The Black N Tweed

[cometazzi]

I've been a bit reluctant to post about this because I kinda went out on a limb on my own. However, I've been working on this off and on for about three months now, doing research and reading and I think I'm on the right track. Today I finally got to the "put on some pants and solder it up" stage. It's about 45% done.

I'm fascinated with Micro amps, and I've thus far only built one- The Heavy Watter (built last summer). It was a lot of fun to build, it's a real gas to play and (most importantly) the power level is perfect for apartment living. It's something I can actually use a lot. I'd wanted to build another Micro in a different style, but couldn't decide between a Blackface Champ, Vibro Champ or Tweed Princeton. Zoning out during a boring lecture in an online class, I was struck with the question of "¿por qué no los dos?"

Ugly chicken scratch, I know. I hope this is the last hand-drawn schematic I post here. The design has changed slightly since I drew this up in May but here's the main idea: A two channel micro amp- one channel is the pre-amp of a Blackface Champ, the other is the pre-amp of a Tweed Princeton. Originally I had all kinds of stupid mods and features involved, but in the interest of KISS I started stripping them away. A few things do remain though:

Tweed Channel has a pull pot (Volume) to engage or disengage the cathode cap on V1A, a pull pot (Tone) to lift the tone stack, and a mini SPST to cut the NFB. In its 'resting state' (no tricks enganged) it is identical to a 5F2-A preamp, but combinations of these three options can make the circuit smell lots like a 5F1 Champ or the Bright Channel of a 5E3. Maybe a few other Tweed-era preamps.

Black Channel is simpler, with only a tonestack lift and SPST to cut NFB. Being the AA764 preamp though, it should smell a bit like a lot of 1960s Blackface examples.

The "FX" you see above in both preamps was a passive FX Loop that has been omitted. Obviously the power section is different so there will be differences. Power section is 12AU7, Fender Reverb OT. Power supply is Weber WRVBPT, diode rectification, and big ole filter caps. Bleeder Resistor added for my own feelgoods.

"You said you wanted KISS, but you built a two-channel amp with a bunch of 'tricks'. That's not simple at all. Are you stupid?"

Nobody ever said I wasn't, but you should have seen the list of stuff I removed. My new take on "mods and tricks" is to have it so that all of them can be turned off and when they are, whatever circuit I've co-opted is in its original, stock form. Tried and true. Also, as I see it (and I'll listen to counterarguments), this circuit is essentially four sections that can be separated out and diagnosed individually: Tweed preamp, Black preamp, power amp, power supply.

Anyways, that was the plan I devised and the path I'm trodding. I welcome any comments or criticisms.


EDIT: Made a correction to the schematic- B and C taps were associated with the wrong preamps. Fixed now!

 

[Phrygian77]

You might have too much voltage gain from the tweed channel since the second stage is bypassed.

 

[Lowerleftcoast]

Hey... why stay in Fenderland? Grab a Gibson circuit for one side.

 

[Mongo Park]

Ga5 is a little ways off the path

 

[sds1]

The 470kΩ channel mixing resistors are followed by 220kΩ grid leak, these form a voltage divider where 2/3 of your signal will be sent to ground prior to the power amp stage. Was this intentional?

 

[cometazzi]

Thanks for all the replies, everyone.

You might have too much voltage gain from the tweed channel since the second stage is bypassed.

The cathode bypass cap on the second stage is true to the 5F2-A design, but do you mean too much voltage gain for the 12AU7's input (instead of a 6V6)?

Hey... why stay in Fenderland? Grab a Gibson circuit for one side.

It's not a bad idea. To some extent this is a sort of 'proof of concept'. If it all works out I might consider another two-channel amp down the road that has a more diverse pairing. In this case, the Tweed and Blackface circuits are still pretty different from each other, as far as I can tell. I will admit that I've never played a Tweed Princeton, nor even seen one in real life (clone or otherwise). I've only heard Youtube videos and builds from this forum.

Ga5 is a little ways off the path

True. Very 5F1-Champ-Like, but angrier. I don't know, but I expect the Tweed channel in this configuration to be a bit on the wooly side at most any level. I bet the GA-5 would be even more so.

