How to set AC load for triode testing

[acheld]

So, I'm troubleshooting a build (Dumble 124) from Ampgarage in which I am seeing a bizarre waveform from V1/V2.   I'm guessing that I've misinterpreted the schematic or wired it up wrong, having ruled out bad valves . . .

My plan is to isolate the triodes so that V1 is working as expected, and then start adding the tone stack back into the circuit one step at a time (and it will be no easy task given the complexity of the switching!).

The question is:   how should I set an AC load for V1 for this test circuit?  I'm not trying to simulate the Dumble tone stack at this point -- all I want to do is get the triode to pass the input signal as expected.

So, for the mini schematic I've enclosed, what would a reasonable choice for the R (AC load) be?   I've seen some schematics where Merlin uses 39k, but ?   I know I'm over thinking this  but would like to hear what others might choose as a starting point for a Fender style amp.

[passaloutre]

I’d use a value from the RCA charts

http://www.tubebooks.org/tubedata/hb-3/Receiving_Tubes_Part_1/Resistance-Coupled_Amplifiers.PDF

Looks like 100k or 220k would do fine.

[PRR]

"Triangles" makes me wonder if you just have too much signal.

[acheld]

First off, Passaloutre, thanks for the reference!  Just what I was looking for -- had totally forgotten that diagram and those charts in the Handbook  . . .

As for triangles, nothing so simple as that.     Actually looks like an EKG tracing from my former life.   If I can get to it today I'll post a PDF.

And with nothing plugged in.   That said, for this amp, there is a FET input stage that precedes the "normal" input -- and needs to be taken completely out of circuit as a part of my trouble shooting sequence (I've already snipped the power to it, but I do need to really take it out . . .). 

[WimWalther]

You can pick a load based upon a typical value for the next-stage grid resistor.

Anywhere from 100K to 500K should be just fine, being driven by a 12AX7. The lower the value, the higher the LF will begin to roll off, of course, but you shouldn't see any crazy waveforms.. just a nice replica of the input (sine is suggested).

[shooter]

there is a FET

^^^ these are good candidates for crazy waveforms when they go crazy

[acheld]

OK, this is confessional, so just between us -- right?   

Thought it would indeed be my FET board, but nope.  60 cycle EKG looking waveform (about 5 millivolts P-P) remained on the input prior to the grid stopper after it was removed.  Hmm.

Here's the mistake:  I had not completed the power board for the relays, so just took an off the shelf 12VDC power supply and temporarily used it (though with grounding, and maybe that was the problem).  After decommissioning that kludge, the amp is now behaving as expected. 

Not quite sure what happened.  Now I just need to build the DC power supply for the relays . . . I'll probably just use one of Doug's boards.

[WimWalther]

Good to hear that you got it sorted, and high marks for making no bones about your error.

So can you tell us more about the "OTS DC supply"? Is it just a wall-wart type? Some of those things have very minimal filtering, expecting that role to be handled by the equipment.

[acheld]

Here you go:    Meanwell RS-15-12.   Looks like a decent design, though I haven't traced the circuit (too damn small with or w/o my glasses).   

When it is running with a small load (about 50mA), there is some switching noise. Otherwise, this unit has been really easy to use on projects.

[shooter]

a switch mode PS will look like a cardiac trace. 
the 300vdc 100A ones we used for work would probably whack a pace maker


I wouldn't recommend one near a sensitive input signal without adequate EMF shielding.