Another AC15

[Ó Draighnáin]

I didn't want to jump in on CJs thread so I thought I'd start another. I just finished this amp for a friend. Still building the cabinet.

This amp has three different Vox-type channels. First, an EF86 channel with switchable coupling caps. The original EF86 AC15 had a switch that gave you either about 240pF or 10pF, I filled in another 4 values in between. Second, there is a "Normal" channel also with switchable coupling caps (6 position starting from 500pF), and third, the classic top boost channel.

I also kept the cut circuit (which will be across any input selected).

The layout is Hiwatt influenced, which I find seems to work well for these simplified Vox-type amps.

There is no vibrato and I did away with all the low inputs that the Vox amps have (guitarists rarely use them - I never did). I decided on a single input and added a three way selector switch between the input pentode and the two sides of the input triode.

 

[user 130681]

Always mount the transformers 90° out of phase to avoid induced hum. You won't be sorry.

 

[CJ]

That's awesome! Very neat wiring.

You have a schematic?

Yes the vib/trem circuit takes up a lot of real estate.

What kind of output? looks pretty beefy!

 

[Ó Draighnáin]

That's awesome! Very neat wiring.

You have a schematic?

Yes the vib/trem circuit takes up a lot of real estate.

What kind of output? looks pretty beefy!

Thanks CJ. I originally tried to get a new production Dagnall output with 4, 8 and 16 ohm taps but was sent the wrong one (one with only an 8 ohm tap) and I had to send it back (my friend is going to use this head with a 2 x 12 cabinet with two original silver alnicos from his 60s AC30 - 16 ohm). This took a long time to sort out (I am in NZ and availability and shipping are a real issue). I thought about it and decided to go with a 15W hi-fi type Hammond transformer (1650E). I purchased from one of the big suppliers and I new that it would arrive quickly. This surety of shipping and knowing it was in stock (a lot of businesses have products listed for sale that are not actually in stock) was also pretty persuasive for me. Finally I was fixated on having an 8000 ohm primary. This output transformer is an expensive option, but did ship for free (which is a big deal to NZ with a chunk of iron).

As I understand it, the Vox amps were never really built like some Marshalls, Fenders or even Valco/Supros with underspecced transformers, more like Hiwatts (which I believe used a hifi type output transformer). So although I could be wrong, I didn't think that output transformer saturation was really a big part of the Vox sound? There is the matter of the extended high frequency response with this transformer but the originals were made with pretty decent chunks of iron as well (I have a '64 AC30 for reference).

Building this amp forced me to make a calculation relating to the power transformer secondary HT impedance matching with the plates of the EZ81. The formulae are available in one of the EZ81 datasheets.

As far as a schematic goes, I only have hand drawn "mud maps". I know that a couple of the voltage dropping resistors were changed slightly. I used 8k2 (originally 10k) for the top boost and 18k (originally 22k) for the normal channel because of the lower B+ (because these are AC30 circuits adapted for AC15). The coupling cap switching is similar to what you might find in a Matchless amp (e.g. Nighthawk) or what I also think is in one of the 90s Korg era Vox amps.

I used 220k mixing resistors for the normal and top boost circuits into one half of the PI and the EF86 into the other half of the PI. The output is standard AC15. I cross-checked the top boost circuit I used against the GA-70 circuit.

This amp is quiet. Like I said, it follows the Hiwatt layouts (with the choke 90 degrees to the power transformer and the output as far away as possible).

 

[Whoops]

Always mount the transformers 90° out of phase to avoid induced hum. You won't be sorry.

+1

 

[Ó Draighnáin]

 

[andYz00m]

I wanna hear it! Sweet design and layout.

 

[dmp]

Beautiful build!
Distance helps a lot with magnetic coupling, as it follows the inverse square law. And if it sounds good it is good!
I'd like to build an amp like this. Curious why no one jumpers (or allows) the two channels together on Vox, EF86+top boost? Do they not sound good together?

 

[Whoops]

Looks great congrats

 

[CJ]

I jumpered across the .005/1 meg network of that filter and it is interesting to see what happened to the midrange. You can do this on c2, c3, c4 and c5 to see how the sound develops. The AC 10 runs the ef86 through this network instead of the ax7, I thing I will try that out.
Maybe a tone switch could be added to do this.

 

[Horizoneer]

Is the EF86 channel out of phase with the top boost, in the same way the "normal" channel is? I think that's why not many people do jumper them, though the resulting phase cancellation can indeed do some cool things in the mids.

Re; transformer orientation, they're probably far away enough that it doesn't really matter. The original AC15 & 30 L-shaped chassis had the PT and OT in the same orientation but at opposite ends of the flat shelf. I had real trouble in my recent 5e3 clone build, in that chassis they're right next to each other and now I know what the worst case is - the PT inducing hum in the OT secondary *even with no valves socketed* Luckily a different PT cured it.

And on the Vox transformers, yes - the OT is a reasonable lump of iron, over-spec compared to many fenders and marshalls of the day. In fact it's about the same size/weight as a Marshall 50 watt OT. HOWEVER... there's still the question of how it was wound, and the old Wodens, Haddons, Albions and Parmekos do have their own voicing. I read that in an OT a lot of the high frequency voice comes from direct coupling between the primaries and secondaries, whereas it's the low end extension that bumps into the limits of the iron stack. I just splurged on a '62 Woden and it's definitely got a different sound than the modern OT it replaced - thicker & darker in the top end even though it's actually a little bigger.

