Hey guys, sorry for the slow reply on this. I did some reading on the original thread above to understand Mr. Cordell's concern and found that in that thread the potentiometer used was 1kOhm. I orignally tested the pcb from TIPD136 with a higher impedance potentiometer and didn't see much in the way of distortion, so I ordered a 1kOhm potentiometer and swapped it into the board. I took FFTs of the output signal for a 1kHz sinusoid at 2Vrms. The FFTs are taken at attenuation levels of 0dB, 40dB and 75 dB. The y-axis in these FFTs are dB relative to 2Vrms for ease of comparison. I also measured the frequency response at a these attenuation levels.
For 0dB attenuation and a 2Vrms signal, the 3rd harmonic is dominate at -117dB, the other harmonics are down 130dB or more. At 40dB of attenuation the 3rd harmonic is still dominant and is -103dB below the fundamental. At 75dB of attenuation the 3rd harmonic is 70 dB below the fundamental.
Frequency response only changed in the case of 75dB of attenuation, where there is about a 10dB loss in attenuation at 100kHz.
Overall I feel that if this performance concerns you, a good option would be to use a higher impedance potentiometer.
Do you have a AP diy kit?