E-Linear, P-P pentode application

[stephenl5]

I was reading a few old threads here regarding E-Linear.

I'm interested in trying this in a P-P Pentode amp. I have a 50W, 6K6 pri. OT with 40% taps. I prefer the sound of pentode over UL... So can I just feed the plates of a 12AX7/AT7 LTPI from the UL taps.

Does each UL tap get connected to the corresponding LTPI triode or crossed over to the opposite triode.

The feedback from the UL taps seems like it would be too much, but I assume it gets attenuated by thr plate load resistors?

 

[Paul Scearce]

You could modify an E-Linear design by wiring the output tubes as pentodes, and connecting the driver tube loads to the corresponding UL taps. This will essentially be the same as using 40% Schade type feedback.

 

[stephenl5]

Thanks for the reply Paul.

I'm interested in how to calculate the amount of feedback in the loop.

It seems like the feedback from the 40% taps would be quite large compared to what is typically taken from the OT secondary. Is it hard to make it stable or do the plate load resistors attenuate it sufficiently? I think I read that hi Ra tubes are preferred...

 

[nerdorama]

Here is an e-linear amp with the output tubes running in pentode mode with a regulated screen voltage. I modified a Dynaco ST-70 to e-linear using both pentodes (6AU6) and high mu triode (6GK5) driver tubes. I liked that audio quality but had a hard time with hiss in the amp. I've never figured it out but the pentodes were worse than the triodes and they were quite microphonic.

6CB5A/EL84 amp | Audiokarma Home Audio Stereo Discussion Forums
Good luck with your project.
John

 

[Paul Scearce]

The feedback, distortion reduction, and impedance reduction are all roughly .4 times the gain of the output tubes. The load impedance seen by the driver tube is also reduced by the same amount. Stability shouldn't be an issue, since the feedback loop only includes one tube.

 

[stephenl5]

Thanks John & Paul,

So do I still use "typical" values for the plate resistors for a given triode in the LTPI... 100k for 12AX7?

 

[Paul Scearce]

Modifying a typical P-P pentode design by simply attaching the load of the drivers to the transformer taps may work, but no guarantees. The ac voltage at the plates of the drivers will be the same as before, but the ac current will be considerably more, and some care must be taken that the drivers can meet this new demand. Otherwise, the amp may not be able to reach full power, and the drivers could distort.

On the other hand, modifying an existing E-Linear design to pentode operation should be as simple as connecting the output tubes as pentodes.

 

[nerdorama]

Thanks John & Paul,

So do I still use "typical" values for the plate resistors for a given triode in the LTPI... 100k for 12AX7?

You may have to recalculate the anode load for the drivers since the UL tap might very likely be at a higher voltage. Otherwise you will probably have both higher current and a higher plate voltage on the driver. I've been through this a couple of times on an old Dynaco ST-70 that I converted to E-linear.

John

 

[stephenl5]

Thanks John,
Would you happen to have a schematic to share for the ST70 you converted?