Originally Posted by toma
yeah, it is a bit complex, but if it reduces the tone drain that is caused by these rack fx when not on then im up for it...
I should have explained this a bit more, I'm not totally sure what your signal chain is, but it sounds like you are going to have a rack effect going to one input and your guitar going straight or through some pedals into the other input, am I close?
As a general rule, we want high input impedances and low output impedances relative to ground, there are a few exceptions, but this fits for about 95% of the time. High output impedance into low input impedance = major tone suckage (muddy and loss of highs). An exception would be an old Fuzz Face, they have kind of a medium input impedance that is very interactive with the guitar pickups, which also explains why they sound better with single coils than hbs. But, you definitely want true bypass in these for clearest signal. Of course, Hendrix used "tone suck" to his advantage, his wah and probably his Fuzz Face were not true bypass, and live he used those funky curly cords which all add up to major roll off of high end. That, along with his fingers and general funkiness, added up to the killer sounds he would get and why other people plug into strats and can't figure out how Jimi's sound was so warm and theirs is so harsh (but I digress OT, I have a habit of doing that, very unorganized thought process is mine).
If you look at the schematic of your amp, you'll see a 1M resistor going to ground, this is pretty much setting the input impedance at 1M, the 68K resistors from the inputs also help a little but for practical purposes the input impedance is 1M. Now, the resistance from 1 input to the other is the 2 68K resistors in series or 136K, which is substantially lower than 1M, roughly 10%. Now if the rack unit follows the general input/output impedance rule stated above, and I'm sure that it does, it has a low output impedance, probably 10K - 100K or so. So, when if you plug your guitar into one input, and the rack unit into the other, when the signal is going from the guitar straight to the amp, the signal no longer sees a 1M impedance, it sees 136K - 250K or so, which is substantially lower. Electron flow is kind of like water, it will take the path of least resistance, so what will happen is that you will get considerably more high end rolloff, know as "tone suck".
Make sense? So, while it would work, for tone purposes, I don't recommend doing it unless you have a way to completely switch out one signal path or the other, meaning input and output. The same principle applies to inputs, if the signal sees 2 inputs at the same time the 2 high impedance inputs are now in parallel to ground substantially lowering the overall impedance. For example, you split your guitar signal to go into a rack unit that has an input Z (Z=impedance) of 1M, and the signal also sees a pedal with an input Z of 500K, both are relatively high Z on their own, but now the signal is seeing the parallel paths to ground and the Z is now 1/((1/R1)+1/(R2)) or approx 333K which is substantially lower than either resistance on it's own.
Whew, that was long, it's slow here at work, but I need to get a few things done here so I hope that helped. Or maybe it raised a few more questions, which I can try to answer.
Yes, I know this is kinda way OT from the original questions, I think those got worked out though, right?