diyAudio (
-   Tubes / Valves (
-   -   The "Stupendous" 6P1P 4PP amplifier! (

kodabmx 13th January 2018 07:49 PM

The "Stupendous" 6P1P 4PP amplifier!
1 Attachment(s)
EDIT: This design has changed many times. Read through the thread. The last posted schematic in the thread is the most current.

So the design uses a PPPP design meaning 8 6P1P per channel.
According to Push Pull Calculator the numbers look interesting.

Since I have the parts laying around I though I might build the following "stupendous" design. Stupendous in the sense that it's about 10% efficient.

In case it isn't clear, the 100k grid resistor is shared. In reality, if this sounds anything like it sims to spec, It's a contender to replace more expensive tubes such as 300b, providing electric power is cheap where you live. I'm fortunate enough to have my electricity included in my rent(!).

What do you all think of this?

Bas Horneman 13th January 2018 08:28 PM

Go for it! Let us know how it pans out.

nigelwright7557 13th January 2018 08:28 PM

I would have just gone for a pair of more powerful output valves than use multiple pairs.

Bas Horneman 13th January 2018 08:31 PM

Reminds me of EAR Yoshino | EAR V12

6A3sUMMER 13th January 2018 08:46 PM

Be careful. Matched tubes may not be matched at the operating point you use them at.

It is much more complex: but separate current sources, and separate bypass capacitors for each tube are suggested (or more easily, separate self bias resistors plus separate bypass capacitors for each tube). The resistors method will give easy measurement of the current in each tube, both when you start new, and when you have used the amp, i.e. for 6 months.

kodabmx 13th January 2018 08:50 PM

I know. But that's what I have. They aren't matched, and they don't need to be. The LM317s take care of that. They ARE all separate, one set for each tube. I guess it wasn't clear in the schematic. The real point of the CCSs is the idle current balance. From my experience these coils are VERY forgiving. I've seen a 10ma mismatch at sound levels that don't reveal it.

EDIT: If they seem to exceed the specs of the 50VA coil I'll upgrade them.
EDIT2: If it's built as shown using only one tube, you can connect the secondary in parallel instead of series for a nice ~3W.

Lingwendil 13th January 2018 09:02 PM


Originally Posted by nigelwright7557 (
I would have just gone for a pair of more powerful output valves than use multiple pairs.

The 6P1P is a fantastic tetrode, and cheap! Basically a 6AQ5/6V6 in a 9-pin bottle. This gets you lots of transconductance (the total Gm will be the sum of the valves being paralleled) at lesser drive voltage than many of the big-boy tubes would require. This also allows for a lower turns ratio in the output transformer, lower copper losses, and often easier to design, also allowing the use of nice power toroidal types for those on a budget.

I love projects like this OP. Reminds me of a big, hot, nasty 5x5 parallel 6v6 triode build I did a few years ago. I fed the grids through an IRF820 source follower in fixed bias per side so I could reach deep into AB2 on high signal levels, and I used 200VA antek power toroids (i think 10volt output) for my output iron. It would go all the way down to 10hz without issues, and sounded fantastic. Efficiency be damned!

I've been thinking of doing something similar with the EL84/6P14P for the lower drive requirements at some point.

Could go full nutjob and run garter bias per each complimentary pair of devices, it would eat some of your supply voltage, but would ensure damn near perfect balance over time. I might do that for the parallel 6SN7 AB2 design I'm prototyping...

kodabmx 13th January 2018 09:11 PM

Remember, I'm using them in triode. You're right, though. They are excellent tubes. However, if I didn't have these tubes already, I would have bought 6P43P. I built a single PP Triode stage (with 6P43P) as a test and now it's hooked up to my TV as the output :)

artosalo 14th January 2018 08:04 AM

2 Attachment(s)
6P43P (EL86/6CW5) is very, very different type of tube than EL84.
It works best with essentially lower supply voltage (300 V) and load impedance. Some 5k is optimum as it is 8k for EL84.

With a pair of UL-connected 6P43P, 5k OPT and 300 V as +Ub, some 25 W is available.
As a triode and above parameters, some 11 W is available.

These are the results with actual amplifier, not simulations.

Attached the test circuit.

kodabmx 14th January 2018 12:38 PM

Great information! I have a push pull 6P43P Triode connected amp running now but with a higher load (8-12k depending on speaker impedance) and it's a great sounding amp.

I have 16 6P43P in the mail, but I have the 6P1P now. Maybe I'll wait for them. I also have 6F12P.

What would the bias current be on your schematic?

All times are GMT. The time now is 08:30 PM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 17.65%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio