Toroidal PT buzzz

[GainFreak]

Hi,

I have a toroidal PT for a 40-50W tube amp with 330V, 45V and 6V3 secondaries. At some point I wanted to add a fan so I connected a voltage doubler to the 6V3 winding (according to the schematic below) to get 12V regulated for the fan. When nothing is connected to the 12V node the PT is dead quiet. As soon as I connect the fan which draws less than 100mA the PT starts buzzing quietly however this buzz can not be heard from the speakers. It doesn't heat up. The 6V3 is wound last on top of all secondaries.
I know toroidal PTs don't like half way rectification and stuff but besides that what else could be the reason for that buzz and can the transformer be operated safely this way?

[Dave H]

It's difficult for me to picture the doubler when it's drawn like that but I think it is full wave. I don't know if this is the cause of the hum from the toroid but you have two ground connections. Try disconnecting the ground connection at the junction of the two 100 ohm resistors across the winding. The other ground connection will put the end of the winding connected to the junction of the two capacitors at 7.5V which will put 75mA DC through the upper 100 ohm if the other end of it is grounded. Also the lower 100 ohm will have a 150mA current pulse as the voltage swings positive to make D9 charge C442.

[GainFreak]

Try disconnecting the ground connection at the junction of the two 100 ohm resistors across the winding.

I did that. I noticed that if you stick to the schematic above they are useless and get hot but that didn't affect the buzz I'm talking about.

[J M Fahey]

Chimed in to say about the same: that circuit is WRONG.

Just pull the useless 100 ohms resistors, the filament and low voltage DC ground is anode D8/negative C443 .

No clue whatsoever.

Is that a Mesa schematic?

[GainFreak]

Resistors are out now. I'll do some more experiments.
Yes, it's from Mesa Mark IV-A power supply.

[GainFreak]

As strange as it may sound it turned out to be a mechanical problem.
Previously I mounted the fan (40x40mx20mm) using those special rubber grommets to prevent vibrations entering the sound through chassis and tubes. However it turned out that with those special anti-vibration rubber grommets the fan actually was vibrating more than when fixed with regular bolts thus transferring the "buzz" to the PT through the chassis (it's around 10cm/4" from the PT). The hint was when I removed the fan from the chassis and the "buzz" stopped. Who would have guessed...
Thanks to everybody who tried to help. Below is a corrected schematic of that power supply which actually is very simple and handy when you need rectified and regulated supply for couple of tubes or other stuff.