The 470kΩ channel mixing resistors are followed by 220kΩ grid leak, these form a voltage divider where 2/3 of your signal will be sent to ground prior to the power amp stage. Was this intentional?

Yes, actually. A lot of the Micro amps tend to dump about 1/2 the signal just before the power tube. In Rob's words "to prevent overwhelming the little 12AU7". Mayhaps my math was wrong but I calculated it to about 53% reduction, and it's actually what combo Rob himself used on the Champ Micro, along with a 1M MV. (5F1 Champ being the closest thing to a Tweed Princeton, most things considered).

That said, this was one of the items I was going to 'twiddle' once it was up and running. One of the things to note is the Heavy Watter simply has a 1M Master Volume control as a grid leak resistor. The Black N Tweed doesn't have a Master Volume, so I may actually 'tune' this divider so that I can hit whatever sweet spot at the optimal SPL output (without causing a noise complaint). It might be perfect. Won't know till I try!

 

[mountainhick]

Curious about the B and C preamp voltage supplies. You have the lower option going to the BF? The schematics I have show 320 supply to BF, and 260V to 5E2 preamps

 

[bebopbrain]

The two feedback paths will, to an extent, fight. This might not be a bad thing.

So lets say tweed channel is unplugged with the volume on zero and black channel is cranking. If you don't open the pull switch then negative feedback is applied to the tweed channel and it will have (out of phase) output.

Maybe just apply the feedback at the phase inverter when signals have been summed?

 

[cometazzi]

Curious about the B and C preamp voltage supplies. You have the lower option going to the BF? The schematics I have show 320 supply to BF, and 260V to 5E2 preamps

Oh jeez. I have those correct on layout I drew up but forgot to update the schematic. Schematic above is wrong, and I've actually got B going to the Blackface and C going to the Tweed. I'm targeting 200-205V on the plates on the BF, and about 165V on the plates for the Tweed (give or take a few volts).

Thanks for pointing that out. I'll correct the schematic later today, perhaps.

The two feedback paths will, to an extent, fight. This might not be a bad thing.

So lets say tweed channel is unplugged with the volume on zero and black channel is cranking. If you don't open the pull switch then negative feedback is applied to the tweed channel and it will have (out of phase) output.

Maybe just apply the feedback at the phase inverter when signals have been summed?

This is secretly why both feedback circuits are on their own SPST ;-)

I had a feeling they would cause the channels to interact. I set it up so I could cut the feedback to the one not being used. It's true that it may not be a bad thing if they do, so I was going to experiment and see.

What will be real interesting is what will happen if I jumper the channels together.

 

[cometazzi]

Curious about the B and C preamp voltage supplies. You have the lower option going to the BF? The schematics I have show 320 supply to BF, and 260V to 5E2 preamps

Corrected now. Thanks again!

 

[Phrygian77]

There is no cathode

The cathode bypass cap on the second stage is true to the 5F2-A design, but do you mean too much voltage gain for the 12AU7's input (instead of a 6V6)?

There's no bypassing on V1B in the 5F2A. There's no tone stack loss like there is on the AA764.

You could also have used the other gain stages to do a Dumble style overdrive circuit.

 

[cometazzi]

There is no cathode


There's no bypassing on V1B in the 5F2A. There's no tone stack loss like there is on the AA764.

You could also have used the other gain stages to do a Dumble style overdrive circuit.

You're right. I posted a link but apparently my comprehension is off today (and when I first plagiarized the 5F2-A schematic). Fortunately that's an easy fix. Thanks for pointing that out!

I'm actually not so familiar with the Dumble amp circuits. Weren't they were all mostly made to order and had lots of variation?

I'm still in Fender 101, apparently.

 

[Phrygian77]

I'm actually not so familiar with the Dumble amp circuits. Weren't they were all mostly made to order and had lots of variation?

I'm still in Fender 101, apparently.

I had mostly ignored them until recently. I would say they are basically just his own evolution of the blackface circuits.

 

[sds1]

Weren't they were all mostly made to order and had lots of variation?

Some variation but not like Fender world. Little tweaks here and there within the same few base circuits, some variations more consequential than others.

