Weird multiple output question

[DRodriguez]

So I have a weird thing with 3 mono outputs.

Is there a way to wire 3 output Jack's so that output one sums all 3 outs unless a cable is plugged in to another jack?

So to just put all the scenarios:

---------------------Scenario 1---------------------
1 cable into output jack 1: all 3 signals are summed
---------------------Scenario 2---------------------
1 cable into output jack 1 - Signal 1 into
1 cable into output jack 2 - Signal 2+3 summed
---------------------Scenario 3---------------------
1 cable into output jack 1 - Signal 1
1 cable into output jack 2 - Signal 2
1 cable into output jack 3 - Signal 3

Also, less important, whats the best way to put a phase shift on a knob?

[crochambeau]

Regarding the outputs, use Switchcraft 12A on the ones you want to normal to the "main" output. Feed the output wire to tip as usual, then wire the output switch/shunt to the "main" mix output. That way feeding the mix is only interrupted if a cable is inserted. I would put mixing resistors in line with my design, as I don't like mixing two sources on a conductor, but your mileage may vary.

My outline above changes the flow you describe (in yours, the mixed jack moves, in mine it's stationary), though I feel my outline is simpler, yours might be feasible - I just haven't done the mental lifting on it, and would expect the parts to be more complex (either more expensive, or more prone to failure).

Regarding phase shift, are you talking polarity inversion, or a real time controllable time delay to align trailing signals? Plenty of ways to go about the former, the latter is nightmare fuel depending on how finicky/technical you want to be.

[DRodriguez]

Thanks, that makes sense.

I know how to invert polarity, but that's not quite what I'm looking for. My primary concern is weird phasing issues from the multiple sources causing a loss in low end. I'm looking for the time delay option. I don't think it's going to be needed, but I'm not quite sure yet. Not looking for something super advanced like the little labs devices, just something quick and dirty. I have a feeling this is a little out of my league though.

[crochambeau]

Yeah, you could try a single repeat delay IC that's clocked at a comfortably high rate, but that still only addresses phase coherence at a particular band - so you wind up just chasing the problem around.

Though if the issue is based in time delay (wavefront delay between spaced microphones) might get away with it. I'm not an IC guy so I'm just speculating here.

[DRodriguez]

Yeah, I think that's what I'll try and look into.

I think I'm going to build a guitar with 3 rail pickups mounted diagonally staggered 2 stings to a pickup. That way I can have a bass out, mids out, and highs out.

I have serious doubt about how effective that will be, but it seems like fun!

Thanks for the help

[eatyourguitar]

we can engineer simple circuits that correct complicated phase problems even if they only collide at some frequency but we can't do it without the data being measured from sine waves and oscilloscopes. sometimes if you post the circuit, we can plot the phase rotation per frequency very quickly with reasonable accuracy.

[crochambeau]

sometimes if you post the circuit, we can plot the phase rotation per frequency very quickly with reasonable accuracy.

This assumes single origin source, given what DR does (for a living, unless I'm in error) I think leaving the design open for multiple sources (like DI plus miced amp ranging to correction between dissimilar source/waveform) is required. So it boils down to a micro-delay that can be tuned by ear. I haven't started playing with delay ICs, so I haven't the foundation to make a suggestion toward a circuit - but everything I've seen tops out in the tens of milliseconds.

I have, however, observed the concept in action with adjustable micro-second delay networks in my own studio, but those are big, clunky, and lossy.

[DRodriguez]

It's a guitar. So 6 strings into 3 rail pickups lined up side to side. So 3 sources with similar but different wave forms. There will be some crossover because of how tight it will be so some phasing issues will occur. My goal would be to choose what frequencies they occur at using a micro time shift. No complex circuitry besides the time shift, pickup to volume pot to outputs.

Rough Plan:


And no you are not in error, I do sound professionally. But this definitely is purely for fun, not work.

[eatyourguitar]

for this you can actually do it with single or multi-stage RC networks as long as the corner frequency of the low pass is well above the range of the guitar and it's harmonic content. obviously we can more easily adjust the resistor on a pot in an RC network a lot simpler than any variable capacitor. the range of the pot needs to keep the guitar part of the frequency response very flat for this to work. I have read that there are phasers built this way. if you do want to try digital, I would suggest DSPic. the processor is designed to do microdelays with onboard AD/DA