Monkeymatic S'express 15-watt

[martin manning]

Reducing load impedance means more current not more voltage. With 400V on the plates it's a pretty good bet that you'll be exceeding the plate dissipation limit on the output tubes, provided the PT can keep up. If it can, I don't think the output power will be reduced much if at all.

[xtian]

So do you think I'm inviting disaster with my OT choice? I'm using a 1974 Schumacher 022913. If you had to place a bet, do you think it will burn up under full load?

EDIT: I have a 25w Deluxe OT, 6k6 primary. Probably a safer option, huh?


And I'm surprised my B+ is so high! I measured 310vac unloaded (no CT). I implemented a FWB rectifier, and multiplied by a factor of 0.9 for "average DC" according to that Hammond data sheet. Guess I should have multiplied by the 1.41 factor, giving me 437vdc. And then accounting for sag under load, I'm actually seeing 405vdc, which is about 93% of unloaded.

[martin manning]

EDIT: I have a 25w Deluxe OT, 6k6 primary. Probably a safer option, huh?

I like that a lot better.

[xtian]

Added Mr. Manning's Presonance control. VERY nice. Also added 10K grid stop to input (was hearing a little cell phone chirping, though it was several feet away!).

[dcribbs1412]

Congrats xtian
Great looking build
Cool idea for the caps,
interesting about your plate voltage.
Any voltage readings on the preamp tubes?
again thanks for the info and layout and
inspiring me to try this circuit

Darin

[xtian]

I installed the 25w OT, just to be safe. It reduced the white noise, and increased headroom a bit. Still sounds great.

However, in scoping the output power (seeing 25watts into 8ohms clean, biased at 50%!), I see a 100KHz at 1.6v p-p oscillation when the PPIMV is between 3 and 4 o'clock, only when the Presonance (Martin's circuit) is between 11 o'clock and max.

This is a problem I need to solve, because while this oscillation is happening, my power tube current jumps from 22mA to 35mA each tube.

I tried pushing the NFB wire around, but can't see any change.

Photo attached: the Presonance control is at right (don't be confused by the bias circuit on the turret board below--it is not connected to the components on the Presonance pot!). My NFB resistor is now 68K instead of 100K. I implemented this per Colossal's drawing, but now I'm looking at Martin's v.3 version, and I see the 0.1 cap is split into two 0.05 caps. Maybe I need to do that.

[xtian]

Well, crap. I rewired the Presonance control, even nicer this time, I thought, but the oscillation is even worse. When I lift the NFB, oscillation vanishes.

[Tillydog]

I tried pushing the NFB wire around, but can't see any change.

I'd still put money on it being the problem - its proximity to the wires from the MV wipers to the output tubes.

FWIW, I would re-route the wires from the MV wipers so they ran down and back the way the wires from the PI come to the MV, then run them under the board to the EL84s, well away from the NFB wire (you could run that against the chassis, too).

If that doesn't help, you may also want to see whether giving the presence circuit its own, direct ground helps at all. (Rather than it sharing with preamp grounds as appears to tbe the case from the photo, although if you're not seeing any oscillation on the output of the PI, then it's probably not that.)

If none of that helps, you might have to slug the EL84 grids with capacitors to stabilise the feedback loop.

Oh, and double check that you're not chasing fairies - you can get oscillation caused by hooking scope leads into the circuit - is there still a problem when you're not looking at it with the scope? (I guess you can still check the current draw?)

£0.02.

[xtian]

Good eye about the ground wire. But in the second photo, you can see I've rewired the ground; now it's grounded along with the reservoir filter.

My scope is hooked to the output jack, along with the 8R dummy load.

I also ran a new NFB wire with more slack, allowing me to move it far from any other wires, including the MV wires. No change in sine wave on scope. Also, the wire from the PI circuit to the presonance control--pushed it around; no change.

How does one snub 20KHz+ frequencies on the PI with caps, as you suggest?

BTW, your £0.02 is worth more than my US$0.02.

[martin manning]

Presonance is tricky. There is a lot of phase shift on the resonance side. In my Express it would break into an audible oscillation at the extreme setting, and even though it sounded good and worked over most of the range I gave up on it.

[Tillydog]

I also ran a new NFB wire with more slack, allowing me to move it far from any other wires, including the MV wires. No change in sine wave on scope. Also, the wire from the PI circuit to the presonance control--pushed it around; no change.

Hmm... (I take it the green wire in the last photo is now the ground, and the new NFB wire is the black one?) you could try shielding either the NFB wire, or the wires from the MV - If the problem only happens at partial settings of the MV, I still suspect that the wires to the wipers on the MV pots are picking something up - likely from the NFB wire, but any OT wires would be possibilities (can't see anything suspicious, other than that NFB wire).

How does one snub 20KHz+ frequencies on the PI with caps, as you suggest?

Have a look for discussions about 'slugging the dominant pole':

You could just try a very small cap across the PI outputs (47pF, 100pF, etc)

If it turns out that it's some instability in the NFB loop (maybe exacerbated by the presonance control?) I believe that putting capacitors between grid and cathode of the EL84s should help this. Something around a few nF direct on the socket. It goes against the grain, but I have tried this, and even relatively large capacitor values are pretty transparent (little or no Miller effect in pentodes). With a 2n2 capacitor and your 1k5 grid stoppers, the -3dB frequency should be ~50kHz.

FWIW, I would carry on working on that NFB wire for a bit longer first.

(LOL @ $0.02 vs £0.02 )

[xtian]

Well, kind sirs, thank you for joining me on this snipe hunt. So sorry--I wired the damn thing incorrectly. Now corrected, no more oscillations, even with NFB wire bundled with MV wires. All good.

Let it be know that I use Mr. Manning's Presonance control, v.3, and I approve this message.

[Tillydog]

Glad to hear you got it sorted!

What did you done do?

[martin manning]

Woo hoo! Nice compact arrangement hanging the Presonance parts on the pot, too!

[xtian]

@Tillydog, the orange and brown caps, I had them wired to the wrong legs of the pot. You can compare the before and after photos. The black wire is ground. Now I've got the pot wired the right way round, so CW is brighter (presence) and CCW is darker (resonance).

Thanks, Martin!