. Just that it communicates the most fundamental. Nuance comes later in the fabric of things.

 

[Vienna]

Bearing wear in the pre purchase period , you got to be kidding .
These high end turntables are likely the most simple strains on a bearing that exist on the entire planet .
They are well designed / machined lubricated machines rotating at very slow speeds indoors at room temperature

Dont tease !!!!!!!

 

[andromedaaudio]

Dont tease !!!!!!!

??? Seems to me you dont have an argument to support your claim , if so i d like to hear it

 

[Phoenix Engineering]

One thing I can tell you that for sure is not useful is anything that checks the speed once per revolution. Anything that does that is useless for actual speed control.

Care to explain this statement?

I've heard people posit that if the platter speeds up by x amount within a rev but also slows down by the same amount within that same revolution that a once per rev speed device will not report any errors. This is true, but unless the speed variations are caused by an eccentric pulley, the chances of the amount of speed increase being matched with an exact mirror image of speed decrease is nil. I have many users of the RR tach report random up/down changes in speed that were indicative of improper bearing lubrication, loose or contaminated belt, twisted belt, belt riding up and down on the platter etc. While the PSU cannot correct for W&F (it was never designed to and would be very difficult with a belt drive), the tachometer is useful for ensuring constant speed over time (most belt drive tables drift in speed as the belt and bearing warm up) as well as diagnosing problems with belt, bearing or motor. I've experienced this first hand. If I handle the belts a lot during testing, the oils from my hands are enough to cause the speed to randomly fluctuate; wiping the belt, pulley and platter edge with IPA will eliminate the fluctuations.

Most people seem to be concerned about absolute speed, as they use a strobe disc to check it. The tachometer does this as well, but to a much higher degree of accuracy. What you do with that accuracy is up to you. I have a DVM that goes to 4 places right of the decimal point; do I need that much accuracy? It depends. If I use the DVM to work on my car's battry charging system, I don't even look to the first position right of the decimal. If I'm measuring DC offset at the output of an opamp in a phono stage, 3 & 4 places to the right of the decimal is important. But I've never heard of anyone complaining to Fluke that they make meters with too much accuracy.

For those who use a strobe disc and assuming you can adjust the speed, do you just change the speed until the strobe spots slow down, or do you go back and forth in an effort to get the speed "just right" where the strobe spots don't move. If so, why?

Measurements don't replace listening ; they weren't meant to. They do quantify performance and can be useful for uncovering gross errors. They also provide a standardized language for comparison that listening tests lack, especially as the differences become more subtle. They are repeatable and independent of the user, are not subjective and are not prone to outside influences, bias or other variables beyond the user's control (mood, health, changes in hearing, intoxication, etc.) Measurements are a valuable tool; like any tool, in the wrong hands or if used improperly, the utility is diminished. Listening is just as valuable of a tool, but used in a different domain of experience and with different rules of correspondence.

 

[microstrip]

(...) i do not agree that the NVS retains noise inside it's structure, as opposed to other turntables draining resonance through their footers, as you suggest. you use the word 'perhaps', so i don't want to respond too strongly. but i think it's more fair and objective to say that we just don't know about what resonance, if any, was inside the NVS, and if there was some, where it went. we also don't know the exact nature of the resonance sensed by the peizo sensors from the other turntables. we only know that some did exist and that it set up a feedback loop.....to the degree that i observed. there are things we know and things we don't know. (...)

You have a point here. The NVS and most decent direct drive turntables do not generate apparently perceptible noise because they have a servo feedback system and supply what their designers consider the exact amount of energy to move the turntable at every moment. However as there are no perfect servo systems they always generate some noise, simply in zones of the frequency spectrum where classical instruments used to measure turntable performance are not sensitive. However people will tell us that our ears are more sensitive than any accepted measurement we can perform - I can't see anyone in WBF disagreeing with th is statement - and that this noise can be channeled toward the LP-cartridge system. And this we know for sure, although we can disagree (and disagree strongly ... ) on how subjectively perceptible is this effect!

In an instrumental sense, your active table display is equivalent to using a stethoscope as Tim uses, only with a different bandwidth and sensitivity - your instrument extends in the lower frequencies.

