Question about potential tone loss/impedance issue

[Sleipnir]

Howdy,

The other day I plugged my Hughes & Kettner Rotosphere using its stereo outs into two amps - it sounded awesome. I then played a few different other pedals (all mono) through the Rotosphere but with the Rotospehere bypassed and I really liked the sound and feel of the wider spread offered by two amps. Now, the Rotosphere is not true bypass and does alter the sound, so I am thinking about a solution that allows me to play it in stereo through two amps but then easily lets me bypass it but also allows me to select only one amp when playing other pedals or no pedals at all. I only play at home, and yes, it would not be too hard to just plug and unplug some cables, but that would be too easy :-)

Anyway, here's what I'm thinking. I currently have a bunch of pedals connected through a 5 loop switcher. If I move the Rotosphere to sit in the 5th loop by itself, I could connect the send of that loop to the mono input of the Rotosphere. Instead of sending an output of the Rotosphere back to the receive of loop 5, I connect each side of its stereo out into a patchbox. Each of these patch boxes would simply have two ins and one out, basically acting as a Y-cable if you will. The out of the 5 loop switcher would go to an A/B/Y box (exact model yet to be chosen, but perhaps  either the Fulltone True Path, Lehle Little Dual, or Radial Twin City). The A & B outs of this ABY pedal would go to the same patch boxes that the Rotosphere outs would go to (e.g. A out from the ABY pedal and LEFT out of Rotosphere to one, B out of the ABY pedal and RIGHT out of the Rotosphere to the other). The single output of each patch box would then go to one amp each. Because the 5th loop of the 5 loop switcher is not a complete loop (i.e. there is nothing connected to its return), when engaged, no signal would be going  to the output of the 5 loop switcher and thus only the Rotosphere would see a signal (nothing would hit the ABY pedal). This signal would go out as a stereo signal via the patch boxes to the amps (left and right). If instead the 5th loop is not engaged, the Rotosphere would be bypassed and the signal would go to the ABY box, and from there I could choose either amp or both combined (as a mono signal split between two amps).

The only concern I currently have with this idea is that the outputs of the ABY box and Rotosphere are always tied together via the patch boxes. There would never be a signal coming from both the ABY box and the Rotosphere at the same time, but I am wondering if there would be some unintended loading going on. Since it is the outputs of the two pedals that are connected, wouldn't that, if anything, lower the overall output impedance which thus should not be an issue? Could it affect the actual signal in any way?

Any thoughts on whether this might work?

[GibsonGM]

Can you draw this out in pic form, man?   It's easier to "get it" sometimes when you're just looking at a diagram showing what you mean.  Doesn't have to be fancy, but should easily get across what you need to know (the 'problem case')....thanks...

[Sleipnir]

GibsonGM - here is a diagram of what I am thinking of doing and would love to hear some feedback on. Let me know if this makes it any clearer?

[slacker]

If the patch boxes and the ABY are all passive it won't do what you want when the Rotosphere is off. You're correct, you will get signal loading issues because the outs of the Rotoshpere are connected to the amp inputs and to the outputs of whatever is coming from the ABY box, some pedals might work happily like this but others won't and straight guitar definitely won't.
You could make it work by adding a buffer between the looper and the ABY so that weak signals like guitar aren't loaded and using resistors in the patch boxes to passively mix the signals.

Another problem is if the ABY box is in Y mode then Out A and Out B are connected together so both outs of the Rotosphere and both amp inputs are also connected together so you won't get any stereo effect. The approach above probably wouldn't solve that.

I can't think how off the top of my head but you could probably build a stereo true bypass box for the Rotosphere incorporating the ABY switching.

[slacker]

Just thought another problem is some AB boxes ground the unused output to reduce noise,  that would mute one of the amps in any setting.

[GibsonGM]

^^  YUP

[PRR]

> the outputs of the ABY box and Rotosphere are always tied together

NEVER tie two outputs together.

> never be a signal coming from both ...at the same time

Just the same, they will fight.

The "live" output wants to put, say, 1 Volt on the wire. The "dead" output is obliged to force Zero Volts on the wire. The same wire.

They fight.

