2204 gain mods

I'm sorta torn, but in an attempt to diversify my home studio's amp collection I am considering "Mesa-ing" my 2204 build. By "Mesa-ing" I mean going a bit nuts on adding gain and bottom end.

Gain mods so far:
The 2nd triode's cathode is 5k instead of the stock 10k.
The tonestack's input side cathode is bypassed with a 4.7uF.
Presence circuit is less feedback like a plexi's (47k, instead of the 2204's stock 100k).

Bass mods so far:
1st triode's cathode is 3uF.
Tonestack's bass cap is 0.1uF, instead of the 2204's stock 0.022uF.
Coupling caps to the power tubes are 0.1uF.

Its gainy, but I'm thinking of going further.

Perhaps:
Going even lower than 5k on the 2nd stage cathode (maybe even bypassing it? IDK).
270k or even 100k resistor to the Pre/Gain pot.
270k or even 100k resistor to the next stage's (tonestack) input.
A larger cathode resistor on th tonestack tube (its 820 now, maybe 1k to 2.2k? IDK).

What do you guys think?

PS
This is not just a studio amp, it'll be my main live amp in a three piece hard rock/punk band. I can always dial back gain on the Pre knob live.

Oh come on, up the anode resistors to 330k. You will have to up the stage voltage to get any current flowing, but remember, for every 3db you up the gain wiht a larger plate load, you increase PSRR 6db

I think that the Mesa to borrow from would be a MkIV. JMO. The trick is to ferret out which options you want to hardwire (those things have a zillion relays).

There's a lot of well thought out cap and resistor combinations in the MkIV - stuff that I haven't seen too often in higher gain amps. Bigger cathode bypass caps, .047uF coupling caps, etc.

The biggest unknown with using that amp as a model is the slider EQ, which I would personally never mimic in a build.

While I'm not a big Dream Theater fan, I really hated it when Petrucci bailed on his MkIV's. His tone is so much more kick a$$ now that he's gone back to them, IMO (last two albums). If you can get past the 8,000 knobs and switches, it's a pretty cool high gainer. IMO, it smokes the rectos (as well as SLO's, 5150's, etc.) because it's got gobs of old school stuff in it.

Grab the schematic from Schematic Heaven and have a gander. It should give you ideas galore!

Oh come on, up the anode resistors to 330k. You will have to up the stage voltage to get any current flowing, but remember, for every 3db you up the gain wiht a larger plate load, you increase PSRR 6db



I was thinking of keeping the 100k and hadn't considered kicking them up a notch. This could be another option for me though. I was seriously considering a bypass on the cascaded stage's cathode and swapping those 470k resistors feeding the plate to the 1M Pre pot and the one feeding the tonestack tube's grid. I figured if I add small bits in several places it'd keep things tameable and still musical.

I'd really like this amp to have the fat midrange roar of my modded plexi build - but keeping the woof out of high gain preamps means I might not be able to have my modded plexi's bottom.

As far as actual Mesa borrowing 11 Gauge, I was looking at a nod towards the Triple Rect in sound, hadn't really considered the MkIV - it has a plate fed tonestack so it sorta just slipped my mind. I always thought the unusual cathode values in Mesa's preamp must have come from a great deal of trial and error, both the MkIV and the Dual/Triple Rect have interesting values.

I'll start small again. I think I'll swap the 470k+500pF grids - I've already clipped the 500pF's and juiced the cathodes of V1 and V2 a bit - perhaps then I'll fine tune towards Mesa values. While I want massive overdrive, I want to take baby steps and avoid overkill in one or more stages that might result in the sonic equivalent of a Smurf's woodchipper.

Boot strap the CF to the stage before it.

Split the anode load in half, return some of the CF's cathode signal to the junction of the load resistors with something around 22nf.

Boot strap the CF to the stage before it.

Split the anode load in half, return some of the CF's cathode signal to the junction of the load resistors with something around 22nf.



I'm still new to this in many ways. Is this essentially what some circuits do with the cathode-to-plate caps?

As it is, I know I can unlock a ton of gain in the 1st stage and cascaded second stage. I'm tempted to simply go with a 1.5k instead of the stock 10k... after this I know I can reduce some of the interstage attenuation by swapping those resistors to get even more gain. I'm just not sure how much I should go with since I don't want a fizzy, buzzy sound for my #1 live amp. My attitude is that I can always use the Gain/Pre know to dial back to more classic tones... though in practice its not that simple since it depends on how much gain and where in the chain it comes. Bypassing cathodes in V1 and the tonestack tubes seems to make bigger differences in bass than changing coupling caps - though they can get woofy if overdone early on, or too many times in the chain.

So my main concern is a fat, full sound - without the subsonic woof and with enough gain to lay down industrial/metal without the need to pull out an overdrive/distortion pedal (which is a ton of gain and fatness - which is what Mesa's Dual/Tripl Rectifiers are already doing out of the box). I know its a tall order, but if I chose my gain for each stage sparingly and trim fat only in specific spots, I think I can get it fine tuned to this tonal area. I already built a JCM800 style amp and don't have the cash (or desire) to build a Mesa from the ground up.

To be completely honest, my 50 watt "plexi" build has 200% the tone I could ever want (fat, beefy, raw, and raunchy - with plenty of dirt, tons of soul, and all musicality) - but like I said, I need to diversify my studio herd... besides the JCM800 style pre PI MV was easier to build into a footswitchable 2 channel amp (I needed 2 channels for live use) with seperate Pre/Master controls for each channel - my 50w plexi uses a PPIMV which is harder to build around 2 independently volume'd/gain'ed channels... unless I wanted to use a million DPDT relays and add a hurricane of extra noise.

