PL84/EL86 push pull

[hidnplayr]

Hi Fellow tube-freaks,

I'd like to present my latest creation, a PL84 push-pull amp.

Perhaps the most interesting thing is that I have tried to use the same output transformers before, with 807's in AB2.
Back then I could not get the wanted results without using gNFB (read: plate to plate feedback - measurements)

Now, I also connected the transformers secondaries to the cathodes of the PL84's.
This has been discussed on different threads here recently, and here: diytube.com :: View topic - Koren Local Hero and Mods

Responce is measured at 1Watt into 8ohms resistive load.
The nasty looking bode-plot is open-loop response.
Full output is more then 20 Watts (again into dummy load)
more pictures: PL84 push pull - My Photo Gallery

Any further suggestions?

[Tyimo]

Hi!
Nice and interesting amp!
How is it sounding?
What is your ballanced source?

Greets:
Tyimo

[hidnplayr]

Thanks Tyimo,

I dont like to desribe sound. Sorry (I think it depends a lot on the speakers too.)
Lets just say it sounds perfectly good to me.

The plan is to build an RTP3C clone (Vacuum State - High End Hifi Equipment)
So i'll have balanced phono.
I also plan to build a balanced TDA1541a dac with a preamp stage similar to the RIAA stage in the RTP3C.
Later on, I might do the same thing to my reel-to-reel tape deck.

Yes, I plan to go fully balanced but have some work to do!

[Tyimo]

Hi!
How do you make the bias for the PL84 tubes?

Tyimo

[gianis]

I think ( on the schematic) that the resistors R16 ,and R20 may be connected at the other leads of R4, R5 , otherwise no exists NFB

[hidnplayr]

Gianis: you are right, I made an error while drawing the schematics.

Tyimo: I used two potmeters between 0 and -30V in a way so that I have balance and BIAS.

Here is a corrected and more complete schematic:

[Globulator]

I tried sitting the cathodes in the output of the OPT like that once, it did sound fairly good but later I decided to take the OPT out of the feedback loop completely (I wrapped feedback around the driver+output tube to present the OPT with a low impedance drive, which sounded better).

Your use of the 16ohm and 4 ohms taps seem to give more feedback to one side than the other - 12 ohm output to the top and 4 ohm output to the bottom - is that supposed to be like that or should you connect the earth to the 8 ohm tap?

[hidnplayr]

Originally Posted by Globulator

I tried sitting the cathodes in the output of the OPT like that once, it did sound fairly good but later I decided to take the OPT out of the feedback loop completely (I wrapped feedback around the driver+output tube to present the OPT with a low impedance drive, which sounded better).

Your use of the 16ohm and 4 ohms taps seem to give more feedback to one side than the other - 12 ohm output to the top and 4 ohm output to the bottom - is that supposed to be like that or should you connect the earth to the 8 ohm tap?

4ohm + 4ohm on an OPT gives 16 ohms
You need to calculate with voltages, not impedances in this case.

This kind of feedback should give a better damping factor, now it depends if your speakers like that or not.
Newer speakers like it most of the time.

[Fenris]

Congratulations, it looks nice. Looks alot like Broskie's Power Buffer that I've been wanting to build: Aikido Cathode Follower & More Triadtron

[smoking-amp]

Not trying to shoot down your design. But I do see a conflict of sorts between the plate to plate local feedbacks and the CFB on the outputs. The Schade fdbk makes for a low Z load for the driver triodes, and the CFB makes for a large voltage swing requirement besides. I would try putting the Schade fdbk resistors back to the driver cathodes (with some resistance between them), and a CCS load on the driver plates. Then the driver triodes can develop the high voltage drive easily and linearly. May require some more input signal though.