Wow and flutter - some figures

I agree with what you've said. Part of the problem is finding the correct belt with the correct tension, another part is the three point suspension which seems to have a definite impact if not set up 100% correctly, and the other part the horrible plastic pulley. I've never yet had mine running without a degree of audible wow and flutter, though an age spent servicing the deck and setting up the suspension did seem to help. I'm going to examine the motor mounting next, and also see if I can have a metal pulley machined somewhere.

The motor is meant to be low torque by the way, as in theory it's the heavy platter that keeps wow and flutter to a minimum. Even so, it should be strong enough to start the platter, though I got in the habit of giving it a helping hand.

All that said, it's a lovely sounding deck, and I've heard few better.

I got fed up with it, and put everything back on the Denon. I will probably just sell it off, as I need the money. My motor pulley is metal. It could be an earlier build. A shame, since I liked the sound of it.

It also ran too fast. They apparently could not figure out the correct drive ratio. I ended up putting rubber bands around the subplatter, which brought it back down to where it should be.
 
Not as good as its bigger brother 2000 (0.005%) but my GT-750: 0.006%.
I'm rather shocked by MMF & Pro-ject data. MMF especially....that's 5.1 & 9.1....not 2.1!
 
As I recall the BIC 960 had probably the best measured specs (wow and flutter and rumble) as measure by several of the popular hifi magazines of the time. I bought one based on those reviews. Problem was it was overly sensitive to movement. I remember that the tonearm could be made to bounce very easily if you touched the dust cover or plinth as the record was playing. The deep bass of the Infinity Monitor Jr speakers I had could send it off into acoustic feedback pretty easily. All that kind of negated those great specs.
 
Last edited:
Sadly the published W&F measurements are not useful. And most magazine measurements either, if they end in a single weighted W&F figure.

Those figures show some idea of the the *amplitude* of the flutter. As pointed out by the AR engineer, flutter is more or less troublesome depending on the frequencies where it's happening. This is the most important observation one could do. So the frequencies are also important, not only the global amplitude. Even worse if it' weighed. To apply weighting to flutter is simply... cheating. It is assuming that the ears will not notice flutter of certain higher frequencies.

Moreover, there are different weightings for flutter.

Then, the cartridge and arm itself do affect the reading of W&F! If the cantilever is too excited by the warp resonances (4-8Hz), then it will move with them and will do a FM (frequency modulation) on the signal, which will affect flutter measures.

Then there are some arms in which the VTF slightly changes with height. This may also have some effect on VTF.

All in all, it's not so easy as just collecting the weighted W&F numbers...
 
Sadly the published W&F measurements are not useful. And most magazine measurements either, if they end in a single weighted W&F figure.

Those figures show some idea of the the *amplitude* of the flutter. As pointed out by the AR engineer, flutter is more or less troublesome depending on the frequencies where it's happening. This is the most important observation one could do. So the frequencies are also important, not only the global amplitude. Even worse if it' weighed. To apply weighting to flutter is simply... cheating. It is assuming that the ears will not notice flutter of certain higher frequencies.

Moreover, there are different weightings for flutter.

Then, the cartridge and arm itself do affect the reading of W&F! If the cantilever is too excited by the warp resonances (4-8Hz), then it will move with them and will do a FM (frequency modulation) on the signal, which will affect flutter measures.

Then there are some arms in which the VTF slightly changes with height. This may also have some effect on VTF.

All in all, it's not so easy as just collecting the weighted W&F numbers...
I beg to disagree. The figures give an indication of wow and flutter, which is useful.

Since, however, you think them invalid, perhaps you would like to share with us what would be more appropriate to measure, and how? Never forget that W&F at a certain level is audible and can affect your enjoyment of records.
 
I beg to disagree. The figures give an indication of wow and flutter, which is useful.

Since, however, you think them invalid, perhaps you would like to share with us what would be more appropriate to measure, and how? Never forget that W&F at a certain level is audible and can affect your enjoyment of records.

Perhaps you are missing my point. But i can explain further.

Let's suppose you have a Garrard 301 with an idler wheel that is slightly twisted. It would probably measure about 0.2% on combined wow&flutter. But most of that amplitude (0.2%) is simply very low frequency wow, while the rest of the flutter components will be (and i can bet my life on this) less than 0.05%. After all, this is a tt with a heavy platter coupled with a motor of great inertia.

This 0.2% wow probably won't bother you unless you hear a sustained piano note that lasts for a full turn of the record. Because the flutter -that is, the speed variation on higher frequencies- is still very low.

Now, returning to the "weighing" thing... A belt-drive Technics SL-BD30, with its light platter and coggy, small DC motor, has 0.05% weighted W&F. A belt-drive turntable with a smooth and heavier AC motor and a circa 2.5-4Kg platter would probably measure 0.03%-0.05% under the same weighting, but you can bet there is an audible difference in speed 'smoothness' or 'timebase stability' or lack of 'jitter' or whatever you want to call it.

