JFet Channel Switching - Negative Voltage Supply

[Boots Deville]

This is my first attempt at channel switching using JFets.  I'm following TUT as my guide.

I've wired up a channel switching board and installed it.  The plan calls for -36VDC to supply it.

Power supply looks like this: HV tap [313VAC to ground] 120K dropping resistor [173VAC] 1N4007 [-5.2VDC] 36V Zener parallel with 10uF cap to ground.

The board is being supplied with -5.2V instead of the desired -36V.

Do I have an incorrect valued component in this portion of the power supply, or is there more likely a problem in the rest of the channel switching circuit causing the low voltage?

Regards,

John

[tubesornothing]

Sounds like it should work.  Measure the voltage without it attached to the switch circuit to see the problem source.

If you search the forum, PRR did a real nice rework of the KoC switch schematic for one of my builds.  Might or might not help.  I got the -36VDC from the bias supply.

[Boots Deville]

Thanks ToN.  I remember that thread, now that you mention it.  I'll paste the link here, to make it easier for me to find the next time! :)
http://www.el34world.com/Forum/index.php?topic=3912.msg33963#msg33963

This amp is cathode biased, so I don't have a bias supply to tap into.

Thanks again,

John

[JayB]

Is it possible to get some thing like that to switch 3 channels or more on the fly? With out going to an IC?

[PRR]

> switch 3 channels ...With out going to an IC?

There is a saying in binary math: 0, 1, 2, many.

If you want more than two states, it is "many". i.e. you won't have the kind of dead-simple plan which is possible for a 2-state system.

In this case, the "two state" function is IN the pedal. The alternate-action foot switch.

The cave-man way would be to make the pedal push a ratchet on a 3-tooth gear, and 3 switches.

The alternate-action switch may be a 2-tooth (or 4-tooth) ratchet. Or it could be like the mechanism in a ballpoint pen, altho this too is usually an 8-tooth ratchet mounted end-wise and every other tooth indented.

The switch on "3-way" lamps is another example. However these are really 4-way (one of the states is "off"). You could try to adapt one. However these things break when I touch them, they sure won't like being stomped.

YOu "could" do a 3-switch "with out an IC"... whole 64-bit computers were built without a single IC.

I could dredge old memories and come up with a plan. But the reason ICs took-over the world is that they make the non-simple things pretty simple.

However this paricular job is complicated because it uses quite high voltages to drive the FETs, because they are clamping quite high signal voltages.

Oh... the next step after "2" is "4". You would probably build a 3-state register as two 2-state registers, giving four possible states. Yes, I bet there is a ring-of-three approach, but it might be just as many parts as two flops.

[jhadhar65]

Also, look at LooseChange's stuff.  He does a lot of jfet switching, IRRC.

[imaradiostar]

There is a neat design on R.G.'s site that uses a simple IC/diode/mosfet arrangement that could easily be modded to run channel switching. You could easily set it up to switch 4 or 5 channels and have assignable loops on relays and such...the sky is the limit. It uses reliable SPST momentary switches to control the logic.

I saw a box in a store recently that appears to use a similar design- eight loops, eight momentary switches and eight dip switches per momentary switch that select the loops that are active when you hit that footswitch. It also has a switchable buffer/amp that can be used if needed.

just thought I'd mention it since it seemed so useful.

jamie

[PRR]

> I could dredge old memories

neon ring counter.

Low connection count, negligible cost, self-indicating. Fussy.

http://www.dos4ever.com/ring/ring.html
http://www.dos4ever.com/ring/neon_ring_counter.pdf
http://www.dos4ever.com/ring/scope1.jpg
http://www.tech-systems-labs.com/books/GE-lamps.pdf

Neon lamps: http://www.mouser.com/Search/Refine.aspx?Keyword=36NE002
http://www.mouser.com/Search/Refine.aspx?Keyword=36NE004

Hoffman neon-box: http://www.dvq.com/geiger/cm_service.pdf

Needs to be flipped for negative supply, or use P-JFETs Mouser J175 J176 J177 J270

B32529C1472K189 Capacitors 0.0047uF 100V 10%  $0.10 
 

[JayB]

eckk.. That is one two many. I guess I'll have to go with the IC just for simplicity and do my own board. Thinking of using a PDIP DM74LS374. I've seen RG's page on a different one. This could get me 8 switching states although I wouldn't know what to do with them all. :o

http://www.datasheetcatalog.org/datasheets/90/375622_DS.pdf

[PRR]

> DM74LS374

Well, in that context, "tri-state" means something else. (The output may be High, Low, or Open.)

You want three (or more) output, only one High, and the High output can be moved around the loop with a single input.

The '374 is several 2-state flip-flops. You could do a 3-state A/B/C switch with that, but the thinking is hard.

Lancaster's CMOS CookBook is GOD for projects like this. Lemme figure out where I left it. Probably next to AMCjim's screwdriver...

[PRR]

me> You want three (or more) output, only one High, and the High output can be moved around the loop with a single input.

CD4017.

Cute animation of a 4017 in action:
http://www.geocities.com/it2n/counter.html