 

[Folsom]

Care to explain this statement?

I've heard people posit that if the platter speeds up by x amount within a rev but also slows down by the same amount within that same revolution that a once per rev speed device will not report any errors. This is true, but unless the speed variations are caused by an eccentric pulley, the chances of the amount of speed increase being matched with an exact mirror image of speed decrease is nil. I have many users of the RR tach report random up/down changes in speed that were indicative of improper bearing lubrication, loose or contaminated belt, twisted belt, belt riding up and down on the platter etc. While the PSU cannot correct for W&F (it was never designed to and would be very difficult with a belt drive), the tachometer is useful for ensuring constant speed over time (most belt drive tables drift in speed as the belt and bearing warm up) as well as diagnosing problems with belt, bearing or motor. I've experienced this first hand. If I handle the belts a lot during testing, the oils from my hands are enough to cause the speed to randomly fluctuate; wiping the belt, pulley and platter edge with IPA will eliminate the fluctuations.

Most people seem to be concerned about absolute speed, as they use a strobe disc to check it. The tachometer does this as well, but to a much higher degree of accuracy. What you do with that accuracy is up to you. I have a DVM that goes to 4 places right of the decimal point; do I need that much accuracy? It depends. If I use the DVM to work on my car's battry charging system, I don't even look to the first position right of the decimal. If I'm measuring DC offset at the output of an opamp in a phono stage, 3 & 4 places to the right of the decimal is important. But I've never heard of anyone complaining to Fluke that they make meters with too much accuracy.

For those who use a strobe disc and assuming you can adjust the speed, do you just change the speed until the strobe spots slow down, or do you go back and forth in an effort to get the speed "just right" where the strobe spots don't move. If so, why?

Measurements don't replace listening ; they weren't meant to. They do quantify performance and can be useful for uncovering gross errors. They also provide a standardized language for comparison that listening tests lack, especially as the differences become more subtle. They are repeatable and independent of the user, are not subjective and are not prone to outside influences, bias or other variables beyond the user's control (mood, health, changes in hearing, intoxication, etc.) Measurements are a valuable tool; like any tool, in the wrong hands or if used improperly, the utility is diminished. Listening is just as valuable of a tool, but used in a different domain of experience and with different rules of correspondence.

That is the thought sort of... If it has speeds that is off and you correct it then you may have somewhat poor speed control over an area that's correcting it. Even a bad bearing could be "worked out" of the speed by doing that if the motor is strong enough - but really it's just slower and faster at different points.

I learned a lot about true speed from this thread.

I'm not advocating direct drive with endless sensing, just saying that a test like above would be a better evaluation for figuring out consistency when making whatever kind of table you are making. I have zero issues with, and think it even wise to just have a "set it and forget it" type of drive, but during development I'd be running that test a lot to check on my choice for the TT.

 

[microstrip]

(...) Most people seem to be concerned about absolute speed, as they use a strobe disc to check it. The tachometer does this as well, but to a much higher degree of accuracy. What you do with that accuracy is up to you (...)

A good strobe disk can be accurate to 0.05% - do you really think that people really need better than that?

As far as I have read professional turntables included a strobe / speed display and adjustable speed because radio stations were concerned with exact playing time of songs to avoid gaps, not because of sound quality.

BTW, unfortunately a proper discussion on Fluke meters needs to debate at less full scale, precision and accuracy - otherwise we can't separate their excellence from cheap chinese meters.

 

[Phoenix Engineering]

I'm not advocating direct drive with endless sensing, just saying that a test like above would be a better evaluation for figuring out consistency when making whatever kind of table you are making. I have zero issues with, and think it even wise to just have a "set it and forget it" type of drive, but during development I'd be running that test a lot to check on my choice for the TT.

I agree and I'm familiar with that thread. In fact, I use a modified version of Scott Wurcer's Python code to measure the performance of both belt drive and DD (HW40) motors by looking directly at the output of the encoder. This tool is invaluable for measuring W&F but is completely different than what a tachometer provides. Both valuable tools however, IME.