So, all in all, i'm starting to think that instead of looking at the weighted W&F figures, we can simply take a look at the construction of the turntable and have a good clue on how smooth it will be. As mentioned before, the arm&cartridge interaction influences, particularly if the combination is overly sensitive to the warp frecuencies (and arm&cartridge combinations are VERY different in this respect, because of differences in mass, compliance, and cartridge damping factor). Then, having examined the arm&cartridge, i'd say the speed smoothness will mostly be a function of the platter weight, motor inertia, motor 'coginess', and belt compliance (what i call 'belt bounce'), firmness, and/or slip on belt drive systems.

Idler wheel systems being similar to analyze with a quick glance of the eye. For direct drive, it is another story...

Of course, what i'm proposed is a purely subjective 'guess' on how smooth will a turntable be. For measurements, simply put: The reviewer/technician should use several test tones, at different frequencies, and give a graph of *linear* flutter vs frequency (that is, a spectrum plot), and this for each test tone. And then, note which arm and cartridge was used.

It's frequent to read posts with comments such as "specs are meaningless", "your ears are better than measurements", "use your ears", etc. It's a partial truth and a partial lie. It depends on what is being measured, how it is being measured, and how the final result is presented.
 
Last edited:
Perhaps you are missing my point. But i can explain further.

Let's suppose you have a Garrard 301 with an idler wheel that is slightly twisted. It would probably measure about 0.2% on combined wow&flutter. But most of that amplitude (0.2%) is simply very low frequency wow, while the rest of the flutter components will be (and i can bet my life on this) less than 0.05%. After all, this is a tt with a heavy platter coupled with a motor of great inertia.

This 0.2% wow probably won't bother you unless you hear a sustained piano note that lasts for a full turn of the record. Because the flutter -that is, the speed variation on higher frequencies- is still very low.

Now, returning to the "weighing" thing... A belt-drive Technics SL-BD30, with its light platter and coggy, small DC motor, has 0.05% weighted W&F. A belt-drive turntable with a smooth and heavier AC motor and a circa 2.5-4Kg platter would probably measure 0.03%-0.05% under the same weighting, but you can bet there is an audible difference in speed 'smoothness' or 'timebase stability' or lack of 'jitter' or whatever you want to call it.

So, all in all, i'm starting to think that instead of looking at the weighted W&F figures, we can simply take a look at the construction of the turntable and have a good clue on how smooth it will be. As mentioned before, the arm&cartridge interaction influences, particularly if the combination is overly sensitive to the warp frecuencies (and arm&cartridge combinations are VERY different in this respect, because of differences in mass, compliance, and cartridge damping factor). Then, having examined the arm&cartridge, i'd say the speed smoothness will mostly be a function of the platter weight, motor inertia, motor 'coginess', and belt compliance (what i call 'belt bounce'), firmness, and/or slip on belt drive systems.

Idler wheel systems being similar to analyze with a quick glance of the eye. For direct drive, it is another story...

Of course, what i'm proposed is a purely subjective 'guess' on how smooth will a turntable be. For measurements, simply put: The reviewer/technician should use several test tones, at different frequencies, and give a graph of *linear* flutter vs frequency (that is, a spectrum plot), and this for each test tone.
So, in other words, you can't really suggest any tests.
I'll stick with the accepted methods, thank you.
 
So, in other words, you can't really suggest any tests.
I'll stick with the accepted methods, thank you.

What's the matter with you Avole? Always the same attitude, month after month.

How about you actually reading what i post? This is what i posted above:

"For measurements, simply put: The reviewer/technician should use several test tones, at different frequencies, and give a graph of *linear* flutter vs frequency (that is, a spectrum plot), and this for each test tone."

I just suggested how the tests should be done. It doesn't require many words to be described: If you knew how the standard DIN flutter test is done, you would know what i mean with "different frequencies" and "linear".

And now, i'll step aside and let other well-informed people give their views on this issue.
 
Not that I consider myself well informed, but I have some thoughts. One of the reasons for the DIN tests to be performed, as they are designed, is they do not consider high frequency variation a possibility, a platter should always have enough inertia to withstand frequencies above 200Hz. If this is the case I don´t know, but it sounds reasonable to me.

I agree with Flavio that presented figures could, in principle, be anything the supplier like to present, if they like to present any at all, and should be looked upon with a healthy degree of scepticism. If they are not provided from a totally indipendent institute.

It is also so that a good tt today should IMO have much less W&F than the record, and there lies one dilemma in measuring.

Personally I think that achieving good figures with a belt drive is tricky, and that is reflected on presented figures, or not presented.
gusten
 
It is also so that a good tt today should IMO have much less W&F than the record, and there lies one dilemma in measuring.

I am not so sure, if you take into account that record cutting lathes have hugely massive platters, easily more than 20Kg of mass!! As a comparison, the technics SP-10MKIII (turntable) platter had an amazing 10Kg of mass, and a huge motor to move it, but it is small compared to the Technics SP-02 cutting lathe motor it was derived from!!

For Scully lathes the belt drive system was massive:
scul-mot.jpg
 
flavio would that mean even published W&F of say CD players is meaningless unless 'several test tones, at different frequencies, and give a graph of *linear* flutter vs frequency (that is, a spectrum plot), and this for each test tone' are used?
 
flavio81
I believe a weighed figure of 0,03% W&F is about the best there is on a test record. (if I remember correctly) Probobly most are not that good.
Significantly higher on a normal record taking into account all different figures in the chain added together.
gusten
 
What's the matter with you Avole? Always the same attitude, month after month.

