Trying to build a splitter box. Anybody able to help?

[butter_side_down]

Hello good people,

I'm having a real head-scratching time with a design that should be fairly simple, and I need some help understanding why it's not working, and, hopefully, finding out if there's any way I can make it work.

First, a little explanation...

I have a Maxon AD999 analog delay pedal that has two outputs (one for the dry signal, and one for the delayed signal). If you just plug into the delay out, it mixes the signals, but if you plug into both outputs, it keeps the signals seperated. I wanted to take advantage of this so that I could run the delayed signal through, say a phaser or chorus, and leave the dry signal unaffected. I realize that I could do this by using two amps, but I really don't want to have to lug two amps to every show every time I play. Plus that would mean that I wouldn't be able to run the two signals together to the same amp. So, I thought, hey, I'll build a little box that will act as a switchable splitter, that way I can enjoy the best of both worlds.

Here's a rough drawing of the unit I built...



Basically, when the switch is in position 1, the signals are split and the dry signal goes straight out, while the delay signal goes through any following effects before going to the output. When the switch is in position 2, the signals are mixed and go through the following effects together before hitting the output to the amp.

When I test it with a jumper cable between the effect out and the effect in (see below), it works like a charm.



BUT, once I hook up an actual pedal into the effects loop (see image below), it gets wonky. When the signals are mixed before going to the effect loop (switch in position 2), everything works just fine; however, when I split the signals (switch in position 1), the delay signal returns from the effect loop at regular strength, but the dry signal is very, very, very faint in the background.



I'm fairly new to electronics, so I'm sure there's something obvious that I'm missing (impedence mismatch maybe?) but I'm not advanced enough to figure out what it is, or, more importantly how to correct it. Anyone here feeling generous enough to explain my folly?

Any help appreciated.

[Eternal Dragon]

Are u testing it with your amp or with a CRO or something?? What i'm thinking is.....

1) When u have the jumper connected, all voltages are reaching the output because there are no resistors to load it down.

2) When you have the pedal connected, the bottom stage is going into the pedal because the input impedance of the pedal forms a voltage divider which divides most of the voltage across the input of the pedal (Ie: More voltage in the pedal than the output of the previous stage). However, output impedances are generally low in order to divide voltage across the input of the next stage, BUT if you don't have a load resistor on the output (If you were connecting it straight up to a CRO or voltmeter), then ALL of the voltage would still be across the output resistor of the pedal, and not where you want it to be.

Actually, that could be wrong :lol:. Something like that.

[Ray Ivers]

I'd suggest adding 10K resistors to the left of the DRY IN and DELAY IN arrows, and a 15K resistor to the right of the EFFECT IN arrow, and see how that works. These values are just guesses to get you started, you may want to play around with them a bit.

Ideally you should be using active buffer stages so that all mixing nodes have the same output impedance and all inputs the same Zin, but the resistor setup can be made to work OK for a given set of effects.

Ray

[Steve Conner]

Generally, you can't just connect outputs of devices together like that to mix them. Connecting one output to several inputs is cool, but not the reverse. I guess what's happening in your case is that the outputs are fighting each other, and the one with the lowest output impedance is winning and swamping the other one. The thing with the resistors that Ray mentioned should fix it.

[butter_side_down]

Dragon, I had a feeling that something like that was happening, but I was only guessing from my very limited knowledge base.

Ray, I'll give the resistors a shot and see how that works. Just to make sure, should the resistors should go in parallel (ie. across the inputs, tip to ground), is that correct?

If that doesn't work then it looks like it's time to start learning about active mixers!

[Ray Ivers]

Ray, I'll give the resistors a shot and see how that works. Just to make sure, should the resistors should go in parallel (ie. across the inputs, tip to ground), is that correct?

No - the resistors should be in series with the 'hot' signal leads, to isolate the various inputs & outputs from each other.

Ray

[butter_side_down]

Gotcha, Ray. Makes sense when I actually take a second to think about it.

[butter_side_down]

Version 2 - Splitter W/Mixer

A simplified drawing for splitter v2 (w/mixer) below. Does this seem more like it?

For the mixer section, I'm planning to use the schem for the MiniMixer available at the General Guitar Gadgets site. I'm sure there's probably a more efficient way to wire the switch, but I'm too brain addled to come up with it right now.