On the PT side, they were barely built to withstand the stresses they'd be put under. They use bigger iron laminations, but a much shorter stack of them than seems wise. The HT secondary was rated for ~280-0-280v at 160ma, these amps *idle* at 200ma and under load go up to 250, so there's major sag and heat build up - I've seen plots that show the PT getting up to 90 degrees C after an hour of use. They also used wire with higher resistance than modern options - I had a PT built to the 280-0-280 spec, but even then the amp didn't get down to the kind of voltages I wanted to see until I added 30 ohms to each side of the HT going to the rectifier - that dropped the plate voltages down to 320v.

@Ó Draighnáin , your build looks beautiful. What size coupling caps did you use either side of the phase inverter, and what voltages are you getting on your EL84s? Is that a choke I can see hiding behind the PT, and did you place it in the original AC15 position where it's smoothing the plate voltages as well, or the more typical way of putting it after the plates so it just does everything upstream?

 

[Ó Draighnáin]

Beautiful build!
Distance helps a lot with magnetic coupling, as it follows the inverse square law. And if it sounds good it is good!
I'd like to build an amp like this. Curious why no one jumpers (or allows) the two channels together on Vox, EF86+top boost? Do they not sound good together?

Thanks DMP.

From my perspective, my favourite Vox channel was the AC30 "Brilliant" high output from a treble voiced-Vox (as you probably know they were voiced differently). See attached photo, the bottom left input ( I had to wipe the grime off, poor old girl hasn't been used for years).

When playing live I never used any other channel and I also never jumpered. The gain was enough (definitely with a Les Paul type guitar. Add a tube screamer or a tonebender for solos and you are away (ok, and an echo and a wah). Even so, if you knew how to use your guitar volume knobs, and how to play the strings with both a light and heavy touch, it was all at your fingertips. Add the flick of a pickup switch and you are going from glistening clean to a creamy driven sound - all amp.




The EF86 channel is a different beast altogether, it is a high gain channel. I converted my '64 AC30 to an EF86-type but then went back. I just prefer that standard brilliant channel.

Frankly, I just never thought about EF86 and top boost. Part of the reason the top boost was added was because there was a demand for more gain and I'm thinking that the EF86 had that gain already?

 

[Ó Draighnáin]

@Ó Draighnáin , your build looks beautiful. What size coupling caps did you use either side of the phase inverter, and what voltages are you getting on your EL84s? Is that a choke I can see hiding behind the PT, and did you place it in the original AC15 position where it's smoothing the plate voltages as well, or the more typical way of putting it after the plates so it just does everything upstream?

Thanks Horizoneer for that very interesting information about the original PTs. I also found that I had to tame the HT as it was running too high (circuit indicated use of a 300-0-300 transformer, for future builds I would drop this back a little and take a close look at the secondary HT winding resistance of the PT used (if using a EZ81). I also ended up adding resistors to the rectifier plates (a mod after the posted photo of the chassis internals was taken). These resistors can get warm to touch.

You are right, there is a choke hiding behind the PT (10H, 155 ohms) and this amp was wired true to the original schematic. As for the coupling caps, after consideration, I used 10n in and 22n out as a starting point (there was always the possibility of tweaking this - I'll wait for the guitarist to tell me what he thinks). The plates are still running a bit high assuming the schematic is not indicating anode to cathode voltage (plates = 320V or 310V anode to cathode). EL84 Cathodes = 11.5V.

Out of interest I have added a picture of the OT in my AC30. Looks like an Albion from the number.

 

[CJ]

Yes the AC 30 was the only amp that could catch on fire all by itself. They added vents but still hot.

300 to 320 seems to be a good voltage for the 84's, they run hot with cathode bias and the smaller surface area. That Elpico used 350 volts , you can do that but you might want to buy Tesla tubes. I hate burning up nos amperex every 6 months

Can you see the wires on the top of that output as to determine the coil structure?

 

[Ó Draighnáin]

Yes the AC 30 was the only amp that could catch on fire all by itself. They added vents but still hot.

300 to 320 seems to be a good voltage for the 84's, they run hot with cathode bias and the smaller surface area. That Elpico used 350 volts , you can do that but you might want to buy Tesla tubes. I hate burning up nos amperex every 6 months

Can you see the wires on the top of that output as to determine the coil structure?

I'll open her up and take a look.

I love the smell of burning AC30.

 

[Ó Draighnáin]

Getting the end bell off requires dismantling the amp a bit (to get at one out of the four bolts). I'm not going to do that until I clean up my workspace, which is a disgrace - even by my very lax standards.

Here are a couple of photos of what can see by just pulling the chassis out.

 

[CJ]

No need to rip it apart, looks like three sec and two pri, measure dcr. If CT to plates is equal, then it is 4 sec , if unequal then it is 2 sec.

Thanks for the pic. Looks like a nice XFMR.

 

[GussyLoveridge]

Really nice work. The wiring is beautiful.