FM Acoustics and their Phono Master

[Gordon]

I have seen ads for this thing on the back of mags before and aften wondered about what it's all about. Today I got the latest Classical Recordings Quarterly and there again is the ad. Also in the issue is an article about the vagaries of pressing vinyl. So, I thought, here we go, get those erudite chappies on the R3 boards to advise.

Given that good design and set up is imperative let's assume that they have done all the right things for vinyl pre-amps etc. They claim to "linearise" vinyl playback without using any digital means. Fair enough. Anyone any idea how they are doing this, it's purely analogue. Whatever they do "the signal remains pristine". Hm, how can it if you've fiddled with it? Sounds like a fancy box that has knobs on it and you fiddle with them until you hear something you like. Where's the fidelity?

Something about the philosophy of this device suggest that there is something fundamentally lacking in the vinyl system, so bad it has to be put right with this box of tricks. Or am I being unkind? Or just provocative?

They also remove clicks and pops etc [OK, but why should they have to] and what is there to linearise?

Do they remove distortion then? I can see how that might work IF you can reproduce the inverse of the original distorting process, which of course comes from many sources and as such are unknown with any precision, and can add back an out-of-phase approximation to the distortion thus reducing it. That method is a well known one used in eg satellite systems that use travelling wave tube amplifiers that are notoriously non-linear. But eg tracing distortion isn't constant across the disc so how is that done? What they do won't be exact but perhaps it doesn't need to be.

Sounds very interventionist to me - the anathema of vinyl purism surely? Anyway I've never heard one of these things so can't really comment one way or the other. Anyone got one? If so does it work?

I haven't found any patents, yet. Then again if it is so marvellous don't publish but risk letting others reverse engineer.

Here is the web site of the company in Schweiz:

http://www.fmacoustics.com/set_domestic.html

A few points taken at random: "Ultra accurate RIAA": well what makes you think the disc cutter was ultra-accurate so what can you do about it? The vagaries of all the devices in the vinyl chain suggest that this is the least of your worries. You keep telling us that the record companies didn't use it anyway!!

"Class A output line drivers guarantee no matching problems to different electronics." Er, so what, don't most pre-amps do that?

"Special dynamic curve-tracer analyzed semiconductors used throughout entire unit." Wow!! That'll stop 'em in their tracks. Must be good then. What on earth for? Most engineers design OUT the devices as much as possible.

"Freedom from usual limitations: no signal degrading IC's, transformers, hybrid circuits or op-amps." well, no-one would disagree with that. Indeed it would be esssnetial for a truly audiophile product.

"Absolutely transparent" Hang on, don't you process the input signal from the cartridge etc and thus change it? That implies the cartridge is lying. How is that transparent? Don't you mean "remove all the defects of the playback process and give out what is inherently cut into the disc"? Well, in that case, how can you say it's "absolute" when you don't know what that inherent thing is because you dont know exactly what has been done to it on the way from the groove and what those defects are? You tweak some knobs until you like what you hear again.

[Bryn]

Provocative or not, nail hit squarely on the head.

[Stunsworth]

It's a phono pre-amplifier, presumably to use with amplifiers that don't have an input suitable for turntables. Amplifiers that have inputs for turntables have something similar already built into them, though without the flexibility that this unit offers.

Variable cartridge loading - of resistance and capacitance is not new, it can have a direct influence on the frequency response. I used to have an Ortofon VMS 20 cartridge that was best with a non-standard capacitive loading.

Quad amps from the 50s had selectable RIAA curves to cope with the different 'standards' at the time, and I think there were modules for the Quad 44 that allowed the user to 'tweak' the turntable to pre-amp interface.

[Gordon]

Quad amps from the 50s had selectable RIAA curves to cope with the different 'standards' at the time, and I think there were modules for the Quad 44 that allowed the user to 'tweak' the turntable to pre-amp interface.

