the v in valvecaster

[duck_arse]

this week's unfinished project on the breadboard is a valvecaster.



I have some questions.

I'm using a 6cw7 (ecc84) - vhf twin triode, mu=24, close enough to the 12au7 for jazz. does a vhf valve have any undesirable characteristics in a circuit designed for an af valve? will any effects be audible and can they be corrected?

I have said valve in the breadboard w/ 7806 reg for the heater, and a cmos inverter charge-pump running at 52kHz producing around 16V at the moment. the 16V shows no switching noise, and less than 50mV ripple at the input frequency, but the output shows 150mV p-p 52kHz w/ 0V input. as the input increases, the output gets very "furry", with the switching noise apparent on the output waveform.

I get about 10V output, and obviously can't hear the noise, but it can't be good for anything attached, surely. do I need to juggle the layout, shield the valve, move or dump the charge-pump, or get an af replacement?

it seems the "sticking-out-valve-distortion" is a rite of passage pedal around these parts, and I have some suitably outlandish ideas for a wood, perspex and aluminium case, but I won't get any further until I can clean up the signal some.

any ideas?

[PRR]

> does a vhf valve have any undesirable characteristics in a circuit designed for an af valve?

Costs more (or did, long ago).

Might be more prone to supersonic oscillation (however the 6CW7 is an exceptionally tame/lame tuner tube).

In fact I'm suspecting that 6CW7 "is" a 12AU7 with small changes.

Put your switcher in one box, put the audio-stuff in another box (breadboards always leak bad).

Put 0.1uFd ceramic from B+ to ground.

Put 470pFd from each plate to cathode, also across the output jack.

> as the input increases, the output gets very "furry"

Then the tubes are mis-biased enough to distort. Yes, it is supposed to distort; but usually not so gain rises with signal. Get the DC bias points sane before you fret the details. There may not be a sane op-point for this tube and load and voltage, but the plates should not be slammed to ground or to B+/

[Jdansti]

You've probably already done so, but check the tube wiring.  The 12AU7 and ECC84 have different pinouts.

12AU7


ECC84

[duck_arse]

ahhh, you know how the ground lead on a cro probe pisses you off because of how it twists and gets in the way all the time? so you get some stranded wire and connect it to the breadboard and the cro front-panel earth, and everything is tickety boo. it's only audio work anyway.

so then you play with a valve amp with a switching circuit, and all sorts of hash appears. after re-layouting everything, and grounding/ bypassing all the breadboard tracks between valve pins, and bothering the people at diystompboxes, you think could it be not really there at all? and you clip the earth lead, and it just about all disappears.

so. oops.

thanks for hints, prr. mine cost about 50c each, clearing. and the audio valves have a slightly longer plate assy. and by furry, I meant the output signal had the 150mV superimposed, so the cro trace looked "overbright" and furry until you looked real close. but not now.

[Jdansti]

Glad you got it worked out!!!