2021 Monkeymatic Black Butte SMPS version!

[xtian]

I finished wiring, powered up, and am having trouble. My SMPS is not working well. If I plug in only the 12AX7, it's OK. HT is 300v. You can see my multimeter in the photo, set up to measure current across the HT-to-ground, and I see ~0.8mA. Both triodes amplifying, and I can see a nice, clean sine wave at output. Total draw from the wall shows 6-7 watts.

When I plug in an EL84, during heater warm up, it's OK, but when it starts conducting, HT sags quickly to 160v and falling, draw from the wall rises to ~30 watts, and HT current rises to ~25mA. (This SMPS is rated for 60mA on the HT.) And the heat-sink'd voltage regulator on the SMPS gets very hot--I shut it off after 250°F.

I think my EL84 is wired correctly.

Any ideas?

Screen Shot 2021-05-21 at 4.45.49 PM.png

IMG_0609.jpg

IMG_0610.jpg

[pompeiisneaks]

I think those circuit boards are designed for 12AX7 heaters, and the EL84's pinout doesn't have the heaters on pin 9, it has them on pin 4 and 5, so that could be what's going wrong? i.e it's jumpering the pins 4/5 for 12AX7 use and that's the start and end for EL84.

~Phil

[pompeiisneaks]

Oh it looks like you did something to remove the tracks underneath the wires, so nvm I guess it's possibly something else...

~Phil

[pompeiisneaks]

No, looking at it agian, I think that could be the problem, what I thought was a trace lifted is on the side I can't see. To use these for an EL84 you'd need to cut like on this, or the heaters are jumpered in the wrong way:

CutTrace.png

~Phil

[xtian]

Thanks, Phil, yes, I cut those traces, and am applying 6.3vDC to pins 4 and 5 on the EL84. And this tube is working--just before it starts conducting, heaters are all good. In fact, heater voltage never sags at all, even when HT drops over 50%!

[pompeiisneaks]

I kinda figured you had thought of it, but it can't hurt to double check. It might be some flaw with the SMPS where it can't put out the current it's supposed to, due to too much heat, try pushing some air over it and see if cooling helps?

~Phil

[xtian]

I have confirmed that my wiring is correct, because the amp passes signal as is.

Next, removed both tubes (thus unloading the 6.3v supply), removed all wires from the HT supply terminals, and instead used a 12K dummy resistor as a dummy load. 300v = 25mA x 12K. Nope, HT sags to 200v almost immediately, and continues to sag lower and lower.

F*ck this SMPS.

[xtian]

I swapped in a different SMPS, a 60w module, and this one works like it should. No voltage sag at all. Here are my readings, heated up and running at idle:

177vAC wall voltage, 27 watts
B+ 303v
screen 279v
cathode 8.9v 178ohms = 50mA (too hot; 100% is 41mA)
with sine wave and 8R load, I get 4.5 watts clean

I've got a nasty oscillation at higher volumes. We'll see if cooling the bias solves that.

Also, chassis has 50vAC to earth, so I obviously need to earth the chassis.

IMG_0611.jpg

[pompeiisneaks]

Very cool, how's it sound compared to old linear power supplies?

~Phil

[xtian]

I tamed the bias with 220R cathode; now running at 44mA (just over 100%). I can't really evaluate this amp yet; I need to solve the oscillation first. It's high pitched (edge of my hearing), happens only when Volume is opened up, and the Volume pot sounds exactly like there's DC on it, but there's not. Schematic attached--where would you look for oscillation first?

Monkeymatic-Black-Butte-Desktop-#2-schematic.jpg

[xtian]

I'm inclined to add a grid stopper to the EL84 first. I see values ranging from 100R to 1K5...

[xtian]

I put in 100R screen and 1K5 grid resistors, and...no change. Still have scratchy Volume pot (not dirty, though), and high-pitched oscillation when turned up moderately. Guess I start snubbing next?

[xtian]

The oscillation appears on scope at 47KHz, and is present at point circled in yellow. But if I remove power tube, no oscillation.

Does V1b (pin 7) require a grid stopper?

Monkeymatic-Black-Butte-Desktop-#2-schematic.jpg

[didit]

Hmmm --

Speculation? EL84 screen dumping a loose signal into its shared B+ with V1b.

Another speculation? That particular SMPS has an internal clock that feeds your high strung control grid just enough resonance noise to provide lift off.

Sorry to hear those teeny SMPS didn't cut it. I grabbed a few. Plans are for preamp only, into beefy class-D. However, had also hoped maybe other applications requiring more current such as tube reverb would work out. Someday somehow someone will get the energy up to design something cheap & satisfactory in this space.

Best .. Ian

[xtian]

Thanks, Ian--I think the original SMPS would be fine for several 12AX7s.

The 60w one I'm using now is working pretty well! Noise is extremely low--if I put my ear right to the dust cap, I can hear a little rhythmic chatter, but no hum at all.

I squashed about 80% of the oscillation by changing the 500p silver mica in the tone circuit to a 470p ceramic. Silver micas have caused me a lot of problems over the years! I also added a 22K grid stop to V1b, and rerouted the grid wire.

The remaining oscillation is low in amplitude, and doesn't respond to a snubber cap clipped across the V1 plate resistors. Fortunately, the amp sounds very good right now! Just like all the other Black Butte's I've built from those Akai M8 units. Great distortion. And the built-in L-Pad is a nice feature.

Phil! I'm calling this a success!

For future reference, I'm sure I can reduce filtering, because there's no AC ripple to worry about.

IMG_0614.jpg

IMG_0615.jpg

IMG_0616.jpg

IMG_0617.jpg