Series/parallel AND order looper?

[Faldoe]

I want to build a loop switcher with 2 footswirches and two loops.

A line 6 dl-4 would go in each loop.

There would be no global bypass. One switche would choose whether the loops are running series or parallel and the other switch would choose the order of the loops (when in series mode).

Is this possible? I've been an effects order switcher before. And I'm pretty sure one can build a series/parallel switcher, but can both be combined as I'm discribing?

Thanks.

[crochambeau]

Wholly possible.

Would you be hoping to do this footswitchable or toggle switching?

If you have already wrapped your mind around changing order in a series string, the only thing you need to do to initiate the parallel function is to short the outputs and inputs together, though the outputs would benefit from mixing resistance (it needn't be huge), so I would use two poles normally open that would connect to two 1-10K resistors already on the output jack and hard connect the input directly.

If you're after toggle, you might be able to get away with a single three position switch of the ON-ON-ON variety, but I haven't mapped that out yet.

[Nocentelli]

...the only thing you need to do to initiate the parallel function is to short the outputs and inputs together, though the outputs would benefit from mixing resistance (it needn't be huge), so I would use two poles normally open that would connect to two 1-10K resistors already on the output jack and hard connect the input directly.

If you're after toggle, you might be able to get away with a single three position switch of the ON-ON-ON variety, but I haven't mapped that out yet.

Is it really that simple? Just short the inputs and outputs? I built a 4PDT order juggler for two loops and it's extremely useful, but I tried for ages to devise a series/parallel switch but it was all way too complex. I'd assumed you would need to buffer the input first, then split it into two separate parallel chains, then do the series/parallel thing with a 3P4T like in a guitar pickup series/parallel switch.

[crochambeau]

It is simple, we just like to over-complicate things. I'm not saying just shorting inputs and outputs is a slam dunk fix-all, but under the criteria given it's probably an option (and easily reversible).

It's going to boil down to what you have in the loop, and what you are using to drive the loop. If you are running a pair of really high impedance input devices (I'm guessing the Line6 qualifies as this), you're essentially just cutting the input impedance in half (or doubling the load, if you prefer). Most (not all) devices run their output impedance in a well under 10K range, so even if the input impedance of one is a relatively low 20K there will *probably* not be detrimental issues while running them in parallel. Old transformer coupled 600 ohm units will certainly want a buffer.

If you've got old leaky BJT front ends, sorry, a couple of hand built Ge boutique creations being fed directly from a passive instrument it might be a different story, but it shouldn't break anything and the first thing you'll lose is treble (and you have the option of loading down the front with intentional resistance at the expense of noise floor to get that back, within reason (see: daisy chaining the input sockets of old tube amps)).

Outputs, like I stated previously, are going to best be decoupled from one another via a resistance, but again, this is a performance function in that you don't want one bullying the other. Put yer dogs on a leash.

Of course, if you want to go whole hog on it, you can plant a buffer on the front end (shorting the inputs of the looped units at the output of the buffer, which is inherently a lower impedance output) and run a buffer on the return of each feeding a blend control, or whatever you want. Ultimately, small signal handling/routing is not rocket science.

[Nocentelli]

Ok, I can see that would work ok for two DL4s with no requirement for bypass. I'm interested in rebuilding my dual looper with order switch to have an addtitional series/parallel toggle, but it seems that it would be more complex than described above since I also want independent bypass stomps for each loop that would work in series or parallel mode, and a blend control for parallel mode. It strikes me that simply jumpering the inputs would make this tricky.

[crochambeau]

Yeah, in the primitive method I outlined independent bypass switches would kill the other side, so you are correct in that simply shorting them wouldn't work and support that function.

Would the bypass need to pass through the looped leg? Perhaps the bypass switch could be considered a third, uneffected and buffered path that would latch in and out BEFORE the primitive parallel function, so if you isolate the series parallel as a block, bypass would simply take the affected loop COMPLETELY out of circuit while simultaneously offering an unaffected unity gain path from input to output.

You would naturally need one bypass switch for each loop, but under conditions of redundancy it would only amount to a mute switch as you'd already have your "direct" path engaged.

[Nocentelli]

Yeah, in the primitive method I outlined independent bypass switches would kill the other side, so you are correct in that simply shorting them wouldn't work and support that function.

Would the bypass need to pass through the looped leg? Perhaps the bypass switch could be considered a third, uneffected and buffered path that would latch in and out BEFORE the primitive parallel function, so if you isolate the series parallel as a block, bypass would simply take the affected loop COMPLETELY out of circuit while simultaneously offering an unaffected unity gain path from input to output.

You would naturally need one bypass switch for each loop, but under conditions of redundancy it would only amount to a mute switch as you'd already have your "direct" path engaged.

Well...... I'm going to need to see a rough schematic to be able to comment on that

thanks for the suggestions.