Low Cost Turntable - Incredible Performance

Or from Project the Xtension 12 Evolution with 12" EVO tonearm

@oberoniaomnia 

Thanks for the 4 Hz cut off value. But what does that mean in reality? 4 Hz = oscillation 4 times per second, but of what? A reference tone, say a’ = 440 Hz. But then what does 0.1% of >4 Hz mean?

You are on the right track, but it gets more complicated!  The reference tone as set out in measurement standards is 3150-Hz.  The % wow and flutter is the maximum deviation from 3150-Hz on a weighted scale.  The weighting is centred on 4-Hz and drops steeply on each side.  3150-Hz is chosen so it can be easily played on most devices!

I would have expected some variation of the platter rotation (33.33/45 RPM) to be the metric of wow and flutter. but then we need a time dimension (seconds) to differentiate short from long term variability.

You are right again!  Low frequencies or wow roughly correspond the with platter rotation speed, or the wobble of an off-centre record (2-Hz).  Look at the frequency spectrum of what is played from the test record, and you will see 3150 smeared out, between say 3148 and 3152.

If only an off-centre record was to blame, the frequency would be spot on twice per revolution (2 x 2-Hz = 4-Hz, right where the weighting curve is highest), and way off in between.

Direct drive turntables tend to have low levels of wow because of the rotational mass and feedback controls of rotation speed, when compared with belt and rim drive.

Flutter is higher frequency, like say the motor speed of a belt drive or rim drive turntable.  The frequency response will show extra frequencies further from 3150.

Direct drive tables tend to have worse flutter than belt drive, because direct drive uses impulses to rotate the platter while elastic belts tend to absorb vibration from higher speed motors.

Of course, the frequency spectrum will fluctuate over short time frames, so the standards say use at least 5 seconds.  Measure the maximum difference over at least that time, after applying the weighting curve from the standards.  So you will have a difference in frequencies which you can convert to a percentage of 3150-Hz.

Oh, if you actually want to read a standard, be prepared to pay around $50 to get a copy.

Either there is more to it, or it does not make any sense. If you (or anybody else) can provide some clarity, that would be great! Thanks!

I hope this has helped explain the measurement and why wow and flutter are combined these days.  But so much for measurement, what can we actually hear?

Surprisingly, it depends on whether we are in an anechoic chamber or a reverberant environment.

Are you familiar with the concept of frequencies ’beating’?  If two frequencies are close enough, you might hear a much lower frequency as the signals reinforce and then cancel.  This interference phenomenon is a cheat’s way to get low notes for an organ.

In a real listening environment, we hear direct sounds and reflections.  There is a chance that a sustained note from say a piano or a clarinet will interfere with reflections of itself.  If the note is subject to flutter, when the reflected flutter is out-of-phase with the direct flutter, you may hear unnatural beats.

These are more likely to annoy classical audiophiles that those raised on jukeboxes full of pop music.

The wow and flutter specs on my Denon is .008% 😎

@richardbrand Thanks for taking the time!

I still do not understand the 4 Hz. Table rotation is 30 RPMinute = 30rp/60s = 0.5 rps = 0.5 Hz. Off center record spinning should be correct 2x revolution (thanks for this!) so record wobble frequency is at 1 Hz, not 2 Hz.

If the test tone 3150 Hz frequency is sampled with a measuring frequency of 4 Hz, so 4 times per second, one gets 8 data points per revolution: 2 at minimum speed/tone frequency, 2 at maximum speed/tone frequency, 4 at in between speeds/frequencies. That then measures the precision of the average speed, how close the platter rotates to the normative 33.33. Given that there is no perfect record, one has to take the wobble of the record out to assess the actual TT performance.
It also suggests that the % metric is derived from a Fourrier transformation of the received tone signal. Then I can see how you can look at maximum deviation above and below 4 Hz in the transform. It appears that it is critical to make the distinction of tone frequency and measuring frequency. 

The deviation above 4 Hz in the transform is microspeed variation of the TT.

Re tone variability of 3148–3152 Hz, so delta of 4 Hz, that is <1 cent (800 cent = 1 octave = doubling of frequency). In sequence, indistinguishable, even to trained musicians. On a good day, I can do 3–4 cents when tuning. <1c is in comma territory, and if that would be objectionable, then any equal tempered piano would sound bad. [It does, but that is my personal dislike of the instrument and preference for HIPP baroque music].

Well familiar with interference. Easy to generate on a violin. Interesting concept re flutter affecting sustained notes. However, considering a perfect TT, direct and reflected sounds are also out of phase due to different travel distances. The phase difference has nothing to do with flutter, only with direct vs. reflection, plus there are various reflections (side, floor, ceiling, back wall ...). Not sure I buy that rationale, but a fun thought experiment.

Additionally, in classical music, sustained notes are smeared with vibrato (as opposed to note shaping with messa di voce in HIPP baroque playing). So any microspeed and reflection variation is lost due to the dreadful vibrato sauce. 

I happen to still believe you’re better off investing in some of the used direct drive tables of the 80s and 90s

^this. 

There are not many recognized experts  on the care and feeding of vintage Japanese DD turntables.

@mikevanloo @lewm 

Here is one. If you live in Perth, Australia, he’s just a short drive away.

https://liquidaudio.com.au/