Negative bias voltage from tube rectifier?

[passaloutre]

Perusing some ancient circuits, I came across this anomaly in an old RCA paper. It derives -30vdc for a grid-biased output section by placing a 200ohm resistor between the HT center tap and ground (bypassed by a 100uf cap).

Anyone care to explain how that works? It looks like the cap on this bias supply is still oriented cathode to ground, but that doesn’t make much sense to me

[passaloutre]

Here’s the next page with component values and tube voltages

[AlNewman]

Quite a few older circuits work like this, and I think instead of creating negative voltage, they elevate the ground.

[passaloutre]

But the cathodes are tied to ground, which should be at a lower potential than the grids? I just can’t wrap my head around it.

I just so happen to have a lot of adjustable 50W 200ohm Dividohm resistors so this is really appealing to me If it’s feasible.

[sluckey]

It's really old technology. It's called "back biasing". Read Aiken's "What is Back-Biasing?" article for more information. Then file the info away in your archives and forget about it.

PS... Here's an old '50 Hammond organ that used back biasing. R101 (lower right corner of schematic) is the back-biasing resistor.

[AlNewman]

They were kinda smart in the old days.

[Merlin]

Back biasing can be easier to grasp if you re-draw it

[sluckey]

It looks like the cap on this bias supply is still oriented cathode to ground

Yes, that's a schematic drawing error. The negative lead of C10 should be connected to the transformer center tap.

But the cathodes are tied to ground, which should be at a lower potential than the grids?

Don't think of it like that. For a tube to operate properly, the grid must be negative with respect to the cathode. Said another way, the cathode must be positive in respect to the grid.

There's really no reason to experiment with back-biasing unless it helps you to understand the circuit operation. Back biasing was superseded by silicon diodes in the '60s. But there's no practical use for back-biasing today. A 10 cent silicon diode works much better than a $5 power resistor and there is no wasted power just heating up the chassis.

FYI, long time ago during the time of field coil speakers, some manufacturers would connect the field coil to the PT center tap rather than use a high power resistor as seen in this older Hammond AO-29 circuit.

[passaloutre]

I’m not so quick to dismiss it. It seems like some elegant old-school engineering. It reduces parts count, provides some of the self-adjusting benefits of cathode biasing, and knocks a couple dozen volts off the B+, which is almost always welcome in my shop.

I’m also in the unique position to have a LOT of 200ohm power resistors, at least enough to conceivably put 2 in every amp I ever build, unless I go commercial.

If a little extra heat is a downside, then I shouldn’t be making tube amps anyways.

Thanks for your help and expertise, as always.

[AlNewman]

It is pretty much cathode biased.  I'm trying to figure how it would be considered fixed bias.  There must be a reason they approached it that way, but I'm at a loss.

[sluckey]

It is pretty much cathode biased.  I'm trying to figure how it would be considered fixed bias.  There must be a reason they approached it that way, but I'm at a loss.

Huh? All the circuits presented above have the cathodes tied to ground and a negative bias voltage is fed to the grid. That screams fixed bias to me.

Back-biasing was born in 1933. There were no silicon diodes back then. If you needed a negative voltage back then, putting a resistor on the PT CT was the smartest way to get it.

[AlNewman]

Except when current increases, wouldn't the negative voltage at the grid increase?

[sluckey]

Except when current increases, wouldn't the negative voltage at the grid increase?

Exactly. And that's one major reason you don't see this 90 year old technology being used today. But that doesn't make this a cathode biased (AKA self biased) circuit. It's just a poor fixed bias circuit. You could use zeners to overcome this, but why keep trying to breathe new life into an ancient technology.

[kagliostro]

If I see a reason it will be to drop a bit the B+ voltage if the PT has a bit high HT voltage

It may vary the bias at the change of current but in a Cathode bias it acts only on the power tube, in the back biasing it acts to the whole amp and this gives different response to the amp

If you want to experiment with not usual configurations, consider to give a try to the Mixed Bias

Franco