Princeton Reverb - ECC823 PI

[gtomax]

Just wanted to share a fun experiment/learning: acquired an all original 1965 PR and had a few issues to take care of. Power recap, reinstall/restore an NOS amphenol power socket (someone had taken it out and bolted an IEC in its place for a "removable power cable"), broken reverb spring, clean/lube pots, store original Oxford and install 66 Jensen C10P (so sweet).

Ended up with a nice clean powerful good sounding PR! It does what PR's do, that upper mids distortion thing.

Tons of people want a baby DR and they mod sockets, change the PI B+ node wiring, replace the OT all manner of thing. That good PR vibe comes a lot from a strong driving cathodyne PI and modding the circuit just doesn't feel right.

Solution: bought a JJ ECC823 which is a 12DW7 in reverse. 1/2 of V4 is for the tremolo and the other is the PI so usually it's a 12AX7 and you get that raucous PR tone. With the JJ ECC823 the PI side is a AU7 triode while the tremolo is an AX7 as intended for proper tremolo wub-wub-wub-wub-wub.

I believe what's truly happening is yes it's an AU7 stage but it's also changing the bias point of the tube to pretty cold. In this case that's desirable for the intended effect.

Result: delicious harmonics, more balanced overdrive that happens a little later and more usability out of the bass knob. It feels like a plug and play PR -> DR mod and generally is very good sounding. Easy to restore back to full vintage PR with a simple tube swap.

YMMV! Live long and prosper

[Tony Bones]

Nice! Thanks for sharing.

So you just plugged the ECC823 into the socket without changing the circuit at all?

I once tried something similar with a 5E3ish circuit, but I changed all of the resistors in the concertina to 'optimize' the operating point. My goal was a beefier driver than a 12AX7... actually quite different than your cold biased tube. Interesting.

[gtomax]

Correct.

A 12DW7 (JJ's ECC832) is half 12AX7 and half 12AU7. Awesome tube for something like a Trainwreck when you want to tame gain from the second stage for example. This wouldn't work in V4 of a PR as a plug-in because it'd put the 12AU7 on the tremolo side and keep the 12AX7 on the PI side. You'd end up with the same sound but a crummy trem.

JJ's ECC823 is a reversed 12DW7 and as far as I know the only one in existence. This puts the 12AU7 on the PI side and keeps the 12AX7 on the trem side.

Plug and play!

[TUBEDUDE]

The 12DW7, and it's commie cousin the 832, are perfect for effects loop use. The 12AU side to cathode drive the loop and the 12AX side for loop recovery. I use at least one 12DW7 in each build.

[M Fowler]

https://www.tubesandmore.com/products/v ... ual-triode

Show the differences in the ecc832 vs ecc823

[Stevem]

You should play around with the matching and unmatching of the PI 56K plate and Cathode resistors to audition the range of clean, crunch and dirty tones.

[tubeswell]

There are other benefits to using the AU triode for the PI.

You can decrease the load resistors to something like 10k* or 22k each (instead of 56k) and get more tube current and much better impedance bridging to drive the 6V6 grids. The lower you go, the more you will want to think about changing the bias resistor to get about 6V to 8V bias (to keep the AU in its happy place). Then you can also decrease the 6V6 grid-leak resistors to 100k (instead of 220k) and still have good impedance bridging, and bias the 6V6s to idle a bit hotter (because there will be better grid leak function).
*Make sure you use 1W or 2W rating

Then you might also want to use a red LED to bias the LFO triode, and decrease the LFO 1M bleed-out resistor (between the LFP plate and the coupling cap to the depth pot) to 470k (or 220k or 100k - you might want to play around with this) to get more hypnotic tremolo. You could also play around with increasing the cap values in the LFOs C/R-C/R-C/R network to alter the term wave form to enhance this effect. Try adding 0.01uF in parallel to each of the caps (one at a time).