Will this work? A/B/C box

[limit6]

Hey, took a stab at an A/B/C box with independent mutes for each out.  How'd I do?  Will this work?

[blackieNYC]

I've been told to put a 1k in series on each branch after the splitter.  Weird loads and short protection.
Anyone?

[GibsonGM]

Good idea, Blackie, but - IMO, this is actually easier (and cheaper as switches COST) and sounds better due to the buffering effect of the FETs!

Check it out - you can rig up a switch system to ground whichever in/out you aren't using....I use a setup very close to this (a A/B/Y box, forget where the schema is tho) to switch between amps.

http://www.muzique.com/lab/splitter.htm

[limit6]

Wow, you guys are awesome!  Thank you so much 

Haha, well, I already routed the box for this configuration.  So per Blackie's advice, is the only thing missing from my layout a 1k resistor going to each output jack?

[blackieNYC]

Now that you bring that up, I have built a three way splitter and three channel mixer. (Gibson-You nailed that?)  I spent many weeks tryin to get his going with just FETs, and determined as I was I could not get it to my satisfaction. 9v was a must. A little boost to make up for mixer loss and then a few dB extra - FETs could not do it cleanly.  This is a box you want to be as neutral as possible. The Tilman FET preamp has 5-6% dist and you can hear that.  The Fetzer preamp can be designed to the measured specs of the selected FET, but it is not clean. I went op-amp.  I could have put together one of the multiple FET boosters, but i threw in the towel. Very happy with the simple TL072 Buffer. You are just splitting so you only need one booster, whatever it is.  Maybe the MOSFET booster would be clean enough.   
Going to three amps? You could address the buzz of colliding 120v circuits you might encounter.  You know that  deal.  Someone else chime in on the split point resistors. 

[limit6]

Ahhh, I hadn't considered the relationship the three amps would be having with one another on the way back.  I suppose that's why the Cornish PCB on Aaron Nelson's page had a 10mf output lytic going to each output jack... so the signal wouldn't backwash.  Yes?

[limit6]

Ok, so I'm gonna go with the Jack Orman splitter, as prescribed.  Thank you, amigos.   You're the best.

[PRR]

> Will this work?

Yes, but.

Where does Input A go to? Probably a HIGH gain amplifier.

What is Input A connected to when NOT selected? "Nothing", or rather several inches of wire.

Plug a cord in your amp, turn-up, and wave the other end around. It picks up ALL the garbage in the room. Power hum/buzz, PC whine, maybe radio, and if you get very near unshielded audio it will pick up that too.

It is basic good audio hygiene that, when you want "no signal", you *connect* to a No Signal point. Audio ground.

I also don't understand why you run Ground to the *center* of each switch, a pole that does nothing else.

I hate to trash-up your hand sketch, but this is IMO a better plan. A(out) connects to Buffer or to Ground. Either way it *knows* what to do.



Isolation resistors may not be essential; but again Good Audio Hygiene... when you split a signal, 'specially in live performance, make sure that an accident on *one* output does no harm to *other* outputs. Say you do a Glam-Rock show in pointy-heel boots. And stomp on A(out) cable so hard it shorts-out. Without isolation, ALL outputs are shorted, dead, you pluck and nothing happens even if you try switching A/B/C. With sufficient isolation resistors, B(out) and C(out) keep on working, the song/show is not ruined. You switch to B or C out.... may not be the "right" tone but you are making music and the crowd keeps dancing.

ProAudio and Hi-Fi tends to isolation like 47 Ohms to maybe 1K. Guitar-cord circuits are generally higher impedance. If one output is shorted, with 1K that's a 1K load. Many guitar-cord buffers can't drive 1K well. One reason is that most guitar-cord inputs are 100K to 1Meg, so heavy drive is not needed. I think 10K or 20K are quite reasonable.

If you must use those meshugganah 3P3T switches, make note of which-way the lugs go, so you don't wire it 90 degrees wrong.

[limit6]

PRR, this is great!  Thank you.  I totally welcome trashing my sketch.  I don't view it as trashing, but moreso that you're protecting me.  I'm learning lots and it's all thanks to good people like you offering constructive criticism.  

To answer your question about why I chose to use the center pole on the 3pdt switches as ground - I was planning on wiring up three LEDs, but omitted them from the drawing to simplify things.  I was going to stick the negative leg of the LEDs on the top middle pole.  I see what you did with the resistors and swapping the ground the upper left hand pole.  Makes total sense.

If I went with the Orman splitter suggested above, should I still incorporate the 22k resistors you advised, or is the circuit all inclusive and I'd be able to drop it into my sketch above?

[PRR]

> should I still incorporate the 22k resistors

I dunno. The experiment is cheap. Build without. Feed A to an amp, pass signal. Short-out B. Is A signal harmed? If so, blow the 36 cents for 3 resistors and try again.

Alternatively just throw them in mindlessly. Seeing the AMZ plan, 10K or 5K may be fine. And I note that bus "i" is un-used. Two cuts (to make 3 isolated strips) and you can put the build-outs on the board.

[limit6]

Great info!  I'll complete the build on monday and report back.

[limit6]

Success!  Giant thanks.  Sounds wonderful and everything worked first time around except for mute switch A.  Replaced the FET on that channel and then everything worked 100%.  Guts shot was taken before the wiring was complete, thus, no wire coming off the 3pdt's going to the output jacks yet.