EL84 PP bias & load setting - schema review

[tristanc]

This post follows on from https://el34world.com/Forum/index.php?topic=32040 - where I set out the motivation to build a small / light amp for small pub jam nights.

What I've built is set out in the attached schematic, with measured voltages. It's been cobbled together P2P, isn't pretty, but low on hum and static. And works.

It's basically Merlin's High Gain preamp with the 2nd and 3rd cathode bypass caps removed - to be closer to Merlin's medium gain design. It's a bit loose at higher gain settings, but that's not the intended use (blues, more clean stuff).

The power section is 2 EL84s with cathode bias. Vanilla stuff.

My questions are around the bias and load. The HT is 340V, with screens at 280V, with the cathode at 10.8V through ~200 ohm resistance. The OT load is 8.2k in to 8ohms.

Using the loadline plotter at https://www.vtadiy.com I get a portion of the line outside the SOA. As I've used a Hammond 125E OT, I can alter the load to 11.6k by swapping taps which takes the line back in the SOA.

Should I bother?

Or should I bias it a bit cooler (12 or 13V) to get that red dot down a bit?

[HotBluePlates]

Should I bother?

No, it's fine.

It doesn't matter if the loadline strays above the dissipation curve for only a short time.  It's the average dissipation over time that matters.

The other thread also says B+ is 325vdc, which would cool things off all on its own.

[tristanc]

Should I bother?

No, it's fine.

Thanks - that saves me some work.

I've yet to scope the output / trace the signal to make sure there are no gremlins. I'm aware of one that may exist: the cathodyne / source-o-dyne phase splitter had a weird 'glitch' on one of the outputs that made the trace jump about. I put that down to a cold joint. Not audible but strange.

I swapped speakers (from a V30 style to a Neo V-type to a Neo Copperback) which has brought the thing to life. Sounds great at low volume and will test at high volume today.

[pdf64]

There's usually protection zeners between mosfet gate and source.
The R8 stopper should be immediately at the gate.

[tristanc]

Interesting - I had left those out hoping the internal one (if there is one? Should check) would do.

Attached is the current PCB layout - the highlighted resistor (R2) is the stopper, so pretty close to the pin. I have been meaning to get more boards made as I made an annoying mistake on the ones I had fabricated. Perhaps worth doing now!

So back to back zeners (BZX84C10L) where the pink line is.

[tristanc]

Forgot to attach!

[Merlin]

They don't need to be back to back, one will do.