advertisement


Turntable speed analysis

Status
Not open for further replies.
It's certainly a different spectrum to the Kuzma. It's a shame we cannot eliminate the recording from the measurement. I'd not make any statement about the level of the higher frequency components of the wow though.

Paul
 
I had a quick go at looking at the wavelet power spectrum in the hercules wav file - let me know if this is of interest and I can email a figure. Not really sure what it's showing, but it's interesting to see the 50 Hz signal come and go with each revolution.

Dan
 
I'm curious how the 50hz is manifest, I don't see how it can possibly get past the belt as the pulley/platter ratio would surely eliminate it. Ie any 50 hz from the motor is divided down by the pulley/platter ratio, surely. Might it be radiated from the motor or somehow cross coupled into the signal during transmission?

There's plenty enough wires hanging down round the back of my deck for that to happen.
 
The 50 Hz signal clearly synched with the platter rotation, so my guess is it is coming in from the belt. Perhaps as the load varies the motor is asked to do more or less work and the 50 Hz noise depends on how hard the motor is working?
 
I'm curious how the 50hz is manifest, I don't see how it can possibly get past the belt as the pulley/platter ratio would surely eliminate it. Ie any 50 hz from the motor is divided down by the pulley/platter ratio, surely. Might it be radiated from the motor or somehow cross coupled into the signal during transmission?
The 50Hz spike in my plot up thread is a 50Hz modulation of the platter speed, it can't be radiated by the motor or due to other interference, it must be a variation in the rotation. There is also a spike at 100Hz, but less obvious.

For example if the phase splitter weren't optimally tuned then you might expect this.

I don't understand dan's plot yet.

Paul
 
I don't understand dan's plot yet.
Paul

Don't understand what I am plotting or what the plot says? The x dimension is time (~12 secs worth but plotted in platter rotations) and the y is frequency. The contours represent power spectral density. If you were to average across time, you'd roughly get what you would see in an fft. It's like a rolling window fft. Apologies if this is obvious. Here you see that the power in the 50 Hz band comes and goes with each rotation - the pattern sort of repeats. Interestingly, there's also a lower frequency (10-20 Hz) spike that seems to precede the 50 Hz spike.

Now what it all means, I don't know!

Dan
 
50Hz (and 100Hz) interference may masquerade as phase noise; even the very best phase detectors have some amplitude to phase conversion, and the cartridge itself is mildly non-linear, so can do this as well.

A simplistic example may help. Imagine the phase detector is just measures the positive going zero crossing time of the 3.15kHz sine wave - I know the real one is better than this, but bear with me. If you get interference that adds to the signal, the crossing is early; if it it subtracts it is late. So simple additive interference can masquerade as phase modulation. Now, the algorithm in use is lot better than timing of zero crossings, but the effect remains at some level.
 
Well that's problem solved, yes this sample was from the Herc and Geddon when both of them used the supplied .22uf capacitors. The Geddon now has 2 x 0.1uf caps and turns more smoothly as a result, bass timing is much improved audibly. I suspect the Herc would similarly benefit from this same swap.

Dan, so that 50 hz band is actually akin to the visible slow speed variation that you can see in Paul's plot. if you look at the bottom right quarter of his rotation plot you can see more than a hint of the humps coming from the motor poles.

All good stuff. I have yet another power supply waiting in the wings with adjustable voltage and quadrature.
 
50Hz (and 100Hz) interference may masquerade as phase noise;
The argument against that being the case, in this case, is that the Armageddon clone result doesn't show anything like as big a 50Hz component.

Also I have some results from my own turntable that show the 50 and 100Hz components being affected by a purely mechanical modification.

I will try an experiment of synthesising a low frequency interference and see what happens...

Paul
 
B is slower than A. Ah ha. But C and D are more or less the same as A. Boo.

On closer examination C, which should be the most draggy, is running minutely faster than A.

The demodulated output of 'D' looks like,

YNWOAN_D_demod_base.png


The dominant effect is the eccentricity, I can't, at the moment, see where the drag change occurred.

So not getting very far at the moment.

I've an idea for loading/unloading the drive of an LP12 without having to use another arm, which might be interesting, but really we need to get the measure of stylus drag.

Paul
 
Bear in mind he's running 'high drag' in the bearing already, that may well queer this pitch.
 
Paul, any chance you get around my 78rpm pre-warmed file?
Sorry for pushing so hard but there was a concerete method proposed (78rpm warmup), I've done it, now we miss the last step-check if it works :)
Post #567 on page 38
Thanks!
phi
 
The idea is that running the idler at speed (78rpm) for a protracted period helps to iron out wheel irregularities; I would say that the evidence (provided by your files), does not support this hypothesis.
 
Oh I see, and we were just expecting it magically deform back to an equal shape despite years of use because it had warmed up a few degrees under load. Quick you better tell the F1 teams that they don't need to waste money on tyres, they just need to keep driving them and they'll fix themselves... :sarc off:

The method proposed utterly defies logic...
 
Paul, many thanks! Dopn't know how I have missed your post...sorry.
Indeed the warump did not work here at all.

Sq, this is not F1 but a turnatble, mind you, so a certain logic may or may not apply
 
Status
Not open for further replies.


advertisement


Back
Top