Phono preamp headroom - why?

[WDeranged]

A lot depends on the title.

I ran an LP mastering operation for about 20 years. What most people don't know is that a lacquer cut, if the cutter stylus is properly set up (correct angle, which varies from stylus to stylus, cutter depth, stylus temperature and so on) properly, it can cut a groove that is essentially dead silent. Play it, and literally you will wonder if the system is turned on or not until the music plays.

We did a few projects thru QRP. They found that by damping the pressing machines so they were more still while the vinyl was cooling caused a much lower surface noise. I had to talk to Chad Kassem directly (I"ve known him for over 30 years) to allow us to do our customer's limited run of 500 copies. The tests we got back were so quiet that just like the lacquers the surfaces were so quiet I was wondering if the system was on the right input when the music started.

Our source for the project was a digital file. You're not going to see those noise floors on larger runs with old analog sources.

I remember hearing Bernie Grundman say that vinyl has no noise floor and I thought wtf is he saying. I wonder if this is what he was talking about.

 

[atmasphere]

I remember hearing Bernie Grundman say that vinyl has no noise floor and I thought wtf is he saying. I wonder if this is what he was talking about.

If he said that, it jives with my experience.

 

[MaxwellsEq]

I remember hearing Bernie Grundman say that vinyl has no noise floor and I thought wtf is he saying. I wonder if this is what he was talking about.

Everything has a noise floor even if it's the random movements of atoms at anything above absolute zero.

 

[JP]

I’d love to see one concrete piece of evidence that a pressing can even be as low as claimed recently in this thread.

 

[MaxwellsEq]

I’d love to see one concrete piece of evidence that a pressing can even be as low as claimed recently in this thread.

Agreed. I think some of the low noise claims are associated with the cut of the lacquer, rather than a pressed LP

 

[levimax]

Agreed. I think some of the low noise claims are associated with the cut of the lacquer, rather than a pressed LP

Even for lacquer -80 dB seems like a stretch.

 

[atmasphere]

Even for lacquer -80 dB seems like a stretch.

As always, my advice is to try it yourself if in doubt. Or you could talk to someone else who also has experience mastering LPs.

Any mastering engineer worth his salt knows how important it is to set the system up to allow for the lowest noise groove cut. End users never get to experience that so no surprise they might be incredulous at how well the technology really works.

 

[levimax]

As always, my advice is to try it yourself if in doubt. Or you could talk to someone else who also has experience mastering LPs.

Any mastering engineer worth his salt knows how important it is to set the system up to allow for the lowest noise groove cut. End users never get to experience that so no surprise they might be incredulous at how well the technology really works.

Even the top of the line Techniques TTs have a rumble spec of -78 dB and that is A weighted. There may be quieter TT's but I don't think I have seen one with better than -80 dB rumble unweighted so not sure how you could confirm a -80dB laquer which is probably why I have never seen a measurement like that.

 

[TriodeLuvr]

I don't understand why A weighting would ever be used for a measurement where the main frequencies of interest lie in the lowest octave or below. Can someone explain this?

 

[levimax]

I don't understand why A weighting would ever be used for a measurement where the main frequencies of interest lie in the lowest octave or below. Can someone explain this?

It makes the published number look more impressive to consumers that don't really understand all the details. Nothing new, my TT from the 1970's published an A weighted rumble spec.

 

[TriodeLuvr]

It makes the published number look more impressive to consumers

I was afraid of that. Seriously, this means rumble components at 20Hz and below are at least 50dB worse than the specification. It all seems rather meaningless, particularly in the absence of measurements that should probably include a spectral display of the unwanted noise.

Jack

 

[Mnyb]

A weighting takes human hearing in consideration. That the weighting part ? But it gives higher numbers for sure .

 

[TriodeLuvr]

A weighting takes human hearing in consideration. That the weighting part ? But it gives higher numbers for sure .

The most significant problem with this is that it's not at all scientific in terms of defining turntable performance. Without knowing the frequencies involved, it's impossible to know whether the rumble present in a turntable with a more favorable number is actually less audible. This is just like the phono preamp headroom testing being performed here. Unless it's an order of magnitude or more worse than average, a single number is meaningless for comparing performance or value.

Jack

 

[Mnyb]

The most significant problem with this is that it's not at all scientific in terms of defining turntable performance. Without knowing the frequencies involved, it's impossible to know whether the rumble present in a turntable with a more favorable number is actually less audible. This is just like the phono preamp headroom testing being performed here. Unless it's an order of magnitude or more worse than average, a single number is meaningless for comparing performance or value.

Jack

Are not the A weighing supposed to track audiability ? And solve that conundrum. And give a number that tells how audible it is ? The weighting follows a curve that penalises you , if the frequencies are more audible ?
Unweighted seems even more problematic as the spectrum of the rumble is not considered at all ?

I don’t know if there are more factors at work ? A problem is often that different manufacturers and different reviewers does not perform measurements in the exact same way and makes a single number even more dubious ?

For example it’s not even specified if A weight are used ore not or how the measurement are performed with what signal ? I don’t know how this measurement is performed, are there different test records to use ? Or is there some other way ?

 

[atmasphere]

Even the top of the line Techniques TTs have a rumble spec of -78 dB and that is A weighted. There may be quieter TT's but I don't think I have seen one with better than -80 dB rumble unweighted so not sure how you could confirm a -80dB laquer which is probably why I have never seen a measurement like that.

Simple: don't use a consumer turntable. The lathe has a better noise floor.

 

[TriodeLuvr]

Are not the A weighing supposed to track audiability ? And solve that conundrum. And give a number that tells how audible it is ? The weighting follows a curve that penalises you , if the frequencies are more audible ?

Is it that simple in terms of system functionality? Isn't there more involved in this than human frequency response? Let's say rumble is listed at -75dB A-weighted, and that it consists solely of 20Hz energy. That would mean the offending noise is only 25dB down. Might that create issues for a system with sensitive bass reflex speakers if the woofers are effectively unloaded at that frequency? In any event, I would much rather see manufacturers publish the actual rms measurement and let me worry about my ears. That's how we rate amplifiers and speakers. No one skews the small signal frequency response measurement in order to comply with Fletcher-Munson curves.

Jack

 

[Mnyb]

That begs the question, what is the spectrum off typical turntable rumble ? I’ll guess mostly low frequency then weighted measurements would not be well suited , not the right tool for the application if audiability is not the primary problem ?

A curve of the rumble would be interesting as a complement too a single number even an unweighted one ?

 

[JP]

Weighted, but not A-weighted.

 

[Mnyb]

Weighted, but not A-weighted.



View attachment 505710View attachment 505711

Are these the weighing curves for DIN-45539 A and B ? or actual rumble spectra .

Must confess I'm unfamiliar with this .

But thank you

 

[JP]

Weighting curves