Ya, and I even wonder about the rectifiers since I've tried rotating the entire unit and even gotten below it. I would think (and could be wrong) that the rectifiers would be more of a electrostatic issue, so a huge chunk of grounded steel in the way would mop that up.
No, only fullwave on 3 windings and only 3 rectifiers. The aux winding supports relays and then a digital uC section, so I step the voltage down from 12V (relays) to 5V (uC) with a regulator to isolate. The ground takeoff for this supply is at the output cap of the 5V regulator.
Thank you! I tried most of this.
Removing c97, c121 and c126 made a huge difference to the 1-5kHz range. Improvement could have been 10dB or more in that frequency range.
10 ohms in series with the slug caps did very little, potentially an improvement in the 3-7kHz range.
The rest were destructive so I was hesitant, but I did put 10ohms in series with F1 and it may have helped a bit but nothing super obvious.
Interestingly, I tried removing C120 and C128 based on trobbins' idea, and it seemed to lower the 120 and 240 Hz range, but it also seemed like I was getting more upper harmonics and that the noise seemed to modulate more. I then tried removing C94 and C96 ("series cap 2 in the HV supply) and the noise was 10dB higher across the entire range....which makes me wonder.
Then I tried swapping the XFMR leads on the Aux supply to see if I could possibly use phase of the charge pulses to cancel at least the LV supplies, no improvement.
But I went back and compared to the professional unit and we are actually beating or tied with it on the spectrum analyzer test. So I'll take this as a victory so far.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.