Return to Home Page - www.GMweb2.net

The 'ACE' a vinyl Phono preamplifier

for MM cartridges.

By Graham Maynard                           graham.maynard1@virgin.net

 

Notes.
Back in the 1970's I became disheartened with audio 'progress', for in spite of amplifiers presenting much superior specifications, the mainstream transistor offerings sounded far worse than the tube (valve) designs they were supposed to replace. KT88s made better power amplifiers than transistors, but transistor circuitry could be much better and quieter for pre-amplifiers. Integrated circuits had also developed well, but these were often thought of as offering a lesser performance when compared to discrete components.

For those who might wish to 'save' a digital copy of their own vinyl record collection to a computer hard-drive, I have redrawn my own 1975 vinyl pre-amplifier circuit. It remains capable of optimising disc playback and offers features still not repeated today.

Many designers opinionate that NFB loop controlled amplifiers are inferior because they degrade the sound, and yet I wonder what it is that these individuals think they actually listen to?  The fact is that most vinyl waveforms and CD pits, as already pressed into their own disc collections, were mastered through a myriad of NFB loop controlled pre-amplifier and mixer stages, and this long before anyone can start listening.  Indeed, the Mullard developed TDA1034NB integrated circuit I used here and which initially cost me £67 for ten, went on to be used by the hundred within what became a classic series of world famous Neve mixing consoles.

Properly designed NFB loop controlled IC gain stages can set standards for excellence, and thus I separately list five elements embodied within this thirty-plus year 'old' pre-amplifier design.

(1) The moving magnet pick-up cartridge uses an inductive transduction coil which must be both reactively and resistively tuned and damped in order to optimise reproduction.
In order to optimise for any cartridge I fitted a sub-miniature twin gang 500pF variable, plus screened and earthed twin gang potentiometers to the input circuitry
It is only *after* you have actually used these input damping and tuning controls whilst listening to music that you can best optimise for your own equipment line-up, and, for individual disc-stylus characterics. You also then come to understand just how much mind distracting spin has been repeated about plus/minus fractional 'dB' variations with respect to the ideal RIAA characteristic, because cartridge loading has a much more significant impact upon tonal balance and reproduction dynamics than does the achievement of perfect RIAA equalisation!!!

(2) Another factor greatly affecting reproduction relates to NFB loop controlled gain stage interactions and terminations.
For example, it is possible to build a moving magnet stage using just one or two gain stages per channel, as indeed is often commercially available, and these can measure near ideal for RIAA accuracy and distortion under steady sinewave examination, but measurements alone cannot guarantee that these vinyl stages will actually sound good when coping with highly dynamic music waveforms.  NFB loop controlled equalisation stages should be buffered at input as well as output because the input terminal of a stage that is called upon to output current can itself not respond with amplitude linearly if fed by a reactive source impedance and when it does not retain a linear gain relationship with input at all frequencies, this being especially so with simple bipolar input circuitry.
Interstage reactions can also arise non-linearly with frequency, and this can be why some power amplifier plus pre-amplifier combinations reproduce less cleanly than expected when tested in other equipment line-ups.)
For similar reasons a separate additional NFB loop controlled line output driving stage is used to ensure that externally fed load circuitry cannot, by its own presence, degrade the final signal output.

(3) Here the components used for upper and lower RIAA equalisation characteristics are separated.
Passive (non distorting) 750 ohm (or 2x 1k5 in //) plus 100nF close tolerance components not only perform the RIAA HF cut between stages three and four, but they also reduce higher audio frequency noise and distortion from the earlier input and equalisation stages.
The line driver stage operates with good input signal level plus falling input impedance as frequency increases, and the resulting improvement of 'quiet clarity' in overall sound reproduction becomes instantly recognisable.
Additionally, the uncompensated series feedback unity gain error as seen on some other vinyl pre-amplifier circuits (which sometimes gives a false clarity due to increased output above say 15kHz) is automatically flattened.

(4) The original RIAA characteristic used with vinyl was (*is*) sub-bass weak, having a low frequency roll-off that introduces notable phase distortion at low bass frequencies.

In its day this roll-off provided for a reduction of LF rumble, though with a good turntable and signal recording instead of live room playback there is much to be gained by reducing the corner frequency,
For this reason I extended the low frequency equalisation to 25Hz instead of 50Hz, though I did provide a switch to option 'standard' reproduction. So do not then try to use the extended bass response for loud real-time playback unless you have solid floors and walls, a high ceiling and your turntable is solidly mounted.
The many 22uF capacitors deliberately used in this circuit form a multi-pole passive roll-off coming in below 20Hz to more sharply cut turntable and pressing rumbles below the extended response.

(5) At high live playback levels an extended bass response can set up cartridge feedback via stereo loudspeakers differentially energising a long wall room resonance. This was however easily remedied by connecting the primary of a sub-miniature transistor radio output transformer between channels, thereby *mono-ing* the sub-bass frequencies without affecting other stereo reproduction. Thus this pre-amp offered a new method for bass feedback reduction whilst having minimal impact upon the overall bass reproduction level, yet which allows higher 'pop-party' sound levels in under-damped rooms.

Don't just look at this circuit and think 'Yeah ?' or 'Sure ?' and say to yourself 'Look at all those capacitors !'.
This analogue pre-amp is an already tested design providing not just both a cleaner and quieter background to the music we are meant to hear, but also a clarity of reproduction that few are likely to have heard before, or even imagined could become a realisable experience.

Today there are many internally compensated audio integrated circuits to choose from, several offering Fet input devices. Power this pre-amplifier with a remote (on the floor) construction having at least four 470uF to 1mF capacitors per 15V rail, and place it close to the tone arm using no more than two feet of screened interconnect.  This is to minimise cable capacitance and provide mid point adjustment for the tuning capacitor.

If you do build one of these pre-amps then I do hope you enjoy the reproduction as much as I did, though unfortunately I had to sell mine (£100 in 1977) when junior came along.  So do please let me know how you get on, and let me know which modern ICs worked for you so that they can be mentioned here for other constructors.

I also remember trying basic 741s and noted that even these worked better than expected due to the way that noise and distortion from the first three stages was passively filtered, though unfortunately I do not have any photographs of my original, and very neat construction.

Good Luck .......... Graham.

PS. After the 1980's I became even more disheartened by the next supposedly superior technical improvement - the CD. Those who do construct this pre-amp might well come to understand the real significance of our loss !!!

 

Return to Home Page - www.GMweb2.net