Can you make an AC30 out of two tubes?

[Mike K]

Of course not... but... can you convincingly make a 5W AC30 top boost tone out of (1) 12AX7 and (1) EL84. I think I can. And the other rub - can you do it with a single tone control and volume?



I think you can. It's not perfect, but it emulates a decent TB tone control setting and a decent sweep of the cut control. It sounds good clean and driven, and definitely has the AC30 tone. It doesn't overdrive as hard, but it's decent. The clean tone is very nice, perhaps a little brittle, but as I recall that's about what it should sound like.


So why the hell would I do this? Long story, but someone asked me to build them an AC4 TV. I've never played one, I didn't have one I could listen to, all I had was that schematic and some stuff laying around that would work for the iron. It wasn't great stuff, and I told them that, but... So anyway I built to that schematic and they didn't like it. OK, I took it back. Honestly I didn't like it - I swapped the OT out, which made a big difference, but I still didn't like that circuit much. It's OK - it doesn't sound much like a VOX to me though. Sounds closer to a scaled down 18W Marshall, or perhaps a brighter Tweed Princeton. Whatever, it's not horrible, but it wasn't what I wanted.

I figured if I was going to be stuck with this stupid thing, I was going to figure something good out. I tried many, many, many different single knob tone control variations. All of them good at certain volumes, but all had issues at others. Hence the crutch of the single knob. Adding switches is cool, but I had this faceplate made up for the thing, and I didn't want to ruin it punching in more holes, so I was determined to find something (other than Tweed Fender) that was different, but good.


At first I was just testing and tuning, and I wasn't really getting there. I then decided I'd try a modeling approach. So I decided to do an AC30 full amp model and see how well I could match that without using a tone stack and keeping the gain as high as I possibly could, but still retaining a clean good clean tone (hard to do with just one preamp tube!). I had an iteration I played for quite a while that was pretty good, but it just didn't have quite enough gain and it didn't quite sound like an AC. The other day I took another stab at it and found a with a couple small tweaks I was able to get more gain and a better match to the preamp frequency response.

I've been playing it for the past couple days and I like it. I don't have a big AC anymore to compare to, but I listened to some clips and compared to my AC30 amplug (which is actually fairly convincing) and I think it's better than what Vox made... well it sounds more Voxy. Might have a bit more fuzz than I personally love at full tilt but I think that's typical for an AC with the treble cut off. It certainly has the chime down low and at mid-volume it sounds very, very close to clips I found online of current production AC30s.

I haven't updated a schematic yet, only what's in LTSpice. But perhaps I'll post it if anyone is interested. If you build it, LMK how convincing it is. You can of course do the spice model like I did and compare it to a big amp. You'll see where it's off (clearly in overall gain).

I've been playing it through my Weber CVA8 speaker and it does well. It's a dark speaker so I thought perhaps it might like this amp but oddly enough this is where it sounds the most fuzzed when pushed. Through a G12M, G12H and G12 Alnico Blue it sounds very composed and more crunchy when pushed. Also brighter through all 3. All sound good too, and all have a very different character. The Alnico Blue obviously sounds the most convincing though to a true top boost AC.

Today I gave it a test with a Ge Treble Booster and it was very nice. Pushes into decent distortion even with the volume set to max clean. Sounds a bit more Brian May when you really push the volume and tone of the amp and back off the booster level.

It was a fun, ongoing tinker and quite a challenge to see what one can get out of just two tubes and not be a complete one-trick pony.

[Mick Bailey]

I'd like to see the schematic - it sounds like an interesting amp to build, especially the tone stack. I've experimented for years with single-knob stacks, especially low-loss ones and never really settled on anything that was a keeper. Mainly the insertion loss was greater than I'd want for the topology you describe, but I could get the gain I wanted with the stack bypassed and always ended back with a 2-knob design.

[Mike K]

Well you might be underwhelmed by my tone control. It's just a cut. But the great thing about them is when you have a ton of boost up front, they actually bend the shape of the curve and introduce a bit of mid-scoop.


Anyway, the whole key was the capacitive network at the volume/voltage divider as well as the bypass cap on the second stage. Those two needed to be tuned perfect to get the right curve and slope from 1kHz on.

I set the controls at noon (treble) and 9 o'clock (bass) on the AC30 model, and it always stays that way. I knew that was an ideal point where I wouldn't have too much bass when pushed.... it could definitely have more at the lower volumes, but it's workable and most importantly, it retains the midrange and upper end chime that, to me, defines the Vox sound.


