MJR7-Mk5. Setup and testing

One small problem with the MJR7 is that setting the output stage operating voltage to half the supply voltage is difficult if this is done with no load connected. This is because there is then only a 470R load and the output capacitor time constant affects the overall feedback in such a way that adjusting the preset control has only a very slow effect on the voltage. I was turning the control from minimum to maximum and finding the voltage at the mosfet sources changed very little, and thought something serious was wrong, but being more patient and waiting a minute or two after each adjustment revealed that it was behaving exactly as expected. Using a small resistor load, e.g. 22R, during adjustment makes setup far quicker, as does connecting a speaker, but until everything is set up and confirmed to be working correctly it is safer to use the resistor.
If using a speaker without a low dc resistance, it would be a good idea to add a parallel resistor, maybe 100R 5watt, to ensure the output voltage stabilises quicker. Maybe some electrostatic and piezo speakers have this problem, and some speakers have an internal coupling capacitor.

I forgot another important point when I connected a current meter in place of the fuse to set the quiescent current. I already had the power supply switched on, so of course I was applying 60V to the uncharged on-board electrolytic capacitor, with the inevitable sparks. It did no obvious damage, but remember not to switch on until everything is connected, including the meter, and switch off and let the capacitors discharge before changing any connections. I use a 100R 5W resistor to discharge the supply capacitors after switching off before making any changes. When first switching on the completed amplifier ensure the 220R presets are set to minimum, i.e. fully anti-clockwise, to start with a low quiescent current. The 4k7 output stage voltage adjusters can be set to half-way at first, and then adjusted to set the voltage at the mosfet sources to half the supply voltage. Using each of the 220R presets the total current can then be increased by 100mA for each channel to set the quiescent current. For example if the supply current is 40mA with both 220R set to minimum then one can be adjusted to give 140mA total current, then the other 220R can be adjusted to give 240mA total current so that each output stage is running at 100mA. It is often suggested that using a higher quiescent current will reduce distortion in mosfet amplifiers, but my findings for this particular circuit are that increasing the current a little further has almost no effect on 2nd harmonic distortion, and just changes the relative levels of higher order components, some increasing and others decreasing. 100mA seems about the best choice, and has the advantage of a close to zero temperature coefficient, so the current will not drift too far.

Provided the correct component values have been used there should be no stability problems, but if an oscilloscope is not available a simple check is to touch the 1R resistor in the output stage Zobel network to see if it gets warm. It should remain cool in normal use, but any significant output from high frequency instability should noticeably increase its temperature. With bad instability it could become very hot and burn out, so be careful with this test.

Starting with the input shorted to earth a speaker can then be connected. It is a good idea to have a cheap speaker to use for amplifier testing rather than connect your best or most expensive speakers until everything is checked and working. Try switching on and off a few times to check that there is no serious thump from the output capacitor charging and discharging. When switched on, depending on the speaker sensitivity there may be a faint hiss audible with an ear close to the speaker. Applying a music signal from a source with a volume control the output can be checked at various output levels, again checking the 1R resistor for temperature increases. Excessive heatsink temperature could also be a sign of problems. If all is well then you can proceed to any measurements or listening tests you prefer.

If comparing the sound to that of another amplifier remember that unlike most amplifiers the MJR7 is an inverting design, and using speakers with significant even-order distortion (which most speakers have) it is possible to hear a difference with an inversion because the even harmonics in the music will then be increased or reduced by the speaker depending on the output polarity. Just reversing the speaker connections at one end of the cable will cancel this difference, but it is difficult to be sure which way is correct, unless polarity inversion has been avoided throughout the whole recording and playback chain, and unless the speaker also has low phase errors. My own experiments were inconclusive, another test subject agreed with me that the difference was clearly audible, but disagreed about which sounded best.