 

[Phoenix Engineering]

A good strobe disk can be accurate to 0.05% - do you really think that people really need better than that?

That would imply a drift of ~18mils at the edge of a 12" platter over a1.8sec time frame; less than18 mils if the strobe disc is less than 12" dia. That's about the thickness of card stock; I doubt the average person could see that with the naked eye and without a stationary standard, but it's possible. I still don't understand your point (unless it's to be pedantic). As I said, use what ever level of precision you are comfortable with and ignore the rest. The tach can be used while the music is playing, most strobes cannot. Or, don't use a tach or any other device; I really don't care either way.

As far as I have read professional turntables included a strobe / speed display and adjustable speed because radio stations were concerned with exact playing time of songs to avoid gaps, not because of sound quality.

How does this explain consumers using strobe discs?

BTW, unfortunately a proper discussion on Fluke meters needs to debate at less full scale, precision and accuracy - otherwise we can't separate their excellence from cheap chinese meters.

I'm sorry, I don't understand your point or what it has to do with this discussion.

 

[microstrip]

That would imply a drift of ~18mils at the edge of a 12" platter over a1.8sec time frame; less than18 mils if the strobe disc is less than 12" dia. That's about the thickness of card stock; I doubt the average person could see that with the naked eye and without a stationary standard, but it's possible. I still don't understand your point (unless it's to be pedantic). As I said, use what ever level of precision you are comfortable with and ignore the rest. The tach can be used while the music is playing, most strobes cannot. Or, don't use a tach or any other device; I really don't care either way.

We can easily integrate drift along, for example, half a minute. Anyway my main point is that we should state a clear objective in our actions in this hobby. In this case the question is clear - do people think that more digits ( or accuracy over .05% ) improve our subjective music quality? IMHO the answer to this question should be clear, otherwise we take it just as a very interesting and technical objective debate, such as debating distortion in solid state amplifiers.

How does this explain consumers using strobe discs?

Honestly IMHO mostly for fancy. My old Technics sp1500 had a strobe and at that time in the early 80's appreciated it a lot when I got it. It looked great! But did not miss it when I moved to the IMHO better sounding Thorens TD160 . Or just that the turntable drive system is so cheap or simple that needs speed adjustment every time - it can help to lower the price.

I'm sorry, I don't understand your point or what it has to do with this discussion.

We always need more digits when the full scale is not optimal for the range you are measuring - the typical case of offset, but not turntable speed.

BTW precision and accuracy are different things. Are you equaly aiming both with the RR tach?

 

[microstrip]

(...) I learned a lot about true speed from this thread.

Thanks - nice to see it is a recent thread on this subject. Unfortunately the references on the subjective part of the subject are AES copyright and not freely accessible.

 

[Phoenix Engineering]

BTW precision and accuracy are different things. Are you equally aiming both with the RR tach?

The timebase is a TCXO with 2.5PPM stability over the commercial temp range (-30~+60°C); in a temp controlled environment like a listening room, the accuracy is most likely much tighter. Aging is 1PPM; it is adjustable if it drifts over time.

The tach uses a 30kHz clock and a 16 bit register to measure the rotation period; at 33.333 RPM, the resolution is 1/54K, at 45 RPM it is 1/40K. The trigger ambiguity of the Hall sensor is most likely greater that either of those, although I have been unable to measure it.

 

[Folsom]

Thanks - nice to see it is a recent thread on this subject. Unfortunately the references on the subjective part of the subject are AES copyright and not freely accessible.

Keep flipping pages till you see the measurements. Then you'll be like "wtf everything I know about turntables is a lie" well, sorta, it's enlightening.

 

[microstrip]

Well, I did it for a few pages. Most measurements seem similar to what I did in my Studer A80 - but with the A80 I have access to a library that tells me exactly what is the component responsible for each peak, life is much easier.

Can you point me which pages are you exactly addressing? Perhaps it is something I missed.

 

[Folsom]

I don't know. Honestly I spent FOREVER, page after page. One thing I did note was the TD124 is excellent. The SP10 was awfully good, too.