I know, I've still got one!! Spent ages fiddling to get the output of the 44 "flat" from a test disc. Not used much now, not since I CD'd most of the vinyl befoe I got rid of it. Still have a few favourite LPs though that will never see the light of CD day.

My confusion is really over their claim of "linearisation". Their box is an all singing all dancing pre-amp-equaliser, designed for flexibility and, apparently, engineered to a very high standard. I'm coming to believe that it does not linearise at all, it is just an equaliser, a glorified tone control.

[Dave2002]

1. It seems to be a phono preamp with selectable RIAA curves, and maybe other tweaks. In itself not new, such things were available 50 years ago.

2. It may also remove crackles. Michael Gerzon and a few others had circuits which could do this without using digital audio channels. However they couldn't strictly be claimed to be non interventionist, or work without feed back or feed forward. They used this to clean up 78s by playing them with a stereo cartridge, and detected high energy in the L-R component.
By incorporating delays into the L+R component they tried switiching out the audio when a click appeared. This wasn't ideal though, as sometimes the silent gaps could be noticed. To fix this a way of inserting a similar section of music audio has to be found. With a digitised signal this isn't too difficult, and can presumably also be done by tapping off sections via an analogue delay line. Definitely interventionist, though allegedly works very well. I never heard the system.

For stereo LPs the easy way of detecting clicks doesn't work, but it should still be possible to use other methods, and then switching of the analogue audio chain (x2) could be done.

The Phonomaster device may actually be quite good, but the claims made for it may be false. Another trick whcih could be used is to widen the dynamic range, which would push the noise floor down, though would introduce extra distortion. There was an analogue system which compressed the audio going on to discs and decompressed the discs on replay. At a time when surface noise was a significant issue this did reduce the background noise. I did hear that system and it worked, but it didn't really catch on.

Switching signals in the analogue domain can introduce clicks so the merits of doing this have to be considered carefully.

Probably easier to do it all digitally after all! Some people just don't want to have digitised audio at all, and it is possible to do quite a lot with switched analoue circuitry. Whether the results sound better than using digital may be difficult to assess. A very good switched analogue system might sound better than an average digital one. I can't say much more than that. Comparisons should be made between equipment and systems of similar quality whether analogue or digital.

[Stunsworth]

Garrard had a click removal module in the 70s. The problem was its effect on the music was audible. As you say it's something that's better done in the digital domain. To see how well some software can do this, take a look at this chap's site and click on the audio samples to the left...

http://www.clickrepair.net/

I remember the dBX system for expanding the dynamic range. I think it only worked with dBX encoded LPs.

[Gordon]

Yes, well, we've had a number of systems for dealing with the issues in analogue channels. Dynamic Pre-Emphasis was one such - otherwise called Dolby B, C etc. There were some other proprietary ones as well that were full bandwidth [Dolby A was professional and wideband Dolby B, most common, was not]. However, this involves something being done at the "send" end of course whereas this Phono Master device doesn't do that. Wideband noise reducers/dynamic range expanders like DBX didn't catch on as you say.

Thinking about analogue means of detecting clicks etc one merely has to differentiate the audio [trivial in principle, it's a low pass filter] to see high peaks when clicks are likely because a click will cause a sudden and rapid "spike" that will be out of the bandwidth of normal audio. Best done before the RIAA correction. This can be done L and R only as well as the L+R and L-R to cover the bases. Any large spike, whose threshold can be set manually or automatically, can be used as an indicator that a click or a crackle is present and so cause an appropriate gain reduction or a muting in a subsequent amplifier stage. This need not be audible and, to be fair, they don't claim it to be perfect.

This idea was used by at least EMI in checking LPs in the factory - they also played the discs backwards, ie from inside to outside - so that natural sudden sharp changes in amplitude were avoided. Can't so that in this Phono Master device!!

[Dave2002]

This idea was used by at least EMI in checking LPs in the factory - they also played the discs backwards, ie from inside to outside - so that natural sudden sharp changes in amplitude were avoided. Can't do that in this Phono Master device!!

