level drop on one delay when A/B/Ying 2 delays?

[dubkitty]

as part of my attempts to work up a Frippertronics-influenced looper/faux-Revox-loop composition presentation, i've been trying to create a setup where i can split the input signal to my two delay/loopers and then sum the two outputs to send them to the amp, so i can have independent loops going on the delays simultaneously while isolating the output of one loop into the input of the other.

i am using an Akai Head Rush E2 and a TC Flashback for this; i generally use the E2 for loop-based compositions and the Flashback for sound-on-sound tape loop stuff. for splitting and de-splitting the signal i have an Ernie Ball stereo volume/pan pedal and a LiveWire A/B/Y box.

here's the problem: no matter what wiring configuration(s) i use, the output level from each delay is normal as long as i have only one delay connected to whichever device is summing the outputs. but as soon as i connect both delays to the summing device, the level of the Head Rush drops by approximately 40-50% including the level of unprocessed guitar signal when the E2 is turned off.

does anyone know why this is happening, and/or have any suggestions as to how i can fix this short of buying a $400 Lehle switch with buffers and shit??

[MEC]

I'm a little confused.
Are you going Guitar>EB Vol.>split to the two delays>into the 2 outs on the A/B/Y>out from the A/B/Y input>the amp?
Is the Head Rush side always the one that loses volume no matter which A/B/Y output it's plugged into? If so you could try a buffer before or after the Head Rush and that may help.

[dubkitty]

Are you going Guitar>EB Vol.>split to the two delays>into the 2 outs on the A/B/Y>out from the A/B/Y input>the amp?

i've tried it both ways, i.e. with the guitar/board going into the A/B/Y in and the delay outs going to the EB volume/pan inputs. with the guitar into the A/B/Y OR the EB V/P, i have the "either delay has normal level/Head Rush level drops when both are connected" problem. when i use the EB V/P at the end to sum the inputs, i'm only getting one delay in the output i.e. i'm only getting the A or B signal depending on which output jack i plug into; the EB V/P has stereo jacks on all its connections, but is SUPPOSED to pan the A/B inputs down to a mono A output.

Is the Head Rush side always the one that loses volume no matter which A/B/Y output it's plugged into? If so you could try a buffer before or after the Head Rush and that may help.

it's always the Head Rush. but the thing is, it's fine when it's in the circuit by itself. it's only when both are plugged in simultaneously that it drops. it seems like there's got to be some kind of ground loop, or phase cancellation, or failure for one side of a stereo signal to be getting through, or something. i think i've eliminated the A/B/Y box or the EB V/P as sole source of the problem per se, because the same thing happens when either is used to sum the signals. the Flashback runs in mono; i'm not clear if the E2 runs in stereo on a single TRS jack or not, but it isn't listed as such in its manual. i'm fuckin' stumped.

[MEC]

i'm fuckin' stumped.

Me too. I am not too familiar with running two delays.
Most of my signal splitting experience is with running two amps, which doesn't involve recombining the signals.

[hbombgraphics]

Can you put a buffered pedal before the head rush?
That may help. It would at least let you know if it is a buffering issue I guess.

Also I am curious as to what is powering all these goodies.

[smallsnd/bigsnd]

let's say you plug 2 drum machines (A/B) into an ABY with the output (Y) going to an amp. with either one on, you'll have full volume. turn both on though and you'll have halved volume. there are almost always losses when going the passive route... what you gain in simplicity of design and parts, you lose in other areas - generally frequency and/or amplitude.

a buffered A/B/Y will probably do the trick.
or get a cheap mixer with more than one send/return.

[Achtane]

Yeah...I've read about this problem before and a tiny mixer is often suggested.

[dubkitty]

Can you put a buffered pedal before the head rush?
That may help. It would at least let you know if it is a buffering issue I guess.

Also I am curious as to what is powering all these goodies.

i'll try this. i assume a Boss pedal would do?

right now everything is running off a OneSpot. i thought there might be a problem due to everything on the board running on the one adapter so i broke it down to 1Spot->E2->Flashback->powered A/B/Y. I tried running the Flashback on a battery; it didn't help. nor did running the A/B/Y on a battery or using it unpowered.

the newest clue that i noticed while trying different things last night was that when i turned off the buffering on the Flashback (it allows you to switch between true and buffered bypass via an internal dip switch) i could have "A+B" selected on the A/B/Y and get full volume on the E2 until i turned the Flashback on.

the drawback to the mini-mixer idea is that it'd be kind of difficult to work with in a live environment what with the lack of light/my poor vision, and the idea is to be able to take this concept out in an opening-act capacity in a few months when i have it worked up to my satisfaction.

[hbombgraphics]

Yeah a boss pedal buffer might do it, It's worth a shot anyway.
I don't think the Miliamps is over the one spots capacity the ABY should draw almost nothing.

I hope this works it sounds very cool!

[eatyourguitar]

if the two delays have different output impedance, one of them can overpower the other one. also, they are interactive with each other. since they are all connected without buffers, electricity can go in both directions without buffers. like a signal from one delay can run right back up into the other one through the output and drain to ground through a signal shunt. its difficult to say whats happening without opening up the pedals. but really, you just need a buffer, or a simple blend pot might do the trick. even though its passive, a blend pot will add 0 to 100k output impedance to one of the pedals. so if one of the pedals has an output impedance of 1k and the other one is 100k, you just crank the blend pot to one side till it sounds right. impedance matching would also help slightly in getting the brightness the same for both pedals. just get a 100k linear pot and wire the middle lug to the output. the outside lugs get wired as inputs. now you have a super simple, cheap blend pot that also does ghetto impedance matching.

[dubkitty]

so i tried placing my RV-3, which of course is buffered, in various spots in the path. putting it on either side of the Head Rush didn't help, but when i placed it after the Flashback, SUCCESS!!! i can now pan the input from one delay to the other and have both looping simultaneously through the A/B box with balanced input. now i'll need to find something to use in place of the RV-3 for this purpose, because i kind of need it for reverb and echo farther up the chain.

thany you for your input, folks. i really wanted to understand why it was screwing up as well as find a solution, and you've supplied both. now i just need to figure out the best way to buffer the Flashback and i'm good.

[MEC]

I'd get a Boss TU-2 that way you get another use out of it too without adding a unwanted effect.
You should be able to find one for $50 or so.
You could also probably find someone here to build you a cheap buffer.

[eatyourguitar]

I'll build you a 100k linear pot in a tiny 1590a with 3 jacks. I got some 1590a left and no idea what I'm gonna do with them. I'll do it on the super cheap.

[dubkitty]

thank you for your kind offer, but by the time i saw your post i'd already found a used Solidgold(fx) buffer on eBay for $50, so i should be good. thank you for your explanation of the technical stuff involved...i really appreciate your help.

[eatyourguitar]

no problem. I think I'm gonna make that blender for myself anyway now that I have the idea in my head.