Yes, I know, it's about 100 small steps and not one big step...
Thanks again for your efforts, but the flux band is gone. In my case I can say for sure, that it has only an effect, if there is no pot covering the OPT and/or PT. Playing 'three shells and a mVrms-value' it is obvious, that there is no difference in the readings w/w.o. flux band, if there is a pot covering the transformer.
Correct again 😀. No difference using isolation washers on the OPT mounting bolts, but on the PT ...
all three pots in place, no isolation washers: L = 9.7 mVrms, R = 3.7 mVrms
OPT pots in place, no isolation washers: L = 11.2 mVrms, R = 4.7 mVrms
OPT pots in place, PT isolation washers: L = 4.9 mVrms, R = 3 mVrms
...not perfect, but a lot better.
Looks like microphonics from PT vibration. Could be transmitted into just about anything but looks to be more L channel centric.
Thank you for your suggestion, but I think this is not the case. When I hold a choke in my hand and move it around in the air (no contact to the amp) I get the same amount of L/R imbalance. Same behavior with the OPTs 'airborne'.
The only two things till now making a noticeable difference are electrical isolation of the bolts+washers and/or increasing the distance PT - OPT.
As I understand it, your testing is just energising the PT, and using each OT primary for passive signal measurement. As such, the OT's do not operate with signal voltages and so won't induce any signals in their mounting bolts. But it appears that an energised PT is generating signal currents in its mounting bolts and through the chassis, and they are a contributor to the signal levels picked up by the OT primaries.
Given that you can't modify how the mounting bolts are fitted to the transformer itself, then modifying how the mounting bolts are fitted to the chassis seems the only tweak you can make for that aspect. Apart from adding steel sheets to the inside wall faces of the PT aluminium can, and possibly the top and bottom surfaces, and making sure the can does not touch the PT or its mounting bolts, then I think that is the limit of your PT related remedial changes for PT leakage flux.
Your other remedial options are as you say quite onerous, like increasing distance between the PT and OTs, and possibly rotating each OT to see if there is a minima for each OT.
Given that you can't modify how the mounting bolts are fitted to the transformer itself, then modifying how the mounting bolts are fitted to the chassis seems the only tweak you can make for that aspect. Apart from adding steel sheets to the inside wall faces of the PT aluminium can, and possibly the top and bottom surfaces, and making sure the can does not touch the PT or its mounting bolts, then I think that is the limit of your PT related remedial changes for PT leakage flux.
Your other remedial options are as you say quite onerous, like increasing distance between the PT and OTs, and possibly rotating each OT to see if there is a minima for each OT.
Last edited:
Perfect summary of the situation. I tried 1mm steel and that didn't show any effect. Inside the can 4 mm is the maximum for the walls (almost touching the core), nothing on top and bottom. I guess I could only put 3mm sheets to the walls. The other (ugly) option would be to add 5mm sheets between the PT and OPT. The real solution (but expensive) a bigger chassis.
Moving the choke around was interesting, on the right channel the choke had to be centered relative to the PT to show the minimum (as the theory predicts), but on the left channel the minimum was skewed to the chassis - the choke had to be close to the bottom edge of the core to get the lowest reading.
Just wanted to tell you that I tried PVC 5/8mm round cable grommets for PT and OPT, which gave me L channel 1.5 mVrms and R 2 mVrms (L is now less than R !), a lot better than the heatshrink + thin washer stuff I had before (setup as in post #81).
Thank You !