I guess this idea could still be used by anyone or any organisation which wants to remaster old disc material. It might improve the tracking of the playback system before digitising, though it might cause other tracking issues. Cartridges were designed for a particular way of playing back and dynamic behaviours would change if these are varied. Did EMI have special decks for playing in reverse, or did they simply reverse the motor? The effect on the stylus/cartridge/arm might not have been very good.

Another idea would be to replay at lower speeds, though that could introduce other problems - speed accuracy etc., though apparently it is possible to correct for some speed errors digitally nowadays.

Perhaps a near ideal digital copy could be made by combining both forward and backward played discs in some way - maybe better than either individually.

It's possible to do so much more once a good digital copy has been made that the case for doing things in the analogue domain would have to be pretty good now. However, not all copies from analogue sources are good, and not all digital versions are good, so some might still feel it's worth working with analogue. Most end users probably don't want to bother. Working in real time with analogue might be good, as there could be an almost instantaneous feedback to the listener, but this assumes that the listener doesn't want to have any digital processing in the audio chain. There as still some audio enthusiasts (fanatics?) who claim that vinyl and analogue are better - and in fairness I have heard some extremely good analogue systems and analogue recordings. I suspect that if money is no object that very good results can be obtained with either, or both, or hybrid systems. For most people, most of the time, digital systems provide much better bangs per buck IMO.

[Gordon]

Thanks Dave. Here is where you'll find out about EMI's disc testing at Hayes:

http://www.amazon.co.uk/Hayes-Record-Processes-Manufacture-Development/dp/0952098407

and here is a scan a a couple of pages, as the book is unavailable at the moment, which talk a bit about the kind of pressing defects that caused the clicks. They obviously used Garrard 301s running backwards and with the pickup on the far side. Also they used real people which was expensive which is why they used the machines in the foreground of the picture.

Interesting that all the defects to do with injection moulding of vinyl apply to polycarbonate CDs and so there must be some issues in common. Worn stampers can leave behind poorly shaped pits and poor aluminium coating can leave holes all leading to poor data recovery that is left to the error correction to deal with. Worst case of course it doesn't and we get splats etc. but minor drop-outs could end up being short enough not to be obvious. Why dont CD player manufaturers put a little lamp that lights if the Error control fails? Very useful to know of you have a dodgy disc or that your laser life is coming to its end.

[Dave2002]

Why dont CD player manufaturers put a little lamp that lights if the Error control fails? Very useful to know of you have a dodgy disc or that your laser life is coming to its end.

I think I have seen some CD players with extra info, perhaps including error rates, but not for a very long while. My first CD player had a light which came on after about 5 years when I put in a Japanese CD which had pre-emphasis. Of course it went off again when I took the disc out again.

Thanks for the detailed post to be savoured later.

[gradus]

The Hayes article is impressive and something of a surprise whern you consider how poor EMI pressings were generally.
Whatever EMI may have done at Hayes was not typical of the industry. EMI pressings made by Phonodisc/Polygram were statistically sampled for pressing defects but no department like the one described in the article existed. The same factory produced mainly Philips, Fontana and some DGG, none of which were consistently of particularly high quality and imv this was also true of Dutch Philips and German pressed DGG. Decca reordings were frequently spectacular but thefactory often produced poor quality pressings; CBS and RCA pressings were worse.
As for the 'independents'....!
In my experience one was lucky to get a vinyl LP that played flat and was free from pressing/handling faults. No wonder the CD was welcomed so warmly.

[clive heath]

Digital procedures can remove clicks and one way they do it by the "interpolation" technique. You select the area of the waveform where the click is visible ( see above) and on command the computer performs a fast fourier transform of the correct wave on either side of the click determining the main frequency components of that small time segment of music. It then estimates how the corrected waveform would look and inserts it. From my experience the click becomes inaudible.