In an extreme case, one of the outputs smokes. If it smokes dead-open, the other output now has full control. In more likely cases, both outputs fight as hard as they can, within their protection limits, and distort the signal badly.

Yes, some outputs mix without harm. Many boxes have resistors on their outputs. I had a fully patchable synth where ALL outputs had a 1K resistor, you could tie any or all of them together, no harm no corruption. 1V and 0V gave 0.5V of clean signal. However in guitar-land these resistors may be a 500 Ohm line-surge dingus or a 50K tone-shape network. Mix these and you get a 99:1 mix. If you wanted the signal from the 50K box, you will be disappointed.

In general, in your case, if you replace the "patch boxes" with two 50K resistors (a mix network), the two sources will play-nice together, and the output will be ample to go to the amps.

[PRR]

me> two 50K resistors (a mix network)

Also-- many (not all) guitar amplifiers have two jacks (per channel). They may be marked Hi/Lo because with one guitar they give different gains. However if they use the classic Fender input jackery (two 68K resistors), if you plug in two plugs the two signals mix =equally=. (So two guys can play one amp-- we used to be that poor.)

So you might skip that "Patch Box" and go right to the amp(s).

[Sleipnir]

Thanks guys, you definitely pointed out a few things I had not considered and thanks for talking the time to do so. PRR - my Super Reverb does indeed have dual inputs, but my other amp, a Peavey Classic 30 does not. I am planning on building a Marshall Superbass clone at some point, in which case I suppose your suggestion would work well (assuming the Superbass uses a similar input strategy). With your suggestion, I could have the Rotosphere outs dedicated to one set of amp inputs, and the ABY box to the other.

Just out of curiosity, could you elaborate on your 50k mix network - are you simply putting 50k resistor after each of the four outputs? Wouldn't that attenuate the signal?

Another idea I had was to put the Rotosphere in a stereo bypass loop such as this:
http://www.loop-master.com/looper-stereo-p-97.html

However, I assume since this is a passive box that each of the signals coming out of the two outs when in bypass mode are attenuated relative to the input (i.e. the input signal gets split in two) and thus there will be some tone loss at the amps. Also, this approach would not allow me to easily select just one amp.

Now if I could build a stereo loop pedal like this but have the signal go to only one of the outputs when bypassed, then that could be interesting. The output that is disconnected when in bypass would go yo my Super Reverb Normal channel, the other, which is always active, would go to an ABY box which in turn would go to the Clasdic 30 and the Vibrato channel of the Super.

[slacker]

Now if I could build a stereo loop pedal like this but have the signal go to only one of the outputs when bypassed, then that could be interesting. The output that is disconnected when in bypass would go yo my Super Reverb Normal channel, the other, which is always active, would go to an ABY box which in turn would go to the Clasdic 30 and the Vibrato channel of the Super.

You should be able to do that with your current setup, patch the rotosphere into loop 5 using the input and the left output as though it was a normal mono pedal. Out from the looper would go via the ABY to the Classic 30 and the Vibrato channel of the Super.
Take the right output of the rotosphere to the normal channel of the Super. Providing there's no noise from the the right out of the Rotosphere when loop 5 is bypassed that should work.

[Sleipnir]

Slacker, that is an interesting option that I will need to try! The biggest question there will be, as you hinted at, whether the Rotosphere will inject any noise into the Normal channel of the Super Reverb when bypassed. I will try this and report back.

Before I realized you had responded, I actually tried to draw a schematic for a loop pedal that would send a mono signal to the Rotosphere and return the stereo outs from the Rotosphere into two return jacks on this pedal. When the pedal is ON, these returns would go to two outputs on the loop pedal. From there, OUT 1 would go to an ABY pedal whereas OUT 2 would go to the Normal channel of the Super Reverb. When the pedal is off, the INPUT would go to only OUT 1 thus there is not signal splitting. If you get a chance, take a look and let me know if you think this would work. My biggest question is whether it is ok to leave the SEND and RETURN jacks "hanging" and not grounded when bypassed. I did short OUT 2 to ground when bypassed - this might eliminate any possibility of noise on the Super Reverb Normal channel? Also, is the overall grounding appropriate?