 

[Gregadd]

Noise and speed variations are the enemy.(wow and flutter
Preferably we want to decpuple the motor from the platter. Iideally the motor should rest on diffetemt platfprm.
The platter is constamtly huntimg for the correct speed. The motor should be t able to quiclly correct for speed variations.
There is bound to be speculators in the market.
.Reputable companoes abound

 

[Alrainbow]

I am really sick with this situation. The dealer accuses VPI for luck of communication and VPI accuses fedex and Facebook. On the other hand the dealer is struggling to pay his monthly Fixed costs to afford such a refund.

Bro are you going to find any thread to pull into a vpi issue. I don’t know you and do get you had an issue. But how is it a vpi thing if you buy it from a dealer completely? Did you bring Legal action to the dealer who sold it to you. I don’t own any vpi products and don’t sell any audio products. But I have been to there New Jersey show room twice and events showing there tables , racks , phono pre amps. Harry is a very good guy in person and has given me solid advice on his and other products. His son Matt is about the most down to earth honest host and owner I have met. I don’t know where this went so wrong for you and I get your upset. But I’m guessing there is more to this as most issues are rarely all one sided. I’m not accusing you but since you keep going some one should push back a bit.
I truly hope you resolve your issue.

 

[Phoenix Engineering]

You have a point here. The NVS and most decent direct drive turntables do not generate apparently perceptible noise because they have a servo feedback system and supply what their designers consider the exact amount of energy to move the turntable at every moment. However as there are no perfect servo systems they always generate some noise, simply in zones of the frequency spectrum where classical instruments used to measure turntable performance are not sensitive.

I find this very interesting and relevant to the discussion of not only measurements, but verification of specifications and claims made by mfrs.

Any system with feedback will have a delay as it requires an error component to create a correction if the process variable (speed in this case) is not constant. The error term is usually run through a low pass filter to smooth the response, but even if it isn't, the inertia of the platter creates a time constant that cannot be eliminated, therefore any correction will take a finite amount of time to be effective. The control loop can be tuned to provide the best performance, but there will always be a trade off between fastest settling time and lowest noise. While the control loop can be tuned, it is constrained within a certain bandwidth dictated by the physical make up of the drive components: Platter inertia, motor torque, available power from the supply, encoder resolution and sampling time. To give an example, using a DSP based controller on a belt drive motor running at 600 RPM with a very low inertia rotor, the bandwidth is high and the control loop can eliminate speed variations in a 100µSec (10kHz) time frame, but this is not possible with a DD table with a heavy platter and a power supply with limited current capability.

It would be interesting if mfrs would give specifics about the control loop they use for a DD table ie: Loop bandwidth, settling time, % overshoot, sampling time, resolution etc., as well as the rationale for choosing the particular solution (lowest noise, best speed stability, most dynamic sound, etc.). While these specs would be difficult for the consumer to verify, reviewers could measure and verify and it does give the consumer a basis for comparison.

Some claims made by mfrs are easily verified or refuted. If a mfr says their DD table does not slow down with applied drag, it is easy to test for this by lightly touching the platter while playing a record and listening for a change in pitch. If the servo response time is too slow to compensate for such a slow pace test of drag (low loop BW), it also won't be able to respond to dynamic drag caused by variable groove modulation which occurs much more rapidly by definition. If a mfr is willing to stretch the truth on such an easily verifiable metric, I would be suspicious of other claims that could be more difficult to verify.

 

[Vienna]

Bro are you going to find any thread to pull into a vpi issue. I don’t know you and do get you had an issue. But how is it a vpi thing if you buy it from a dealer completely? Did you bring Legal action to the dealer who sold it to you. I don’t own any vpi products and don’t sell any audio products. But I have been to there New Jersey show room twice and events showing there tables , racks , phono pre amps. Harry is a very good guy in person and has given me solid advice on his and other products. His son Matt is about the most down to earth honest host and owner I have met. I don’t know where this went so wrong for you and I get your upset. But I’m guessing there is more to this as most issues are rarely all one sided. I’m not accusing you but since you keep going some one should push back a bit.
I truly hope you resolve your issue.

its not the right thread to answer you ”Bro”!