more on heater bias

[Andy Le Blanc]

Bill Ellis sent me a PM asking to explain a little more about using the heaters
of a couple preamp tubes to obtain cathode bias for a tube amp, here goes.

Tube electronics is an old technology, nearly all DIY'ers have to try to figure it
out for themselves. Every now and again you find something that's not very
common in the technical resources you scrounge up while self educating.
Pre internet, one of the few available sources for old schematics was
Aspens Pittmans tube amp book. We seem to start with old fenders, if its not
fender based, too many shy away from trying other notions of amp design.
There's a couple ward's airline, danelectro, and silvertone schematics, I think,
that certainly don't look a fender. You look at the bias, once you've got that figured,
and it don't look right... I finally built a couple, being tired of fender, the power side
of the amp in the schematic were biased with heaters of the preamp tubes.

The amp I have set up and used for data has a 022798 equivalent PT,
and a 018343 equivalent OPT and is set up with 6CA7 power tubes,
biased through the heaters of two 12ax7 dressed series to ground.


Plate voltage for the pair of 6CA7 is 402v ref. to ground and 375v ref to the cathode.
Cathode measured 26v and tot. current over a 1 ohm resistor on the
ground side of the heater string is 157ma.

So... 375 x .157 = 58.875..... \ 2... give a static dissipation of around 29w for each tube.
The pre amp heaters glow and run DC with first stage heaters dressed to ground.
It works and is gig ready, it even sounds good, with less parts to boot.
The heaters in the pre are simplified to just a couple wires, no big resistor to mount in the chassis or
heat worries around other components.

The 6CA7 have a 25w plate rating and 8w screen grid rating making the 29w figure possible
with some room to spare, and the cathode is rated at 150ma., all design center values.

I've seen example with 2 6l6 and 4 6v6, and with the addition of a resistor 4 6l6.

Now the draw backs are you have a longer warm up before using, but its a soft start for the power tubes,
and you can only design around pre amp tube types a .15 A current rating

[Alexo]

Thanks for posting this Andy, I've been considering doing this in a build where, by some warp in the space time continuum, heater noise is coming in through the grid of the pi.

Using Ohm's law, it looks like each triode in a 12AX7 has a resistance of 42 ohms, (21 ohms parallel, 84 in series per tube). For your example, this would indicate a 168 ohm RK for your EL34's, which is pretty close to what you would choose with a traditional resistor on that application.

These values should give a lot of flexibility in designing a cathode "resistor," and it seems like if you worked out the series/parallel math correctly, there's no reason you couldn't run 3 or 4 preamp tubes as RK.

...but what about dissipation and wattage? My brain is having a hard time with this at the moment... stock AX7's dissipate just under 2 watts per tube, seems a little low for this application, but if they are working as a semi-constant current source, I suppose you have nothing to worry about in this regard?

EDIT: Wait - are you only using a single 12AX7? How do you have 26 volt heaters? If you are using 2 12AX7's, how do you not have 300 ma total at 12V each?

[Jana]

The 2 ax7's are in series. 150 ma. is flowing through each one and since they are in series there is a total of 150ma. flowing.

[Andy Le Blanc]

yep... its two 12ax7 with their heater run in series, run in series...
Its the inverse of what you think, like running two caps of the same value in series .

Its current thru the heaters and not the valve, the plate dissipation has nothing to do with the heaters.

the hot and cold filament resistances are different, you have to have enough
current to light up the filament, and then a portion of the current is dissipated as heat.

whats the trick for first start ups of a new amp, put a light bulb in series with the PT
to regulate the current in case you screwed up, to save the PT from smoking.

The thing that gets me is the happy coincidence of it...
2 6l6 on a 200ma. PT, biased on the fil. of 2 12ax7, and it works very well.

[Tone Lover]

So I need to study some of those schematics so I can ask a few more questions. Thanks Andy

[pula58]

yep... its two 12ax7 with their heater run in series, run in series...
Its the inverse of what you think, like running two caps of the same value in series .

Its current thru the heaters and not the valve, the plate dissipation has nothing to do with the heaters.

the hot and cold filament resistances are different, you have to have enough
current to light up the filament, and then a portion of the current is dissipated as heat.

whats the trick for first start ups of a new amp, put a light bulb in series with the PT
to regulate the current in case you screwed up, to save the PT from smoking.

The thing that gets me is the happy coincidence of it...
2 6l6 on a 200ma. PT, biased on the fil. of 2 12ax7, and it works very well.

Then, you have 25.2V across the two 12AX7's heaters for 150mA ? I.e., 4 42 ohm heater resistances (for a total of 168 ohms) in series?

[Andy Le Blanc]

its series 2/.15 or around .075, in practice you get around 100ish.
it becomes more about you PS, rig up a cathode bias with a couple bench tubes and clips
It two tubes with a 9a basing diagram, a couple 12ax7, pins 4 and 5 series to ground

[pula58]

its series 2/.15 or around .075, in practice you get around 100ish.
it becomes more about you PS, rig up a cathode bias with a couple bench tubes and clips
It two tubes with a 9a basing diagram, a couple 12ax7, pins 4 and 5 series to ground

like this? (see attached drawing of section of Silvertone 1474 amplifier)

[Structo]

How have you been Andy?

Haven't seen you around these parts for months!

[John_P_WI]

I concur, good to see you again Andy.

[surfsup]

So if something goes wrong, can the current increase and fry all your tubes in the amp?

[Andy Le Blanc]

Yes thats It, and no it wont cook anything its just about the safest you can run a power side , you get very nearly regulated DC for a couple pre tubes.
The "But" is that you cant use a standby, because the unit goes thru a slow startup just like an old TV, a controlled warm up period,
which is great for the amp but not to a modern consumers need for gratification.

I've been doing OK, just working too hard for too little. Plenty of work, OMG...
200 gigs contracted this years, should be more next, its great, but with with the cost of every thing going up by multiple factors it don't mean nothing.
but doing alright...

[Gaz]

Maybe this is a stupid question, but you would lose bias if you pulled a preamp tube, right?

[Andy Le Blanc]

you'll turn off the power side no ground return, haven't tried it to see what happens.
good question, just like disconnecting a cathode resistor.

[gingertube]

The trick is to get the output tubes idle current to match the heater required current - as stated above that means 150mA. For 2 output tubes in push pull that means 75mA each, that is a fairly hot bias for most guitar amp circuits.
4 off 6V6 at 37.5mA each would be OK.
I would be a bit worried about 4 6L6 or EL34 or KT88 as peak cathode currents are so high.This was used a lot in HiFi Amps of the early 70's. Usually it was the phono preamp tube heaters which were used.

Articles and threads on it often have the label "Free Heater Supply".

You would have worked this out for yourself BUT it does not work using the output tubes own heaters althouigh I actually saw a guy try it and not understand why it did'nt work. It of-course does'nt work since no heater current means no tube current means no heater current etc and around and around you go.

Cheers,
Ian