How about you actually reading what i post? This is what i posted above:

"For measurements, simply put: The reviewer/technician should use several test tones, at different frequencies, and give a graph of *linear* flutter vs frequency (that is, a spectrum plot), and this for each test tone."

I just suggested how the tests should be done. It doesn't require many words to be described: If you knew how the standard DIN flutter test is done, you would know what i mean with "different frequencies" and "linear".

And now, i'll step aside and let other well-informed people give their views on this issue.
Flavio my friend, you criticised a long accepted method for judging how a turntable performs. I asked you to suggest a better alternative, which you were unable to do, except to say rather vaguely feel the quality, if I may paraphrase the TV show, then , as aside, waffled vaguely about test tones.

That, to me, says you had nothing to suggest, and, frankly, your personal attack seems to indicate such is the case.
 
avole: Actually some of the hifi magazines over here (Audio, Stereoplay) have already been using w&f spectrum graphs for quite a good while - usually for only one test frequency, though. I like these graphs for pretty good visualisation of the actual distribution. And I'd expect good manufacturers to use these, too - 'cause with the spectrum graphs it's easier to indentify sources of influence.

That said, I also like the good old DIN method, as it's a real world scenario measurement - although, als gusten already wrote, probably limited to some 0.03 % at best on everything but the very few tables that allow for an optimisation of the centering. Whereas I don't really like those drive-only measurements using signals of the motor control, as these can be quite misleading - that's a bit like spec'ing the maximum speed of a car in free fall instead of on the road... ;)

Greetings from Munich!

Manfred / lini
 
Lini, it's actually this part of the post I thought meaningless:

"So, all in all, i'm starting to think that instead of looking at the weighted W&F figures, we can simply take a look at the construction of the turntable and have a good clue on how smooth it will be. As mentioned before, the arm&cartridge interaction influences, particularly if the combination is overly sensitive to the warp frecuencies (and arm&cartridge combinations are VERY different in this respect, because of differences in mass, compliance, and cartridge damping factor). Then, having examined the arm&cartridge, i'd say the speed smoothness will mostly be a function of the platter weight, motor inertia, motor 'coginess', and belt compliance (what i call 'belt bounce'), firmness, and/or slip on belt drive systems.

Idler wheel systems being similar to analyze with a quick glance of the eye. For direct drive, it is another story...

Of course, what i'm proposed is a purely subjective 'guess' on how smooth will a turntable be. "

Looking at the construction of a turntable isn't going to tell you anything, other than whether it is well built or not. According to flavio's method, the Clearaudio Emotion would probably, and as he admits subjectively, score well, yet the w&f figures aren't especially good.

I queried this with Clearaudio who confirmed the figures and explained the problem was the motor, which they have changed for the newer model.

Another example is Pro-ject, whose lower-end turntables appear reasonably built with decent Flavio factor, yet again have w&f figures that are less than good until you use the speedbox.

Of course, he is correct in what causes w&f, though he missed centering of the hole in the record, but to say construction will reveal information about w&f is either wrong, or misleading. Yes he kind of mentioned spectrum graphs, but I took this to be a throw away line rather than an acceptance of the importance of measurements.

As it stands, w&f figures however measured are the best guide. Also, if you are someone who is sensitive to this, then the figures are all the more important.
 
avole: Well, you're certainly right that in the end the proof will be in the pudding - nevertheless an experienced cook would already have a good idea about the potential quality of the pudding by just checking out the recipe and the quality of the ingredients...

Greetings from Munich!

Manfred / lini
 
I believe it´s well documented (by Bruel&Kjaer) that when playing a normal record, the W&F coming from the record is the largely dominating factor.

That is not to say that if the tt´s contribution would be zero we wouldn´t notice a difference, we most likely would, at least with some tt´s.
gusten
 
avole: Well, you're certainly right that in the end the proof will be in the pudding - nevertheless an experienced cook would already have a good idea about the potential quality of the pudding by just checking out the recipe and the quality of the ingredients...

Greetings from Munich!

Manfred / lini
Not really. You assemble the pudding from the ingredients using the recipe as the design. You'd have to disassemble the turntable, then check the quality of each component, which might cause some raised eyebrows in the shop:)
 
I'm in agreemnt with those that suggest the record itself is the determining factor for W&F audibility, namely through warp and eccentricity and the resulting interaction with the tonearm/cartridge suspension.

Published figures are useful in establishing manufacturing benchmarks and for bragging rights, but in the end they pose little on what the consumer will hear. Personally, I've never bought a turntable based on these specs.
 
If I look at the specs for my Pro-Ject Perspective II it seems like a woeful .15%.

I added the Pro-Ject Speed Box II and it's specs say "Quartz oscillator Speed stability +/- 0.01%"

The funny thing is I never noticed anything wrong prior to the Speed Box, it did "tighten" up the sound once I got it though.
 
Back
Top Bottom