Since digitisation like this works so well, my own phono pre-amp is just a choice between RIAA and DIN equalisation curves although I have introduced a little balancing circuit to make sure the signal of a mono LP is exactly the same at both pre-amp outputs so that you might expect the best possible separation between the two stereo channels with a stereo LP.

http://www.cliveheathmusic.co.uk/vinyl.php

[Dave2002]

That article about the Hayes factory is very interesting, and a great surprise. I drove past the factory on occasion - a back road round the Uxbridge/Hayes area near the London to Reading railway line. I had no idea that it really was were the LPs were made and tested. I also have vague recollections of having had to visit someone there once, in the computer department.

It took me a while to check out the images of the turntables. The basic units seem to be Garrard 301s though the direction of rotation has been changed, while the arms are specially made since the offsets are to the left rather than the right. Those pickup arms seem to be very long.

The whole operation appears to have been much more labour intensive than I realised. Presumably this was because faults were so common - I had assumed that post production batch testing would have been sufficient. Also note that the runs for classical music were so small that they tested during the runs.

Were the tested LPs only used within the process - or were those that passed the tests packaged for sale? Presumably the playing in reverse only applied to the automated testing which was brought in some while after the development of LPs. Human listeners might have found it strange.

[Gordon]

I nearly ended up working for EMI at Hayes when I graduated. I remember the interview and the tour. Didn't go there in the end. My In laws had a friend who did work there but in EMI Electronics and for a while we got LPs through him at a big discount. I seem to remember the risk was that you could get the discs that had been through the tests. He gave me a tour of the pressing plant one weekend in the early 70s just before the oil crisis hit vinyl production. That and a visit to Abbey Road around the same time. Quite by chance when we got married and had a flat in Harrow the young guy in the next door flat worked as a technician at AR and got me an informal visit to the techy dept to see the equipment and how it was set up. Quite an insight. Happy days.

Those arms were long - quite why is a puzzle - and had extensions from the same pivot bearing that linked into the recording box at the bottom R of each turntable. When the click was detected this marked some kind of log in that box that located the blemish so that it could be examined afterwards.

The type of faults are listed in the text, some related to too rapid a pressing cycle. Note that the presses were not operated in a clean room as is used for CD!! Given the dimensions of the quiet grooves in relation to say dust particles - a matter of microns, human hair about 50 micron and stylus tips around 12 - the need for very fine vinyl composition - ie no lumps or small granules, the polymer charge was a finely ground powder that flowed under heat and pressure - was great and so apart from pressing faults, like the metals adhering to the plastic, there were issues with the smoothness of the vinyl compound and the cutting tool's ability to leave a smooth wall without abrasions. Cheap re-ssues used reground discs requiring a fine degree of compound processing which of course did not always happen. Reading that description of the process makes you wonder how anything decent ever got pressed!! That whole book I gave a link to is a good read if you can find one somewhere.

Warps were caused by removing the still warm disc too soon with consequent too rapid cooling without support allowing the thermal stresses to bend the disc. People used to put their warped LPs into the oven [!!] between glass sheets and then bake them and allow them to cool very slowly. Sometimes it worked and the warp was reduced - my experience of this was that although this did happen there was a low level rrrff sound when the stylus passed where the warp had been. Some warps and "dishing" were dealt with by means of a clamp.

My real bug bear was eccentric centre holes that produced wow, sometimes one side was OK and the other not. I had to ream out the centre holes of quite a number of discs and then use a clamp to get rid of the pitch variations. Decca was a paticular culprit, my set of Solti's Gotterdammerung had several clearly audible misaligments like this. Ah!! the joys of vinyl eh!

[Don Petter]

Those arms were long - quite why is a puzzle - and had extensions from the same pivot bearing that linked into the recording box at the bottom R of each turntable.

The longer the arm the less the geometric (angular) tracking error when compared with the ideal of the cartridge always being tangential to the disc, since the cutting head moved along a radius. That is why SME, for example, offered a 12" arm as well as a 9" one.