|SHUNT FEEDBACK PHONO PREAMPLIFIER|
Implementation of this phono preamplifier is based on two main principles - the used circuitry is discrete, transparent, operating in pure class-A and the applied feedback is exclusively of shunt type. These requirements are vital for the live sound reproduction, this is confirmed first of all subjectively, the shunt feedback contributing maximum effect to this reproduction because it suggests very accurate, in one point comparing of the feedback signal and the direct MM-cartridge signal which doesn't loose therefore its most important subtle "live" details.
The shunt feedback allows to achieve easily an accurate RIAA amplitude- frequency characteristic necessary for vinyl reproduction. This characteristic is formed by a network C3, R6, C4, R7 just in the first stage of the preamplifier whose full circuit diagram is depicted in Fig.1.
Fig.1. Shunt feedback phono preamplifier.
It's very important to choose an optimal gain distribution between the preamplifier two stages to exclude overload issues in each of them and to make the negative feedback maximum effective in reducing distortion to the levels of below 0,001%.
The used inverting four-transistor amplifying block provides a considerable voltage amplification of about 4000 within the whole audio range (the input cascode on Q1, Q2), the circuit output is represented by a Q3 emitter follower fed from a Q4 current generator. The configuration has an excellent linearity, it is free from common-mode problems and switch-off output distortion, it features an absolute DC and HF stability that permits very great amounts of negative feedback to be applied.
The preamplifier second stage is built similarly, it has a constant 29dB closed-loop gain to bring the circuit output signal to a typical 500mV. There is also a possibility to configure this stage as a rumble filter effectively suppressing vibrations and infra-low frequency garbage caused during the vinyl disc playing and being harmful for loudspeakers. The biasing elements D1, D2, R15, R16, C10 are common for both stages, but they can successfully supply two reference voltages for a greater number of such amplifying blocks, for example, for six.
The phono preamplifier objective characteristics are obtained by computer simulation of its circuit with the help of the Multisim 10 software. AC analysis is performed by monitoring the preamplifier output while its input is fed by a 10mV test signal sweeping in a 2Hz-100kHz frequency range. The output voltage is expressed in dB relative to a 1V signal (see Fig.2), so at 1kHz we have the gain exactly 1V(0dB)/10mV=100.
Fig.2. AC analysis of the phono preamplifier.
Most commonplace notion concerning the shunt feedback application is about its unsatisfactory noise performance in phono preamplifiers. Indeed, inserting a 47kOhm resistor in series with the cartridge produces a frightening -63dB noise, but this noise isn’t the same as -63dB produced by a microphone and a linear preamplifier. The obtained noise spectral density graph (Fig.3) reveals considerable attenuation of the most annoying high-frequency noise components due to the action of RIAA reproduction characteristic. Subjectively this frequency weighted thermal noise is always swamped by the vinyl disc surface noise and becomes notable only when the record playing is stopped.
A cardinal solution of all noise problems at once gives my blameless noise reducer which is included in all my audio devices as one more filtering stage and here may be placed just after the phono preamplifier.
Fig.3. Noise analysis of the phono preamplifier.
Given that the signal generated by a typical moving-magnet (MM) cartridge is usually about 5mV at 1kHz, it reaches 500mV at the preamplifier output, but next I would like to test the circuit linearity in my standard conditions, with a 1V output signal delivered to the connected 2kOhm load. Three main frequencies 20Hz, 1kHz and 20kHz are chosen for measuring distortion, the work is done in the interactive mode with the help of the Multisim distortion analyzer.
It is a simulation in time domain when the instruments and probes placed at some points of the circuit give detailed continuous measurement information (voltage, current, frequency, distortion) until stopping the simulation. As can be seen from the screenshots of Fig.4-6, the analyzer registers the preamplifier distortion of less than 0,001% within the whole audio range.
Fig.4. Distortion of the 1V output at 20Hz.
Fig.5. Distortion of the 1V output at 1kHz.
Fig.6. Distortion of the 1V output at 20kHz.
I considered so far the preamplifier as a stand alone device, not taking into account the MM-cartridge it will work with. The electrical signal is generated within the cartridge coil which has a considerable inductance, in the ShureV15 TypeIV it reaches for example 500mH that creates a reactance XL=2πfL being added in series connection with R1=47kOhm (Fig.1). The first stage closed-loop gain becomes therefore at 20kHz:
The gain reduction caused by the cartridge inductance L at 20kHz is calculated by comparing the above two expressions for K1 : 47/78,4 = 0,6 = -4,44dB. The corresponding curve is shown in Fig.7 in green color (initial RIAA curve - in red).
To partly neutralize this negative cartridge influence, at least in audio frequency range, a capacitor in parallel with the cartridge should be added, its optimal value in our case is 150pF and the corrected amplitude-frequency characteristic is depicted in Fig.7 in grey color. Remember that this is total 150pF including also the capacitance of tone arm wiring and the leads running from the turntable to the preamplifier.
Fig.7. AC analysis of the system Phono preamplifier - MM-cartridge.
The offered phono preamplifier was first built on breadboard twenty years ago when I continued experimenting with my audio inverting amplifier. It was a scrupulous work, I tested and adjusted the device in my laboratory by using real instruments whose measurement data are absolutely identical to those obtained in the course of the present simulation. At that time I had already an excellent phono preamplifier created in the best traditions of high-end: special matched input JFET transistors, huge overload capability (±30V power supply), extremely low noise (-87dB). Its sounding was too familiar for me to confidently compare it with what I heard from my shunt feedback preamplifier. This new unusual preamplifier demonstrated quite phenomenal vinyl reproduction, each listening session turned into incessant discovery of new "live" details my old LPs were rich with.
The second revelation was that these details are easily transferred to the computer-made files after properly carried out digital conversion of analogue vinyl music. To confirm that I offer to download and to compare two music recordings of the same song: the first, "live" rip was created via this phono preamplifier, the second, typical rip – via an ordinary audio equipment. Alternate playing of these two audio files from the same chosen point in Sound Forge 9 allows to clearly hear the difference in sounding - not in level or tone balance, but just in the "live" content whose maximum this preamplifier easily retrieves. Of course, all should be reproduced via decent audio equipment and acoustics to enjoy the sound in full degree.
Now I give everybody the possibility to build this phono preamplifier and enjoy its performance, I also hope for full understanding of both the whole concept and the circuit details as well.
To complete this DIY project, I offer the preamplifier printed circuit board (77mmx82mm) whose picture is shown in Fig.8. The picture clicking produces more detailed animated images of various color performance. The recommended for this board polyester film capacitors – WIMA MKS2 (pin spacing 5mm, voltage 63V), metal film resistors – 0,25W. The board's Gerber files can be downloaded from here.
pdf version here