 

[cometazzi]

As an update, I finished the assembly last week. Since you folks kindly pointed out a few obvious errors I had made, I went back and traced everything and checked it against the schematic. Twice, on two separate days. Fired it up on Saturday with a local Tube Doctor friend and it works!

Guess which channel is Blackface and which channel is Tweed ;-)

I originally wired up the OT based on a diagram I found on the Internet but it was wrong, so the NFB phase was wrong. I swapped the OT secondaries on the jack and all was well. If it weren't for that, this would be my #2 in a row of "worked perfectly the first time".

It sounds great! Each channel has its own personality, and both personalities are great. The BF channel is not quite as loud as the Tweed lest you defeat the tonestack, but that's probably to be expected. I might beef up the plate resistors from 100K to 200K (one at a time) to see if that will even them out.

Otherwise, with the Tweed channel on 5 or the BF channel on 7-8 they're about as loud as someone playing a big ole dreadnaught acoustic guitar and singing. Going full up on the Tweed channel would drown them out. Power was measured at 2.6W, give or take.

Overall, I'm pretty pleased. The only thing I wish it had was the Reverb that I scrapped from the previous design iteration.


On top of that, I'd like to give a thanks and shout out to our own local @Mongo Park, who bent this chassis on his home made metal brake. Last fall he sent me three of them (to my specs) for a price I couldn't refuse. He also sent some IEC sockets and vintage tubes, the latter of which are a pretty hot commodity these days. What a guy!

 

[woodandglue]

This looks great! Can we get a sound clip, and or a gut shot? Are ya building a cab or stickin with a head?

 

[cometazzi]

This looks great! Can we get a sound clip, and or a gut shot? Are ya building a cab or stickin with a head?

Thanks! But "looking great" from the top is why I didn't do a gut shot. Mainly, I feel it looks about like this:



But with some Instagram filters it's more like this:

I did my best to fly HV signal or noise wires above other wires, and where any other wires crossed I made a point to have them do it at 90 degrees. Still, I think I could have done a lot better. There are a lot of things about the layout and arrangement that were teaching moments for sure. Everything worked, but there were many instances where I said "that could have been done a LOT better...". Namely how far away the preamp tubes are from their parts on the board, placement of the OT and 12AU7, arrangement of the speaker out and line-out, etc.

Lots of stuff on the board itself could have been done better. Next time I use one of these boards I'm going to try staking a turret or eyelet in the center row in some spots. That could improve things and make the circuit more compact. All of the stuff above just barely fit on a 28-row board.

Anyway, it all worked out and this thing is _dead_quiet_ which is amazing.

 

[woodandglue]

if it's dead quiet, there's nothing anyone should say about your wiring. I like that you had a cool idea, and went for it. No one is making this amp so you had to. Well done!

 

[cometazzi]

if it's dead quiet, there's nothing anyone should say about your wiring. I like that you had a cool idea, and went for it. No one is making this amp so you had to. Well done!

Yeah, I'm totally surprised! I expected it to be a buzzy mess of hum and squeal. none of that! Thanks for the kind words.

Today I swapped the 100K plate resistors on the BF circuit for 220K resistors, as that channel was a little anemic compared to the Tweed channel. They're a lot closer now, and it also improved the tone by a longshot. It's lots brighter and clearer sounding. When I used to dime the volume it would distort, but sound kinda ratty and jagged. Now it gets more hair, and it's smoother and warmer sounding. I like this channel lots more now!

Turns out the NFB switching with the ability to cut each channel is sort of a no-go. They're too interactive, and both channels sound way better with the NFB cut- brighter and louder. I'm going to remove the NFB circuit from the amp, and now I've got to determine what to do with the holes the switches are in. I could stick a bolt in them. I could put a bright switch on each side. I could add a diode clipping circuit, but that's probably illegal. I could also drill them out for Dwell and Reverb knobs and see if I can shoehorn reverb into the circuit.

Much to consider!

 

[Mexitele Blues]

For the switch holes, I would use a high cut at the power amp, maybe a small cap in parallel with the grid leak.

For the other cascading the channels would be cool.