18Watt.com

I've been thinking about the gain structure of the reverb stage that I'm adding to my 36W amp project. As it stands, the reverb will use a single 12AT7; one section to drive the transformer, the other to amplify the spring output. However, I've noticed that quite a few of these circuits use a 12AX7 for recovery, so I'm thinking about what to do if the 'AT7 doesn't have enough gain. Rather than install another tube, maybe this is one of the few places in a guitar amp where a solid stage device might not be a disadvantage. The signal level is too low to ever overload the stage following the springs, and a FET would increase the gain enough to more than compensate for the lower mu of the 'AT7. With all this in mind, I'm wondering if anyone here has ever tried a FET/tube cascode circuit like the one below? It's been used successfully in phono preamp stages for hi-fi, and I'm thinking it might be a good option for this as well.

Have you considered using a ECC832/12DW7, one triode is 12AX7 the other is 12AU7.

geoff 1965 wrote: ↑

Sat 01/09/21 2:21 pm

Have you considered using a ECC832/12DW7, one triode is 12AX7 the other is 12AU7.

Yes, that's another option, maybe better. I modeled the hybrid cascode in SPICE, and it turns out to have a big issue with power supply rejection. In fact, it's almost non-existent. This means the cascode would require a B+ regulator to keep the stage quiet. That's not difficult, a simple LR8 IC can handle it, but it adds one more thing to go wrong. I won't spend more time on this for now, maybe the 'AT7 will be enough. Thanks for the suggestion!

Jack

Here's another possibility, should have thought of this sooner. The cathode of a 12AT7 is biased to +9V with a Zener, and that voltage powers a FET preamp stage connected in cascade (not cascode) with the tube. This eliminates the issue of PSRR and should be totally transparent to the low-level signals coming off the reverb tank pickup. Where gain of a 12AT7 is typically about 33, this configuration bumps that to nearly 90. I'm sure someone has done this before, although I don't recall seeing it. THIS is the solution if reverb recovery gain becomes an issue.

here's a schematic of the 12DW7 reverb Jack,might be of use to you.

geoff 1965 wrote: ↑

Sat 01/16/21 9:41 am

here's a schematic of the 12DW7 reverb Jack,might be of use to you.
MraG58I.jpg

Thanks Geoff! The medium-mu triode in the 7247 is capable of driving the reverb tank with roughly the same current as the paralleled 12AT7 that Fender used. I'm cutting the chassis for a single 9-pin socket at the reverb, so your suggestion will give me another option when the build is being finalized.

BTW, I'm making some progress. The front panel is now anodized and engraved. The rear panel is in the final stages of same. All the front and rear pilot holes are drilled in the chassis. I'm really looking forward to finishing this mechanical work so I can start wiring!

Jack