As it is, I know I can unlock a ton of gain in the 1st stage and cascaded second stage. I'm tempted to simply go with a 1.5k instead of the stock 10k... after this I know I can reduce some of the interstage attenuation by swapping those resistors to get even more gain.

The good news is that all of the "3 Plus 1" high gainers tend to use very similar preamp structures. My suggestion is to grab a schemo of a Recto, SLO, and even a Bogner Ecstacy.

With the "3 Plus 1," the equivalent of the 10K cathode stage is moved to slot number 3, with the add'l stage inserted before it. You can typically run the 1st two stages as hard as you dare, but you've got to pad things at that third stage. Then things are nice and happy when they hit the CF.

...In the case of the Recto, SLO, etc. - it's always achieved with a 39K cathode resistor on the 3rd stage, but they're monkeying around with the B+ rail a bit more than your standard Marshall (hence the need for 220K plate resistors). You have to pay attention to the voltages at each node on the rail - don't be surprised if you have more at the plate of a pre stage than you do at the PI, which is counterintuitive.

Almost all of the "3 Plus 1" interstage stuff is very similar - the typical grid attenuation stuff piggybacked with a treble peaker. The difference is the values - the peaker caps are occasionally 220pF, 560pF, or even .001uF. And of course they tend to use 1M stoppers and liberal use of leak resistors as well.

The other "3 Plus 1" necessity is to typically put small caps on the plate resistors - 500pF to .001uF. This really helps to keep down the harsh high end, but the tradeoff is that you lose a bit of sparkle when you drop your volume knob. If you can keep it to just one stage, the effect is minimal.

Something else not to rule out would be a stage with local neg. feedback. Fender did it with the SF Bassman heads on the "additional" gain stage. It's easy to wire up, easy to tweak, and adds LOTS of tightness and compression. There's a reason why the rockers like to grab those heads and tweak them for raunchy guitar.

You might also want to check out some of Steve Ahola's mods, since he's done comprehensive work playing around with local neg. feedback, conjunctive filters, and similar things that allow you to bring down the sledgehammer, effectively.

But getting back to your original query - I haven't seen any sort of experiences where anyone has gone below 4.7K on the stock 2nd stage with much success. It's a real PITA to try and minimize the blocking distortion when going much lower. Not to say that someone hasn't done it, but the consensus is that it's a real hat trick.

With the "3 Plus 1," the equivalent of the 10K cathode stage is moved to slot number 3, with the add'l stage inserted before it. You can typically run the 1st two stages as hard as you dare, but you've got to pad things at that third stage. Then things are nice and happy when they hit the CF.



I'm starting with the interstage 470k resistors first. I've played with thsoe in a cascaded tweed build and I feel like it won't harsh things up.

If I don't find what I'm looking for in gain and tone, then perhaps I'll go this route. 1st two stages are the cascaded preamp, 3rd stage is the tonestack... I really want to keep things simple.

Perhaps I'm looking for something that can't be done simply - I love my JCM800 build, maybe I should I just leave it alone and forget about diversifying my studio amps. I gotta tell you though, last weeks tweaks on my plexi turned it into a monster, and now the Normal channel is really cool for Bass guitar. The sound I really need for my band is ACDC (chimey, nearly clean, musical overdrive), and my plexi does that on steroids now.



You might also want to check out some of Steve Ahola's mods, since he's done comprehensive work playing around with local neg. feedback, conjunctive filters, and similar things that allow you to bring down the sledgehammer, effectively.



I haven't checked out BlueGuitar in a while. I'll dig around in what what Steve's got for more ideas.



But getting back to your original query - I haven't seen any sort of experiences where anyone has gone below 4.7K on the stock 2nd stage with much success. It's a real PITA to try and minimize the blocking distortion when going much lower. Not to say that someone hasn't done it, but the consensus is that it's a real hat trick.



I've done it on an 18w Stout build. Even before cascading, the PI had that terrible "18w buzz" which I fixed with a combination of diodes and zeners to the power tube's grids. After I fizxed that I spent a good week tweaking the gain to get a monster metal amp...

Its got 2.2k+1.2uF on the 1st stage cathode and 2.2k on the second. Its a filthy amp at 2, Monster Magnet at 3, and Manson/Rob Zombie on anything higher than that. Its only got two stages (1st and cascade), no tonestack (just a tone knob), and the EL84's aren't even driven hard since I use a PPIMV to dial them back.

OK, this is where I'm at as of today:

1.5k with 1uF one the first cathode. 5k on the second. Tonestack cathode is bypassed by a 4.7uF and is hit with a 235k (two 470k in parallel) input instead of the 470k.

Perhaps not the outladish gain factory I was hoping for, but very musical, fat, and ACDC "chimey." The tube in my "Channel 1" is darker and less gainy than Channel 2's, but I still have nice fat, warm, cleans on either.

What sucks is that I need to still be able to dial back one channel's gain for warm fat Cleans and get insane crunch on either channel if I need a Clean/Dirt channel setup - this means finesse in my approach.

I might want a bit more gain, but I'll give the a few days as usual, before making another tweak.

I'm gonna have to go through this thread with a fine-tooth later.

For now, I'll mention that at one point I dropped a .0033 cap in series with my NFB wire, and eventually reversed it, because it gave me much more low end that I really needed, but for that over-the-top thump, it might be just the ticket.

I tried 1.5k on the 1st cathode once. I thought it was a little too buzzy. We're talking Mesa though...