I had someone Email and ask me about testing VTA amps for distortion and square waves. First of all square waves do not correlate directly with good sound from an amp. Case in point > Solid state amps usually give excellent square waves when tested, yet many solid state amps don't necessarily sound that good. What square waves do relate to is an amp's bandwidth and SS amps usually have a wider bandwidth than tube amps.
A few years ago I had a friend with the proper gear test my own personal VTA ST-120 (with both cap upgrades). As part of the test he ran a function generator into one channel of the ST-120 and used a 100 watt non-inductive resistor for a load. He set the input to give about 20 watts out (about 1/3 power). I had my camera and took photos of the square wave output on the screen of a Sencore SC-6 oscilloscope at various frequencies. Below are the photos for the 7 frequencies that he chose > 20 Hz, 100 Hz, 500 Hz, 1 KHz, 5 KHz, 10 KHz and 20 KHz. You can see that the input on the oscilloscope is not perfectly at those frequencies because as the function generator warmed up, the frequencies drifted very slightly above and below the target frequency. I was told that in order to produce a near perfect square wave, the amp must have a fairly flat bandwidth of 1/10th to 10 X the target frequency being tested. IOW to produce a near perfect 20 Hz square wave the bandwidth of the amp must extent from about 2 Hz (!) to 200 Hz. Well the VTA ST-120 doesn't produce a 2 Hz frequency at any level high enough to really contribute to the bandwidth. Same thing for the 20 KHz square wave photo. It's kind of lumpy because the bandwidth of the amp does not extend out to 200,000 Hz. There is still good output on the VTA ST-120 to about 40,000 Hz but only a bat could really comment on that. See photos below ...
Bob
VTA ST-120 square waves
A few years ago I had a friend with the proper gear test my own personal VTA ST-120 (with both cap upgrades). As part of the test he ran a function generator into one channel of the ST-120 and used a 100 watt non-inductive resistor for a load. He set the input to give about 20 watts out (about 1/3 power). I had my camera and took photos of the square wave output on the screen of a Sencore SC-6 oscilloscope at various frequencies. Below are the photos for the 7 frequencies that he chose > 20 Hz, 100 Hz, 500 Hz, 1 KHz, 5 KHz, 10 KHz and 20 KHz. You can see that the input on the oscilloscope is not perfectly at those frequencies because as the function generator warmed up, the frequencies drifted very slightly above and below the target frequency. I was told that in order to produce a near perfect square wave, the amp must have a fairly flat bandwidth of 1/10th to 10 X the target frequency being tested. IOW to produce a near perfect 20 Hz square wave the bandwidth of the amp must extent from about 2 Hz (!) to 200 Hz. Well the VTA ST-120 doesn't produce a 2 Hz frequency at any level high enough to really contribute to the bandwidth. Same thing for the 20 KHz square wave photo. It's kind of lumpy because the bandwidth of the amp does not extend out to 200,000 Hz. There is still good output on the VTA ST-120 to about 40,000 Hz but only a bat could really comment on that. See photos below ...
Bob
VTA ST-120 square waves