Favorite mods to the 18W?

It's time to build another low watt beast. My neighbors are already hating me for shaking the crap out of everything with my EL34 amps...

So it's going to basically be an 18 watt job, but with a few twists. 6V6 output tubes and SS recto will be part of the build.

...I'm just not sure how I might want to alter anything else in the circuit. I know that it's always a huge debate between the original 1974 vs. a T/M/B variation, which many consider to be heresy. I'm actually not hung up on any particular tone stack at all for this build.

I do want a bit more punch than a standard 18 watter, hence the SS recto. But B+ will still be low, probably topping out around 320VDC or so. I don't think that I want to do anything as drastic as adding a choke. And probably no NFB loop, either.

The build won't have trem. I'm not opposed to trying fixed bias in place of cathode bias, but prefer the latter, even if some punch is lost.

So any fun twists to try would be appreciated. I'm particularly interested in cathode bypass cap alterations, as well as coupling cap tweaks. Filter caps are also open for experimentation, within reason. Maybe a bit heavier at the mains for more punch, but really light at the pre to keep some bounce in there.

Another identity crisis amp will no doubt result, but I like to start with a loose foundation that's common knowledge, and the 1974 certainly is, at this point. So any input on tweaks that you've found to be at least interesting would be wonderful contributions!

which value for bias resistor would you use? 150 ohms maybe?

a mod to have more punch and not so high bias es to use 180 ohms bias resistor with a 1000uF o 2200uF cap. This way the amp would act as it was FIXED BIAS RESISTOR... reducing the voltage swing of the POWER TUBES.


besides, do you want it for high gain or normal clean/overdriven sound?

besides, do you want it for high gain or normal clean/overdriven sound?

Nothing high gain, but I'm not expecting cleans out of a low watter like that. Just trying to exploit the punch factor as much as possible, and maybe throw in a few other twists while I'm at it.

270 ohms with cathode bias seems to have served me well with 6V6 builds. I've gone semi high with the bypass cap (~220uF), but not much higher. IME, pumping up the filter caps on the mains seems to work a bit better, for my tastes. But I'm willing to have a go at the 1000uF+ bypassers as well, as they just might end up in some of my SE amps.

I like to set my low watters to about 1:00 on the volume dial and augment the bark with the guitar controls and maybe a very light OD or boost as well, but I tend to keep those rigs sparse. I just really dig them that way.

I like to set my low watters to about 1:00 on the volume dial and augment the bark with the guitar controls and maybe a very light OD or boost as well, but I tend to keep those rigs sparse. I just really dig them that way.

this is also the way I play my amps.... excelente.:mrgreen:

I'd definitely try fixed bias (maybe -20V or so) with your lower B+ and see if it sounds good. No reason to commit to one or the other yet! :lol:

I'd definitely try fixed bias (maybe -20V or so) with your lower B+ and see if it sounds good. No reason to commit to one or the other yet! :lol:

I would have to agree, give fixed a try. You are pretty lean on the a-k voltage for a 6v6, so fixed gives you a bit more.

If you go cathode biased, you might want to give "zener augmented" a try. Parallel the cathode resistor with a zener string of the desired bias voltage.

I would also try out the larger bypass caps, 220uf is not really that large for a output stage.

With hugemongous cathode caps, isn't there a risk of building up enough residual charge to lead to crossover distortion following a big signal? Or is that not much of an issue in practice?

- Scott

You could make a stock el84 18 watter, then get some Yellow Jackets to use the 6v6s. That way you kill two birds with one stone.:grin:

I would have to agree, give fixed a try. You are pretty lean on the a-k voltage for a 6v6, so fixed gives you a bit more.

Yeah - I'm now leaning towards this - might as well use the bias tap on the PT.

With hugemongous cathode caps, isn't there a risk of building up enough residual charge to lead to crossover distortion following a big signal? Or is that not much of an issue in practice?

- Scott

It's a crapshoot. In the case of the 1974X, the 470K feeders have me a little concerned. I'm going to drop them to 220K. I know that the PI is run a bit cooler than most, and the .01uF couplers to the power tubes should lean things out, but I've gone as big as 220uF on the cathode w/o any negligble effects.

But if it's fixed bias, a dinky cap is usually sufficient. My PT has the standard Fender hookup, so it s/b a breeze.

But I'm still torn between a trem-less w/ fairly stock preamp w/single tone control, or the extra gain stage w/the T/M/B after the cathode follower. A real tossup there.

It's a crapshoot. In the case of the 1974X, the 470K feeders have me a little concerned. I'm going to drop them to 220K. I know that the PI is run a bit cooler than most, and the .01uF couplers to the power tubes should lean things out, but I've gone as big as 220uF on the cathode w/o any negligble effects.

But if it's fixed bias, a dinky cap is usually sufficient. My PT has the standard Fender hookup, so it s/b a breeze.

