Marshall el34 PI values to el84

[roberto]

Your B+ should be around 370V, or at least this is the mean value of the JCA20 I've seen.
Just to be clear, the suggestion by matt h is to connect a 68k to the anode, a 33k to the supply node, and take the signal from the point between the two resistors.


EDIT: thank you matt h!

[matt h]

(deleted)

[matt h]

(deleted)

[martin manning]

If the intent is to reduce the PI output by more than half via split-load, then the smaller resistor goes next to the supply node as Roberto said.

FWIW, using a 47k in the tail and 15k plate loads as in the scheme Steve showed will produce very well balanced signals with about 8dB (~60%) reduction.

Any of these schemes that reduce the output post-PI will probably want some increased FB, i.e. make the FB resistor proportionally smaller.

[rock_mumbles]

I've messed with a couple of JetCity 20 watt amps ...

We redid the PI to 18 watt values and in one amp reworked the preamp to be like an XTC (I think) it turned out to be a nice amp but the pcb in that one was bad ... the traces were lifting all over the place ... I need to get back in and redo it with a turret/eyelet board.

The NFB in the stock JetCity circuit is way too much to have any power tube character ...

[matt h]

(deleted)

[Colossal]

If the intent is to reduce the PI output by more than half, then the smaller resistor goes next to the supply node as Roberto said.

FWIW, using a 47k in the tail and 15k plate loads as in the scheme Steve showed will produce very well balanced signals with about 8dB (~60%) reduction. Any of these schemes that reduce the output post-PI will probably want some increased FB, i.e. make the FB resistor proportionally smaller.

A 47k/15k anode resistors probably gives real nice balance. Way too much of it. Completely neutering the goddamned glory of an overdriven PI stage, central to the cranked amp sound. By the time you're done removing voltage drop across the tail and presence circuit, with an already lower than original (read: 50 or 100Watter) voltage, you're running out of head *and* foot-room. And canceling out more harmonics generated within the PI.

From Mark Huss' site. Even though it pertains to his original 6V6 Plexi, the content is relevant because he looked at various PIs and their effects; the November PI included.
http://mhuss.com/Plexi6V6/Scope.html

"I have tried plexi PI (12AX7 100k/82k/15k), "BF Deluxe" PI (12AT7 100k/82k/22k), and November R10 (12AT7 15k/15k/47k) values so far. I subjectively liked the November sound best before I started the scope-out."

[matt h]

(deleted)

[Colossal]

The november PI he's talking about is also a 12AT7, not 12ax7. Sonic difference. May sound good, but it's moving away from what OP has/is doing.

It's also interesting to think about the 6v6 plexi on the whole... I wonder what'd happen if someone built one of those at sane voltages to begin with, and scaled the rest of the design appropriately the first time around, as not to have to resort to mutant PI values to bandaid the snot out of the thing.

Yeah, just figured I'd mention it...since this discussion involved the November PI vs. split load vs. full traditional PI.

Mark went with the traditional Marshall PI values for the 6V6 Plexi, as did I and that amp rocked...hard and loud. 6V6s though, not EL84s as is the topic here.

Yes, it would be good to revisit/reboot the 6V6 Plexi. That is a killer amp. I built mine with high voltages.

[Phil_S]

But EL84's at, as has now been contributed to the thread, 370V, fixed bias, are not exactly to be known long to the world. If you're gutting the amp, I'd seriously suggest a b+ reduction technique...

This is exactly what I was getting at when I asked about the unloaded secondary HT VAC reading. A PT meant for a pair of EL34's is not a good match for a pair or even a quad of EL84's. While probably too high for ordinary 6V6's as well, you might want to look at using JJ 6V6's. Of course, this points the whole idea in another direction. I am not a big fan of lowering voltage by making heat from it. You might also consider a pair of 6189's or Sovtek EL84M, both of which are EL84 variants that should fare better at such high B+. Just my 2 cents.

[matt h]

(deleted)

[RockinRocket]

Try maybe a 33k/68k resistor split for each 100k plate resistor value to start with and adjust up/down as necessary.

My concern with adjusting the plate resistors is now the PI sounds different than what would be driving el34s. No?

No. Using split-plate loads maintains the sound of the original, but at a reduced level. Because 33k+68k=100k (ok, mathematically not. but close enough. given parts tolerances, they're identical) it keeps the total voltage drop from B+ node to plate the same *and* (this is the important one) keeps the AC load-line the same, which preserves the same, for better or worse, harmonic content.

Ok thanks I understand what you mean. Kinda like how the fender pro junior has a 91k/ 9.1k on one half of the PI plates.

I was really hoping to be able to reuse the circuit board without having to cut a lot of traces. Im leaning towards using the split load though.

The jca really isnt a good sounding amp stock. Seemed like a really good platform to mod. I got it for 160$ with some Amperex rx4 el84s. The output transformer is huge and bigger than the power transformer. Maybe that is one part of the bad sounding amp problem.

[matt h]

(deleted)

[Phil_S]

It's a 2xEL84 amp. He never had EL34s.

Hmmm....obviously, I missed that completely. Still. B+ is too high.

[RockinRocket]

On paper it could be confused with a 800 but its really not. Its a little on the brighter side like an 800 but thats pretty much it.

The preamp stock simpilized -
V1 220k plate, 1.8k 1uf cathode. No bright caps to next stage
V2 100k plate, 2k2 cathode. 470p bright cap to next stage
V3 100k plate, 1.8k 1uf cathode

All Ive done so far is-
V1 100k plate
V3 820 cathode removed bypass cap

Sounds a whole lot better but the amp has to have the master on 10 and the pre amp set to tast (anywhere from half to 3/4 up). But I still prefer to play my 18 watt or ac 15. The Jca is supposed to be my practice amp I can attenuate through my rockcrusher (as my other amps arnt to nice on el84s).