• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

New? Plate-to-plate feedback from UL tap output transformer

I found a different way to apply the plate-to-plate feedback, which I never seen in other schemes and sounds great!

Instead of taking the signal from the plate, we can take it from the unused UL tap of the output transformer.

(I don't use anymore UL config, as I prefer triode/penthode sound).

If you have documentation please let me know! Strange not to have seen it elsewhere..

I'm using plate-to-plate negative feedback because it helps stability at HF allowing to use a much higher GNF (about 20db).


  • plate-to-plate.jpg
    20.9 KB · Views: 320
The word I couldn't remember was E-Linear. There are SE and P-P versions scattered throughout the web and this forum. Some run the output tube as a pentode, some use UL. Pete Millett has a UL version on his web site.

Terrific! .. and congratulations for your vast culture in the field. It seemed strange to me (although I might like it) to be the pioneer of such configuration.

I arrived late. Now I look for all the documentation on the internet.

Thanks so much
Usually the UL taps are more closely coupled (via interleaving) to the secondary than the plate taps, in order for standard UL to work well. So this has some advantage for linearity and output impedance when used for "local" N Fdbks. Beside providing some attenuation needed for the N Fdbk besides. However, the plate signal has to traverse part of the OT to get there, so this can cause some phase issue for the N Fdbk loop.
You just have to experiment to see how much N Fdbk you can safely put thru this route.

E-linear was popular for a while some years ago, but hasn't been too prominent lately.

Jan Veiset, I believe, had a rather unusual take on this with N Fdbks from the UL taps (or was it CFB taps?) back to the driver stage screen grids. With a special loop gain arrangement such that the driver screen grids and driver plates swung with a near constant V ratio between then. Causing the screen grids to draw a constant fraction of the cathode current, and so becoming effectively constant resistances to drive. Just about everything imaginable has been tried in the past. But simplicity always seems to predominate in the end by default.
Last edited:
The E-Linear concept is generally credited, as I undertand it, to Douglas Picard (Pak Protector) and Pete Millett who each came up with the idea independently but around the same time, Doug for PP and Pete for SE.

The only commercial application I know of is the Emotive Audio Vita push-pull.
Interesting. Can you share some details about the amp design in which you tried the P-to-P feedback through UL taps?

I built some amps with KT-77 outputs and 6AC7 pentode drivers and they were lively and dynamic. I also have a pair of Emotive Audio Vitas using KT-120 outputs driven by 5687 diff amp, and they too are very lively and dynamic. One factor that PakProtector mentions is the ratio for the UL taps. Some output transformers have UL ratios that work better for E-Linear than others. Unfortunately I don’t remember what ratio he concluded works best.
I tried feedback from the UL Taps on my Baby Huey design and ended up going back to taking the feedback from the anodes. Purely subjective decision from listening.

With feedback from the UL Taps:
I thought the response was too speaker impedance vs frequency curve dependent, that is - not enough of the reflected impedance was inside the feedback loop(s).