example of bjfet switching on cathode?

[Mickey_C]

I've been looking for any schematics that illustrate using a jfet or bjfet for controlled switching on cathode bypass capacitors. Has anybody seen this, or anything close?

Thanks!

[sportster4eva]

You might want to check out some of the Boss effect schematics. They all use fet switching. Momentary button latching on and off. I believe it's called a bistable latch. Here's a pretty good article on relays and fets used as switches.


http://zone.ni.com/devzone/cda/tut/p/id/2774

[Mickey_C]

Surprised nobody has done this on a tube amp! It's hard for me to believe this is an original thought (I'm too damn old for original thoughts).

Imagine if there were a cathode cap, that got its ground path through the bjfet, you could do this with cathode resistors as well (well, if it would work, you could do it). Individual devices for each element. You could add them or subtract them to the circuit that way.

The capability of some of these jfets & bjfets, gets up to 30 volts, and I've seen some with some excellent dissipation over a watt, so why not? Anybody know why?

If it's true they would work, it would then be easy to control these remotely, like from an MCU, reconfiguring gain stages of an amplifier. Combined with relays or shunts, you could mimic just about anything. With an MCU you could control it all via MIDI. A MIDI programmable tube amplifier.

Just thinking out loud here. I'd really like to find if anybody's done this in a schematic before.

[Mickey_C]

Here's a simplified illustration of what I'm trying to do:

[klingo]

blues jr, cathode/fat switch, the hot-rod too...

[klingo]

blues jr, cathode/fat switch, the hot-rod too...

[mhuss]

I think it would be better to leave that largest value of cathode resistance in place at all times, and just parallel (much larger than normal) resistors to get the various values you want. As drawn, if all transistors are open (for even a microsecond) the cathode voltage will fly up towards the plate voltage and (a) produce a hellacious pop and (b) require that all your switch transistors can withstand 250-300v VCE.

I think the reason most folks use FETs is that they behave more like resistors both when conducting and non-conducting. The "shut off" BJT will have a much lower impedance then a "shut off' FET -- not really a problem in a cathode circuit, but it may be in a signal switching situation.

--mark h

[David Root]

I made a "bass dump" switch on my first stage cathode on my latest build, with a 5V DC SPST relay. This switching doesn't need a relay, or a jfet for that matter, just a mechanical footswitch, but I want to see how reliable they are (using a cheap miniature telecom relay made in china).

It's a fat switch, basically, operated in reverse as a bass dump. A .15 uF cap is always in parallel with a 1k5 cathode resistor. A 22 uF cap with a 100K tail resistor to ground is also in parallel, and switching the 100K resistor in and out accomplishes the bass dump. 22 uF direct to ground, fat or normal Fender tone. 22uF thru 100k to ground, the .15 uF takes over creating a 700 Hz step. I probably should have the .15 uF .1, and/or increased the 100k resistor a bit, as it is still a bit too bassy when the boost footswitch defeats the tonestack.

There is a whole chapter on switching in TUT1, I highly recommend it, it covers relays, jfets, mosfets etc in considerable detail and complexity of switching options.

[Mickey_C]

I think it would be better to leave that largest value of cathode resistance in place at all times, and just parallel (much larger than normal) resistors to get the various values you want. As drawn, if all transistors are open (for even a microsecond) the cathode voltage will fly up towards the plate voltage and (a) produce a hellacious pop and (b) require that all your switch transistors can withstand 250-300v VCE.

I think the reason most folks use FETs is that they behave more like resistors both when conducting and non-conducting. The "shut off" BJT will have a much lower impedance then a "shut off' FET -- not really a problem in a cathode circuit, but it may be in a signal switching situation.

--mark h

I agree... and my intention is to have a base default value of a largish resistor and smallish cap always in place. I just wanted to make sure that what I was speaking about was viable to begin with. A picture says a thousand words, as they say.

So, you're suggesting an FET would be better suited? Any examples or references to illustrate this adaptation would be greatly appreciated!

[Mickey_C]

blues jr, cathode/fat switch, the hot-rod too...

Excellent tip - thanks, I am off searching for a schematic. I'm doing my fat switch right now with a pull switch, and wanted to remote it.. which got me to thinking about how I do a reconfiguration of a plexi input arrangement into a 2204 structure, using a DPDT, and how it would be SOOOO much better if I could use transistors, and toggle them all between the modes with a SPST.

Kevin's examples in TUT seem to go from basic shunt to very complex footswitching matrixes without much else in between.

[Mickey_C]

I found the Fender schematics... they're using a wimpy little J111... pretty impressive. Is that the recommended part for something like this?

(updated drawing, using fender example with J111):

[mhuss]

Almost any JFET is fine, as the cathode will only ever get 2-3 volts max across it, and 1-2mA of current through it.

--mark

[Mickey_C]

I'm hoping the JFET will not affect the sound; it should be just the same as running right to ground, correct?

[drz400]

You might have less popping using a Vactrol like a 5C1
Put a 47K resistor across it as well.
If you are switching resistors it will also have pop problems
It will make some noise though as you switch depending on the size of the cap, like going from nothing to a 5uf it will be hard to get rid of the noise.
BTW, it has been done before many times, it just has it's problems and that is why it is generally avoided