[AMRadio] GG 813 Linear in GE Ham News -12AT7 12AX7

Bob Bruhns bbruhns at erols.com
Sun Oct 2 14:58:22 EDT 2005

Wow, 8W of audio from a 12AX7. I assume the
distortion is pretty high at full output.  8W, 8K
p-p, that's pushing the specs with a 12AX7,
because of cathode emission and plate dissipation.
And with zero bias, you would need to set up a
phase splitter that that can provide grid current.
I guess it's possible, but I'd try a 12AU7 or
preferably a 12BH7 at 300V with -20V grid bias in
class B1.

Cathode emission

6SN7 filament: 6.3V 0.6A, 3.78W filament power,
70mA cathode emission per tube.  90 mA would even
be pushing that a little bit.  12AX7 filament:
12.6V 0.15A for 1.89W filament power, expect lower
cathode emission. 90 mA peak is pushing that. 8W
from 8K p-p, not even considering output
transformer efficiency, needs almost 90mA peak.

6SN7 dissipation spec: 5W per plate, but 7.5 w
total when both plates are hot.  12AX7 dissipation
spec: 1.2W per plate.  Whoa.  Maybe this figure is
low, anticipating a shield around the tube that
would hold in the heat.

Plate Dissipation

8W RMS, that's 16W peak.

8K p-p, that's 2K plate to B+, that means:

16W peak, 2K, 178V peak to zero
178V peak, 2K, 89.4mA peak per plate.

178 V peak, plate at least +250V, maybe +300V, to
give that much swing.

250V plate and -2.2V grid bias would have 1mA
resting currrent per plate, for 0.25W resting
dissipation per plate.  Minimum instantaneous
plate voltage at the audio peak would be +72V.

300V plate and -2.6V grid bias would have 1mA
resting current per plate, for 0.3W resting
dissipation per plate, probably OK.  Minimum
instantaneous plate voltage at the audio peak
would be 122V, so dissipation would be
significantly higher with audio than it would be
with 250V at -2.2V bias, but the 90mA peak plate
current would be easier to achieve.

8W out, class B, at this low voltage means at
least 5W dissipation, 2.5W per plate, for a
continuous sine wave at full power.  That's
pushing it.  Maybe it's OK if the 12AX7 was rated
very conservatively, or if the duty cycle is low

Zero bias?  250V plate, zero bias, 7mA resting
current, 1.75W per plate resting dissipation per
plate.  300V plate, zero bias, 9.2mA resting
current, 2.8W per plate resting dissipation.  Can
the 1.2W per plate dissipation spec be this
conservative?  Maybe with a high grid leak
resistance it would self-bias a bit, and work.

  Bacon, WA3WDR

----- Original Message ----- 
From: "Patrick Jankowiak" <recycler at swbell.net>
To: <amradio at mailman.qth.net>
Sent: Saturday, October 01, 2005 10:15 AM
Subject: Re: [AMRadio] GG 813 Linear in GE Ham
News -12AT7 12AX7

> don't laugh too hard, In the ARRL handbook it
claims 8 watts audio from an
> 12AX7 in zero bias class B push pull into 8000
ohms. I have never
> understood that one, if you look at the currents
stated in the book, they
> seem pretty high for the tube.
> PJ
> AMRadio mailing list
> Home:
> Help: http://mailman.qth.net/mmfaq.html
> Post: mailto:AMRadio at mailman.qth.net
> AMfone Website: http://www.amfone.net
> AM List Admin: Brian Sherrod/w5ami

More information about the AMRadio mailing list

This page last updated 24 Jan 2018.