Actually, at very low clean volumes, the effect of the tone control seems minimal even though, based on simulation, it can do a lot. It just doesn't change the character of the amp a lot. It's mainly because we aren't creating a lot of harmonics there and most of the effect is on the upper mids and highs. It generally sounds best between 3 and noon though, and as you increase volume, you increase tone. I could probably tie these together on a gang pot and use another pot for a different tone shape - and I've thought of that as my next progression.

Every cap in there is providing some level of tone shaping though - none are for stability.


I compared this to a single tube design with the Vox tone stack between the second stage and the output tube, there's a significant increase in gain. And even if you throw a cathode follower in there, again, still large. I can look up the numbers. I've built them that way too but I don't have one to compare directly ATM, so I'm using sim numbers.

Also you might think... well... that thing doesn't have a lot of gain. It does actually. Even with my relatively low output strat it's starting to break up on the high end around 8 o'clock. EL84's with no NFB don't have a lot of headroom.

Of note - I'm using a 12k primary. That has little effect on the overall tone except it does reduce the headroom a bit compared to a 5k that Vox uses. What it gives is a much better driven tone and a crazy amount of touch sensitivity when clean (it effectively makes the gain of the EL84 higher). Driven you get a harder clip and more compression on the grid current limiting end, and on the cutoff end, a softer cutoff with a less abrupt inductive spike. The plate current fluctuates less, but the screen current more severe. This is why I've used 100u/100u. It also has zero hum with the config, even with no dead node. 100u is probably more than necessary for the screens, but I used a JJ can. I know these amps tend to hum a bit with only 47u at the first node and no dead node, so I think that's at least necessary.

[bill0287]

I'm interested in the LTSpice schematic if you don't mind sharing. I could reenter the who thing, but...

[Mike K]

Sure. Pm me an email, I don’t want to load it on the forum. I assume you have tube models? If not, lmk and I can attach those.

[Mike K]

That's the actual amp. Been modified a few times, so not my cleanest work.


I realized there's a mistake in that schematic above, but it's of no consequence really. The voltages shown are for a different PT, of which I have another, somewhat similar 12AX7 + EL84 amp. I use all the same values, the bias is not compromised but the Hammond 269GX on this amp has about 15V lower B+.

So far this thing is holding up for me. It's one of my favorite 2 tube builds both clean and dirty. I really like it cranked - much better and more versatile tone than say a champ or Princeton type. Not nearly as midrangey and very articulate on the high end.

[Chuck H]

Hey Mike,..

FWIW I built an LTSpice model of this amp (as shown in the schematic, so 15V higher than actual I guess). The performance looks very impressive for simulating the TB design of larger amps. The overall circuit behavior at clipping is similar, EQ is alright with the exception that the full TB circuit has a lot more mid cut. The asymmetry of the clipped wave form is similar to what the full TB circuit does also. Overall it looks pretty good even just simulated. Neat how you adjusted the EQ in both triode stages to get the overall tonality correct. That said...

What is sacrificed would be the big mid scoop typical of the TB design. I don't expect it's missed in a small amp like this though. In fact it would probably sound too thin if it were there. And you just can't have the mid scoop without extra stages for recovery or you'd lose too much gain. Even without hearing it I'm going to say 'nicely done'.

[Mike K]

Thanks Chuck!

Yeah - as I said, it's not perfect. I don't know if you noticed but it can get a mid scoop similar to the big amp at low tone and volume settings, but that tends to flatten out as you increase the tone and volume - I couldn't figure out any way to get that in without using notch filters or the like, and they suck too much gain. Anyway, even if it's not a *perfect* emulation of the big amp, it's a lot more pleasing to my ear than the AC4 TV I posted in the OP.

WTBS, I've built this, as I said both with a single preamp tube with the tone stack between the 12AX7 and the EL84 - if you're familiar with the AX84 P1 topology, that's what it is, but I've built all sorts of configurations. None of them all that great except one I built that used an IRF820 as a CF and a plexi tone stack and values. Also I believe on that I may have run the voltages in the 250's to keep the headroom down. The clean tone was not great though.

I've also built more faithful two tube preamps with the SE EL84. They work well enough but without a master volume you really can't keep them clean.

If I have some motivation, I'll make some clips.