But I'm still torn between a trem-less w/ fairly stock preamp w/single tone control, or the extra gain stage w/the T/M/B after the cathode follower. A real tossup there.



I've built an "18watt" 6V6 (TMB version) as well as normal 18w, Hoffman's high gain 18w Stout, and a a few other variations.

As far as the 18w 6V6 TMB, there were several people having issues with boomy bottom, I was one of 'em. I did all sorts of stuff (smaller couplers, swapped TMB cap values, you name it).

The one-knob is OK. I prefer a bit more control. For more punch, I'd actually stick with the classic EL84s and PI with a V/MV/TMB tonestack, a fixed negative feedback loop, SS rectifier, and a large cathode bypass on the power tubes (to make it act more like a fixed bias amp).

It sounds like you're trying to keep the volume down and are shying away from a high-gain preamp. My 18w Stout has been tweaked to death and it works great - though it does have a high-gain preamp (which I tweaked the gain on a millions times before finding what I wanted) - with the high gain pre, I was able to use a standard PPIMV control (something stock 18w circuits can't really use very well).

If I've got iron that spits out lower voltages I lean towards EL84's for lower volume rockin. If it spits out 350+, I lean towards 6V6's and more volume/cleans.

As far as the 18w 6V6 TMB, there were several people having issues with boomy bottom, I was one of 'em. I did all sorts of stuff (smaller couplers, swapped TMB cap values, you name it).

Yeah - this is what I concluded today. Even after "slimming down" the circuit wherever possible, it's still got too much woof. Really hard to get that trademark 1974X bark and snarl, unless I drop the bass to nearly zero.

So tomorrow I'm gonna try and slim down the tone stack itself. If that doesn't cut it, next step will be the "Minimalist" 18 watt, w/6V6's.

But it's barking pretty good right now - it's just a very tiny usable window. Sounds awesome clean(ish) at about 9:00 on the volume dial.

Yeah - this is what I concluded today. Even after "slimming down" the circuit wherever possible, it's still got too much woof. Really hard to get that trademark 1974X bark and snarl, unless I drop the bass to nearly zero.

So tomorrow I'm gonna try and slim down the tone stack itself. If that doesn't cut it, next step will be the "Minimalist" 18 watt, w/6V6's.

But it's barking pretty good right now - it's just a very tiny usable window. Sounds awesome clean(ish) at about 9:00 on the volume dial.



Now that I remember - mine was a cascaded 6V6 TMB... this might make a difference.

As far as the circuit you're using - which one is it? The one on 18watt.com, or did you work out your own stack?

If its the one on 18watt.com, try a 56k slope resistor instead of their 47k (in my "normal" Marshalls I like more mids with the 33k, but in this case it might add more woof), a 500pF Treble cap, and 0.022uF caps for the Bass and Mids. You might also want to reduce the TMB channel's input cap to the PI from 0.022uF to 0.01, or even smaller.

From there you can decide whether to cut the tonestack's input side cathode to 820 ohms to reduce gain. This seems to be a big "woof" zone in my normal Marshall builds (plexi and JCM800) when I play with a bypass here - so you could try an 820 with a 0.68uF bypass to shave the Bass at the TMB (and add gain, unfortunately). I think that "stock" value of 1.5k is too much.

If I remember correctly, you prefer slightly larger preamp cathodes than the classic "Bright channel" 0.68uF. I like 1uF or even 2uF on the TMB channel's pre and in the Normal channel I like a 10uF... there's still plenty of the classic 18w mids either way.

Also, the 18watt.com TMB layout is flawed in my opinion. Instead of designing a new one, they sorta built on the old classic 18watt circuit's (1974x) layout. This means the PI tube sits between the input and TMB tube and the Preamp tube, resulting in very long PI plate load resistor runs (can be noisey). This also means the TMB and preamp tubes share a filter/B+ node, which I think they should get seperate filter/B+ nodes.

did you work out your own stack?

Yep - fairly typical (bigger) Marshall:

500pF treble cap
.022uF bass & mid caps
250K treble pot

Atypical stuff:

50K slope
250K bass pot
10K mid variable resistor to ground

And yeah - I would never sandwich the PI in between gain stages like I've seen done by others. Just plain stupid, IMO.

After playing around with hotrodded SE amps, I've come to appreciate the usefulness of .01uF caps, especially for reducing blocking distortion in pigmy amps. Before I do anything else, I'm going to put a pair of those in for the bass and mid caps in the tone stack.

Well, I went back inside the chassis today and did some extensive changes.

After reviewing the Hoffman 18W Stout, it got me to thinking about Doug's use of the extra gain stage (and sandwiching a plate loaded TMB at that point in the circuit). If I used it as a hardwired recovery stage after the TMB, I could lose the cathode follower and much of the stuff that was added to the TMB variation of the 1974X. So that's what I did - V1B has a 220K grid stopper and a 2.7K cathode resistor.

