Low THD of Opt Cathode feedback

[siliconray]

A new modle EL34 PP is under developping. It's a complicate design: input transformer, 12AT7 preamp, EL34 Push-pull, Optimus series OPT with cathode feedback. In order to find the benifits of OPT cathode feedback design we did a compare testing, the result is quite positive.

The amp under testing

An externally hosted image should be here but it was not working when we last tested it.

THD measurement at 1W, 1KHz, 8Ohm

An externally hosted image should be here but it was not working when we last tested it.

THD datas, THD1 is THD with OPT cathode feedback, THD2 is non OPT cathode feedback

An externally hosted image should be here but it was not working when we last tested it.

The curve comparing

An externally hosted image should be here but it was not working when we last tested it.

Close looking of the input transformer (black capacitor like cylinders)

An externally hosted image should be here but it was not working when we last tested it.

Freqresponse from 10K to 100K, 3dB BW is up to 75kHz, awesome

An externally hosted image should be here but it was not working when we last tested it.

Shcematic will be posted when the testing is finished.

 

[jonusgrumby]

What percentage cathode feedback do you have? I've used 10% with good results. The only drawback is that in a triode output stage you need a large output voltage swing, but it is well worth that trouble.

Are you running the EL34 as pentodes or ultralinear?

 

[siliconray]

EL34 is in pentode mode. There's a stand alone coil on the transformer for cathode feedback, same turns as the 4Ohm output coil, center tap connected to GND, the other to pin 1 and 8 of EL34.

You may have a close look at the transformer and wiring.

 

[ThorstenL]

Hi,

A new modle EL34 PP is under developping. It's a complicate design: input transformer, 12AT7 preamp, EL34 Push-pull, Optimus series OPT with cathode feedback. In order to find the benifits of OPT cathode feedback design we did a compare testing, the result is quite positive.

Yes, cathode feedback works well.

My current Tube Amp is EL34 (350V/60mA per tube) PP with screengrid (ultralinear) and cathode feedback and no overall negative feedback. ECC83 input/gain stage and 6SN7 "concertina" phasesplitter.

Sound is very different to common push-pull Amplifier, much more natural and relaxed.

Ciao T

 

[revintage]

A distortion reduction of far more than 12dB can not be accomplished with a 4ohm cathode feedback winding alone.

 

[Ejam]

Thorsten - Do You Have a Circuit?

Thorsten

Do you want to share some more details of your amp. Would be appreciated. Why the shift from output triodes?

 

[Miles Prower]

A distortion reduction of far more than 12dB can not be accomplished with a 4ohm cathode feedback winding alone.

It doesn't look like he's making that claim, and the THD figures he gives looks like there's gNFB involved as well. gNFB with cathode NFB could get that 12db reduction in distortion.

 

[siliconray]

Yes there's a 47k Ohm for gNFB. Too much gNFB will bring more phase distortion to ruin the sound quality.

 

[revintage]

So Siliconray,
You actaully said:

THD datas, THD1 is THD with OPT cathode feedback, THD2 is non OPT cathode feedback

This doesn´t mean you only add GFB in one case.

Give us the correct figures about what happens with GNF applied in both cases. What you show doesn´t tell us anything about the cathode feedback effect.

Miles: The vendor made a claim and anyone could understand it was with and without GFB. Had to ask to get a correct answer.

 

[ThorstenL]

Hi,

Do you want to share some more details of your amp. Would be appreciated. Why the shift from output triodes?

Following my Breakup/Divorce I was left without any system (and money)...

I have been slowly building something up again, using was was available to me at low cost.

So my speakers are now only around 89dB/2.83V/1m and with around 6 Ohm Impedance. So an amplifier with more power is needed than a 300B can deliver and preferably cheaper.

My one was build on a "special sale" (scratch/dented/old series) chassis of the diyhifisupply Ella, that I got a good price on.

My circuit is brutally simple.

ECC83, both sections in parallel into a 150K Anode load (Kiwame), around 90V or so on the Anode, there is RC decoupling circuit there, 10uF Ansar MKP cap's, can't remember the series resistor value, or the cathode resistor (unbypassed).

This is DC coupled to a concertina splitter using 8K2 anode/cathode resistors (mint green silver leadout NOS Painton wirewounds) and either 6SN7 or 6N12P (kind of like an octal 5687 from russia, with love).

I use 0.1uF Silver/Mica caps as coupling cap's, bias comes from the diyhifisupply active bias module via 470K.

Output stage uses the 16 Ohm and 0 Taps to the cathodes and the 4 Ohm Tap to ground (so each tube has 4 Ohm feedback tap) and ultralinear feedback to the screengrids. Tubes run at 370V/60mA, so around 20W anode and 2W screengrid dissipation with a 5K anode-anode load. This is fairly far into class A FWIW, Class AB only on peaks.

Output Transformers and the rest of the Powersupply is original diyhifisupply Ella, CLC filtered per channel and with upgraded diodes and capacitors.

I measured around 35W for 3% THD, due to the circuit the Amp clips and distorts a little asymmetrical, not unlike a good SE Tube Amp.

It suits my speakers well and sounds quite good.

Ciao T

 

[Ejam]

Thanks Thorsten

Thorsten

Thanks for the added information on your circuit. I remember Eisenson did a modification of the Dyna Stereo 70 back in the 70's using the same cathode feedback arrangement. The front end was similar to your's but with a pentode first tube to a triode concertina phase splitter. I think he retained the 7199 for these duties. Perhaps, an ECF80 would be a better choice. Nice to hear things are working out.

Regards