After tweaking the basic Ceriatone 18W TMB circuit I have finally achieved near perfection in sound.  Having built four amps in the past two years this amp gets it right sound wise.   Plexi sound at ear friendly volume.  A pleasure to play!

Enclosed is a schematic to share.  OT subbed with a Mercury full stack which gets better bass extension without a lot of fartiness.   Highs are smooth and not harsh.  Finally solved a too much gain issue in the TMB channel by substituting a 12AU7 in the tone stack driver (V3 cathode follower).   The uF signal caps were swapped with Sozo vintage and the plate resistors NOS Piher carbon film.  Line level FX loop allows the use of time domain effects (G Force in my case).  The rest is all Ceriatone and Nik's wonderful interpretation.  It sounds great!

Guitar Amp Schematic Microsoft Cloud Drive... Click below in a browser, choose "Ceriatone 18W TMB" directory, and download .pdf files.   "18Watt_TMB_Custom_A6.0.pdf" has all the changes made. 

https://onedrive.live.com/?id=636DC4E8060D66A1%21338&cid=636DC4E8060D66A1

Thanks HH.  The tube placement was a head scratcher for me as well as I followed the Ceriatone layout which swaps V2 and V3 for some reason.  So that is the reason for swap in the schematic.

I did not try a 12AT7 in the PI as that would effect the normal channel's gain as well as the TMB.  The 12AT7 is only about 0.2dB less gain than the 12AX7 anyway in this PI circuit with the selected components.  I did some gain calculations and it boils down that the 12AU7 cathode follower gain is about 9 dB less than 12AX7.  This lower gain helps balance the overall amp gain, drive the FX loop close to nominal levels, and keeps noise low. 

For some reason this 18W amp gets the trebles right for me.  So it has gone from a ragged sounding amp to a lovely gem.  Thanks for the kudos... Scott 

Edit: A year after gain balancing and mods I have nothing but praise and love for this amp.  I have used Mercury Magnetics OT transformers in the past with OK results but this one is exceptional and perhaps a large reason I like this amp so much.  The OT full stack with added iron increases inductance so it is harder to saturate the core.  Core saturation distortion is not my thing.  This amp does not have the fizzy sound you hear with so many low wattage amps.  You know, the fizzy decay of notes when turned up.  So does it sound stiff and overly clean as a result?  Not on your life.  It is big and open with a warm sound at any volume that begs to be cranked up.  Bass is tight, upper notes smooth, and quite loud before full on distortion.  With Nik's nice head case and super looking panels it is hard to tell what is under the hood of this amp.  Friends are amazed it is only a lowly 18 watt EL84 amp.

No rewiring necessary. A 12AU7 is a drastic reduction in gain, and it doesn't just affect the CF driver stage, but also the 2nd gain stage. That's probably where the reduction in gain is happening...cathode-followers don't add a lot of gain, they just convert the signal to low impedance and add drive current. You can also try a 12AY7, 12AT7 or 5751. But it won't hurt to try it and/or leave it if you like it. You won't get any more clean headroom (increase clean stage volume), you'll just clean up the preamp tone. The power-amp is still going to be roughly 16 watts RMS.

As for the bass, that's pretty inherent in the TMB design, the single tone control stack on a traditional Model 1974 is much more balanced, but everyone wants the TMB controls.

Things to try...
A.) Lower value cathode-bypass cap (V1, Pin — .68uF, .56uF, etc.?
B.) Lower value coupling cap going into the PI (V2, Pin 2) — .01uF
C.) Lower value Bass pot

Been a while since I was around this forum but still using and liking the 18 watter represented in the schematic.   The open loop gain between 12AX7 and 12AU7s is large but in this circuit yields only about a 10dB difference.  See the note on the schematic. Yes it is just a tube swap.  If you don't like the sound or need more gain pop a 12AX7 back in.  The EL84s need less drive than EL34s so the TMB stage adds unnecessary gain IMO and was hard to control.  Plus a lot less circuit noise and more clean range with a 12AU7 cathode follower.  Not sure about the EF86 preamp 18 watter but give it a try (V3 is also cathode follower in the EF86 version).  

I don't use old stock tubes as I don't think it buys me much in these amps.  For the 12AU7 I use a JJ ECC802 which is a long plate variant of the 12AU7.  All I had at the time but I like it and works well here.

Interesting about the ECC832 but the cathode follower section (2nd stage of V3 or V2 in regular Marshall amp speak) are almost unity gain no matter which tube type selected.  I would just use a standard tube for desired gain/sound provided by the 1st stage of the cathode follower circuit and the 2nd stage driving the tone stack will be near unity gain regardless.

Wyatt has some good ideas about bass response.  I changed the V1 cathode bypass on the TMB channel from 1uF to 2.2uF to add a little extra bass for a Fender strat.  Gave it a little more balls.  But I wouldn't say the amp is overly bassy by any means with the 12AX7 front end.  The EF86 version I haven't heard.  But of course I encourage experimentation so go for it!