...But this indeed did still require slimming down the tone stack. I now have .01uF caps for bass and mids. The bass knob can now be turned well past 12:00 and still be usable. And even when it starts to get loose, it still has a cool vibe to it - not tons of farty blocking distortion. Very old school.

So I guess it's a 1974X in spirit only, but at least I tried to keep somewhat true to certain aspects of the original. :wink:

The funny thing is that I bet if you plugged this particular tone stack into Duncan's calculator, there might be some overlap with the stock 1974X tone control. And I know that there's a percentage of folks who replace the caps in the stock stack with a 500pF/.01uF combo anyways.

So now the final thing to do is go back in and pull the toggle for the big cathode bypass cap for V1A, which broke on reassembly, and which I won't need any more - plenty of bass w/o it. It's simply a 22uF cap that's switched in along with the a hardwired 1uF/1.8K combo. At least it snapped with the contacts open - thank goodness for that!

What a cool little low powered beast. I even hear some occasional hints of JTM45 in it, which is surprising. And I was having fun doing my best (lame) version of some Live at the Fillmore East stuff.

Well, I went back inside the chassis today and did some extensive changes.

After reviewing the Hoffman 18W Stout, it got me to thinking about Doug's use of the extra gain stage (and sandwiching a plate loaded TMB at that point in the circuit). If I used it as a hardwired recovery stage after the TMB, I could lose the cathode follower and much of the stuff that was added to the TMB variation of the 1974X. So that's what I did - V1B has a 220K grid stopper and a 2.7K cathode resistor.

...But this indeed did still require slimming down the tone stack. I now have .01uF caps for bass and mids. The bass knob can now be turned well past 12:00 and still be usable. And even when it starts to get loose, it still has a cool vibe to it - not tons of farty blocking distortion. Very old school.

So I guess it's a 1974X in spirit only, but at least I tried to keep somewhat true to certain aspects of the original. :wink:

The funny thing is that I bet if you plugged this particular tone stack into Duncan's calculator, there might be some overlap with the stock 1974X tone control. And I know that there's a percentage of folks who replace the caps in the stock stack with a 500pF/.01uF combo anyways.

So now the final thing to do is go back in and pull the toggle for the big cathode bypass cap for V1A, which broke on reassembly, and which I won't need any more - plenty of bass w/o it. It's simply a 22uF cap that's switched in along with the a hardwired 1uF/1.8K combo. At least it snapped with the contacts open - thank goodness for that!

What a cool little low powered beast. I even hear some occasional hints of JTM45 in it, which is surprising. And I was having fun doing my best (lame) version of some Live at the Fillmore East stuff.



This is exactly what Phil (AX84.com and now 18watt.com) did with the 18w Superlite. A plate fed tonestack (a la blackface Fender) in front of the 18 watt power section. I always think parallel preamps are wasted crunch for not much fatness, and for this reason I've always liked the Superlite TMB concept - though I haven't really built one yet (tried it for a few days, but always tinker). 1.8k with a 1uF sounds a lot like the Mesa values I've been considering for my 2204...

Sounds like a great use of a "wasted" triode in my book.

Do you still have 6V6's in this one? I might have to try a 6V6 Superlite TMB now... I have an extra OT, but not extra chassis PT or 6V6 pair...

Do you still have 6V6's in this one? I might have to try a 6V6 Superlite TMB now... I have an extra OT, but not extra chassis PT or 6V6 pair...

Yep - still 6V6's. And I'm debating adding a switch to drop the power tubes' cathode resistor, so that I can pop in a pair of EL34's. Kinda a stupid gamble, with such low plate voltages. And my PT filaments are only good for 4 amps - just enough for those power tubes and 2 12AX7's.

...But I'm curious, because I have a schemo of a Trainwreck Express variation where this guy pushed a pair of EL34's with only about 325VDC max B+. And I've heard of success stories like the Octane and similar, and just want to try it for grins. Once again, stupid since the amp sounds smokin' right now, but I just don't know when to stop, sometimes. I think I need an intervention.:shock:

But the 6V6's really sound good, IMO. Much more brash than I would have thought. Maybe I got lucky with my choice of them - bulletproof JJ's. Maybe the new Tung Sol's would have sucked - too "round and musical," as they are constantly described in comparison to the 6V6S.

I have a 2x6V6 amp that only puts about 250 plate-to-cathode volts on the tubes, and most of the preamp tubes have under 100 volts on the plates. Yet it still sounds good! :lol:

Design the amp right, and you can use voltages that are off the beaten path. :wink:

- Scott

Design the amp right, and you can use voltages that are off the beaten path.

Word. 'Nuff said!