Again, your suggestion would be the easiest approach so I will try that.

[wavley]

I've never seen the schematic for the Rotosphere, perhaps a simpler way is to check out it's bypass schem and see if it can be improved?

[slacker]

That switching schematic looks good to me. You could connect pin 4 to pin 9 that would ground the send in bypass, which would stop any possible noise issues.

[Sleipnir]

Slacker - I like that addition and have added it here:




I also realized I prefer not to use the Normal channel on my Super - I usually to play with it disconnected which gives me a bit of a boost on the Vibrato channel but it renders the Normal channel useless. Soooo, I came up with an alternate approach that might work with two single channel amps. In this version, the Returns are both grounded in Bypass mode. The ABY pedal's B output is connected to the loop pedal. In bypass mode, the B output from the ABY pedal simply gets sent to Amp 2. While activated, the B output of the ABY pedal is grounded within the loop pedal whereas Return 2 now goes to Amp 2. The ABY pedal needs to be set to "A" or "A & B" when the loop pedal is active for this to work - the B output is simply going to ground, so if in "B" mode, there is no signal going to Amp 1 when the loop is active.

I cannot figure out a way to get an LED added here - I take it that is not doable?

[Sleipnir]

Whether I end up using two amps in the long run remains to be seen, but as my first pedal build I plan on making one of the two designs listed below (unless they are both flawed). I am curious if anyone could provide some thoughts on which design is better:

In both cases:
"Bypass" mode
1) Guitar IN sent directly to OUT 1 which goes to the input of an ABY pedal (this would be an active ABY pedal which I have yet to buy)
2) The "A" output of the ABY pedal goes to Amp 1
3) The "B" output of the ABY pedal goes to ABY "B" IN on the pedal which is routed directly to OUT 2 on the pedal which in turn is connected to Amp 2
4) This allows me to play either of the two amps or both together, completely bypassing the Rotosphere

"On" mode:
1) Guitar IN routed to SEND on pedal, which then routes the signal into the Rotosphere Mono input. The lush Stereo outs are then routed to RETURN 1 and RETURN 2 on the pedal.
2) RETURN 1 gets routed to OUT 1 which goes to Amp 1 via the ABY pedal
3) RETURN 2 gets routed to OUT 2 which goes to Amp 2
NOTE: although it might depend on the (active) ABY pedal used, it might have to be switched to "A" only rather than "A&B" when using the Rotosphere (obviously, it cannot be set to "B" only). That could be easily determined once everything is connected (i.e. check for any signal loss or noise). If a passive ABY pedal is used, it would probably have to be set to "A".

The two designs differ in that Design #1 does not have an LED and grounds both the two return jacks (in bypass mode) as well as the ABY "B" IN jack ("on" mode) when not in use, whereas Design #2 has an LED, but only grounds the SEND jack on the pedal when not in use (bypass mode). In an ideal world, having the LED would be nice, but since it would be pretty obvious if the Rotosphere is on, it is not a necessity. If the grounding scheme in Design #1 is better, I would definitely sacrifice the LED. I just don't know if it matters whether I leave the RETURN jacks and ABY "B" IN jacks "hanging" when not in use in Design #2....

Design #1




Design #2




Any thoughts would be appreciated!

[Sleipnir]

Just to follow up on my own thread. I ended up building the design with the LED and picked up a Fulltone True-path ABY-ST pedal. It works as intended except I had to change one thing. Where it says "ABY/B In" I actually connected Channel A from the Fulltone True-Path. The reason for that is that in order for the transformer isolated output on the Fulltone ABY pedal to work, and thus also the phase reversal and ground lift (both of which I need), output B has to have a direct connection to one amp, otherwise the grounds "connect" again at the pedal I built which defeats the purpose of having a transformer isolated ABY pedal. I did not consider this when doing the design. No big deal except the incorrect labeling on my pedal.

Here is a photo:



And another:


I am not a good graphic artist, so I just put the signal flow on top of the pedal 

Thanks for everyone's help earlier in the thread!

Edit: Fixed the pics

[mth5044]

Well done! Your graphics are useful and look good.