6,3V, center tap, 100 ohm to ground, or nothing?

[Tdale]

I started changing pots on my first DC30 clone, that I built a couple of years ago. The pots were of poor quality, and I have som Clarostats lying around, and I'm putting them in.

Anyway, when I was soldering, I discovered something strange.

There are two secondaries for 6,3V. I used one for the preamp tubes and one for the output tubes, to spread the current draw. None of the secondaries have a center tap.

The preamp heater wires are grounded on one side, in stead of using two resistors to ground. Not sure why I did it this way, but it works.

However, the heating wires for the output tubes are just wired to the filaments, and that's it.. No center tap, no resistors to form an artificial ground and none of the sides are grouned.

I don't know why I did it this way..I suppose I forgot.. but it still works...

Why should there be a center tap or artificial center tap for the heaters...?

Tommy

[VacuumVoodoo]

Heater voltage should be referenced (conected) to some dc potential, either ground or a potential positive respective to cathode.

This is recommended for 2 reasons:

1. If the heater suply is left unreferenced it will assume any arbitrary potential rel ground and cathodes which may or may not be higher than max allowed heatre-cathode voltage for the tube used. This is always specified in the datasheets and varies betwenn 50-200V.

2. Referencing heater supply symmetrically (with resistors or a pot with wiper grounded) around chosen dc potential tends to reduce hum.

[Tdale]

I see, so it'll work without any reference to ground, but it's not recommened.

Any feelings about just grounding one side of the heater circuit? Should I use two resistors in stead?

Tommy

[skyboltone]

Some schemes show the cathodes of the output tubes hard grounded, while cathode biased pre-amp and output tubes operate with their voltages somewhat above ground. Could this be why it is possible that the filaments in output tubes in grounded cathode schemes can operate without reference to ground at all? Logically it would seem that if the filament windings are separately wound there is no reason for them to assume any voltage differential compared to the plate. On the other if the the windings are part of an autotransformer set of winding big trouble could result without a reference.

These are questions. Emphatically not answers!

Dan