Current required for tube/valve amp bias circuit...

[g1]

If you are talking about changes due to 50 or 60hz mains, it is a moot point unless you have a PT that allows you to change over between 120 to 240V.
Even then, your little bias transformer will also be susceptible to variance depending on mains frequency.
The marshall amps using this circuit have no such issues. Anyway, the numbers have already been calculated, the cap value will not require any adjustment.
But I think you have decided a while ago anyway, so I won't "doth protest too much" anymore.

[yankeerob]

If you are talking about changes due to 50 or 60hz mains, it is a moot point unless you have a PT that allows you to change over between 120 to 240V.
Even then, your little bias transformer will also be susceptible to variance depending on mains frequency.
The marshall amps using this circuit have no such issues. Anyway, the numbers have already been calculated, the cap value will not require any adjustment.
But I think you have decided a while ago anyway, so I won't "doth protest too much" anymore.

Yes g-one... as it goes... the primary on said trannies is 0-100-110-120 x 2... and the extra wrinkle is that we significantly reduced copper (and thus cost/weight) by going with 0-12.6 heaters... which works incredibly well... provided you ensure there is double the copper on the series link between the output tubes... otherwise one filament draws more current than the other ... simply double up between the pin 7 and pin 2 and the job's a good 'un ... doesn't seem to affect the novals...

If there is always going to be an issue with having to re-bias according to your location then there's no advantage at all in having a secondary tranny... I suppose it's just that 'what if somebody does something stupid' scenario in my head...

I'm going to run with your advice and see how it goes... and will post the results... many sincere thanks for your continued input!!

[Gregg]

...because with a single HV winding all you can use an SS bridge.

An interesting option is also a "hybrid bridge" provided you have 5V secondaries for the recto tubes.

[yankeerob]

An interesting option is also a "hybrid bridge" provided you have 5V secondaries for the recto tubes.

always one for learning something Gregg... these don't have a 5V rectifier winding but you've intrigued me now

[Gregg]

Check Merlin's site for more info:

The Valve Wizard

[J M Fahey]

The caps do indeed provide isolation, but of a kind few, if any, notice.

Fact is, in a dual diode "full wave rectifier" , the windings always have some reference to ground, which is the center tap, and are out of phase, so when one swings to the positive peak (thus feeding the amp) the other swings to negative peak .

So it can provide negative voltages and you just need a diode to rectify and a series resistor (often 100k to 220k) to tame the extra voltage.

But a bridge full wave rectifier has a single winding, which is not connected to ground, and which is always positive respect to ground.

So no way to pull a negative voltage out of it.

Until you cheat, that is.

One end of the winding, will swing between + 0.7V (one diode drop) and plus peak voltage (say, +500V) , as I said always positive.

So here you have a waveform which shows a voltage varying in time, 100/120 times a second
(hey !!! that sounds like AC ) but which is always positive (hey !!! that sounds like DC )

You will have this waveform there:



The "contradiction" can be explained by saying that in fact that is: "an AC waveform with a DC component".

Now, if you connect a capacitor to one end of that winding, one end will swing from +0.7V (you can call it 0V) to +500V ; DC voltage will not pass through it but the AC component will, so now the other end will swing from -250V to +250V so now we have a negative voltage available.

Simply brilliant.

So now tou add a diode, a cap or two and a couple resistors and you have "free" bias voltage.

The cap value can be calculated, of course, but is somewhat complex because that waveform is not a sinewave, but a very kinky one, with peaky harmonics; it also does not swing symmetrically from +250V to -250V as I simplified before, simply because that waveform is very asymmetrical, so I guess Ampeg,Marshall and others simply built the circuit and tested a few caps until they found a good value.

Besides it does double duty: it is smaller than optimum, on purpose, so it also helps drop voltage.

So the idea is even more brilliant.

I am not impressed by complex, kludgy, NASA type solutions but by simple and brilliant ideas such as this one.

[yankeerob]

Mr Fahey - that is a lucid, intelligent well thought out explanation... Thank you!!

[Chuck H]

That's our Juan... Lucid.