GROUNDING BACK TO BACK TRANSFORMERS

[Niquel]

Hi,
I am trying tu build some kind of preamp based in the Matchless Hotbox with 2 Russian 6N2P-EV (military equivalent to 12AX7 but can only be fed at 6,3V).
I use two back-to-back transformers to get the B+ voltage. The first one (for 2 serial tube filaments) is 220/16VAC 32VA with a rectifier bridge and voltage regulator. The second one is 12/220 VAC 2,4VA rectified as well (total maximum anode consumption should be 12,8mA).
Following several schematics (Matchless Hotbox, Real McTube 1 and 2, Tube Reactor and Evil wishper), I´ve seen different ways to refference both circuit, so I really don't know if I should use the GND1 or GND2 labeled in the attached schemo for the cathode resistors.
Any help will be wellcome.
Thank you,
Miquel

[Osvaldo de Banfield]

Both can be wired directly together, or separated up to +-100V or the cathode-filament maximum isolation recommended by the manufacturer of the tube.

[Niquel]

Hello Osvaldo,

I have tried to link both grounds but the current flowing through the secondary of T1 and the primary of T2 is about 1,5A. It is not a problem for T1 (32VA) but forces T2 to be bigger than I desired since this transformer has to be PCB mounted. I really don't understand why is this large current since I filament is about 350mA and I R17 is about 4,5mA.
Any suggestion to make it more efficient will be welcome.

Thanks (gracias),

Miquel

[DF96]

You may be pushing the second transformer into saturation by putting 16V onto a 12V winding. Few mains transformers can cope with 33% overvoltage. Saturation means the transformer loses much of its inductance, so will take a heavy current and get hot quite quickly.

The cathode resistors should go to GND2, as should the LED resistor. You can connect GND2 to GND1 too, but ground connections and supply rail connections should generally parallel each other.

[Niquel]

Thanks DF96 for your reply,

In fact, I was suspecing of core saturation but I needed somenone to corroborate it.
It is true that the second transformer gets really hot but, after putting the filament circuit with a half wave rectifier with a common ground, the total current consumption has lowered to 1 A which is an acceptable value.
It also has decreased the 2nd transformer temperature, the voltage regulator remains almost at the same temperature (1,5 V ripple) and hum has decreased a lot.
Now I am using a scavenged wall wart 7VA trafo. I'm not sure if the 2,4VA would work similar.
What would be the benefits of having independent grounds?

[DF96]

I don't know what you mean by independent grounds. The heater supply must have a voltage reference. Easiest is ground, so just connect GND1 to GND2.

You seem to be mixing up transformer saturation with rectifier behaviour. These are separate issues.

[Niquel]

Originally Posted by DF96

You can connect GND2 to GND1 too, but ground connections and supply rail connections should generally parallel each other.

With this sentence I thought you meant independent grounds.
I know that core saturation has nothing to do with rectifiers but I saw that in the Tubeman II power supply they used the same system with a half wave recto. for filament and I have decided to try it, I don't know why, but current consumption has lowered, any idea?

[Niquel]

I put serial filaments because 6n2p only accepts 6,3V and there's no option like in the 12AX7 to feed them at 12,6V.