I am trying to design a DPDT-Relay-based bypass box for a delay. This bypass should provide two selectable modes: Either true bypass or buffered with spillover/trails. The mode can be selected with Jumpers or dip switches, for the switching I want to use the same relay, independant of the mode.
Best shot so far was this:
(https://i.postimg.cc/bY2pVYGV/Bypass-Schematic.jpg)
The Relay would be FX/BP1 and FX/BP2, the mode jumpers would be TB/BUF1, TB/BUF2, TB/BUF3. Not sure if it would work, as I am not sure, if the remaining connections in true bypass mode (via C3) would impact the input signal significantly. Also, I would like to ground the FX input in bypass mode (which seems to be impossible?!)
Haven't come up with a better idea yet, though it must be possible: The TC Flasback uses a single dip switch to change modes and I am sure they did not put two relays in there...
Feedback or questions welcome.
I don't have anything technical to offer. I just want to say that I think this is a great idea--one of those things that should have been kind of obvious but evidently wasn't or it would have been done a long time ago. I admire the subtle kind of cleverness it takes to see refinements like this. So kudos.
+1. This is too cool
Could you get the same effect by using a spdt to short the output side of the relay? That way the output stays on no matter what, only the delay input gets to be on or off. Remove the short and it goes back to being true bypass.
Here's an easier way to do this. This method will work with your tb/tails switch being a slide, toggle or even relay. The TB/Tails switch selects between straight through bypass or a buffer to mix the effect output.
It doesn't have an output pulldown resistor (after C2) since it uses the internal one from the effect you are spilling over.
The only instance where this design would be an issue (that I can think of) is if the delay circuit reverses phase at the output. But, you can easily design around that option with the second half of the IC.
Also, if you are doing this as a standalone box, you could make it even more interesting: add a mix knob either as a pot or trimmer for an adjustable amount of spillover.
Great stuff, brother bean. I will incorporate that in my Looper project, potentially with a phase reverse option :-)
Test it first though. ;)
Would the phase reverse option simply be a selectable op amp stage on the wet return?
Quote from: benny_profane on November 07, 2018, 04:47:03 PM
Would the phase reverse option simply be a selectable op amp stage on the wet return?
Yes.
Quote from: Marshall Arts on November 07, 2018, 04:49:25 PM
Quote from: benny_profane on November 07, 2018, 04:47:03 PM
Would the phase reverse option simply be a selectable op amp stage on the wet return?
Yes.
Gotcha. Yeah that'd definitely be handy if you're blending possibly inverted signals. Very interested to see how this plays out!
Quote from: Marshall Arts on November 07, 2018, 04:49:25 PM
Quote from: benny_profane on November 07, 2018, 04:47:03 PM
Would the phase reverse option simply be a selectable op amp stage on the wet return?
Yes.
An inverting stage, of course ;-)
Quote from: benny_profane on November 07, 2018, 04:47:03 PM
Would the phase reverse option simply be a selectable op amp stage on the wet return?
Two ways you can approach it - reverse the phase of the guitar signal or reverse the phase of the wet return. The first one is easier. The second one is probably better because it doesn't change the phase of your primary signal. However, that would only matter if you were running other more complicated parallel loops at the same time.
Anyway, here's how I would do it.
Quote from: madbean on November 07, 2018, 04:56:35 PM
Quote from: benny_profane on November 07, 2018, 04:47:03 PM
Would the phase reverse option simply be a selectable op amp stage on the wet return?
Two ways you can approach it - reverse the phase of the guitar signal or reverse the phase of the wet return. The first one is easier. The second one is probably better because it doesn't change the phase of your primary signal. However, that would only matter if you were running other more complicated parallel loops at the same time.
Anyway, here's how I would do it.
Man, you are a quick drawer. One could save R6 an C3 though with the phase switch in front of them :-)
Quote from: madbean on November 07, 2018, 04:56:35 PM
Quote from: benny_profane on November 07, 2018, 04:47:03 PM
Would the phase reverse option simply be a selectable op amp stage on the wet return?
Two ways you can approach it - reverse the phase of the guitar signal or reverse the phase of the wet return. The first one is easier. The second one is probably better because it doesn't change the phase of your primary signal. However, that would only matter if you were running other more complicated parallel loops at the same time.
Anyway, here's how I would do it.
Forget my last post... Stupid me...
Very cool! Thanks for the quick responses here! This has always been an interest of mine so it's nice to see ideas floating around.
Quote from: madbean on November 07, 2018, 04:56:35 PM
[...]Anyway, here's how I would do it.
(http://www.madbeanpedals.com/forum/index.php?action=dlattach;topic=28875.0;attach=19390)
Not sure, if I fully understood this: spillover mode, bypass, lower Opamp used... How does the spillover from the return jack get to the output jack?
In that context, you can achieve the phase reversal even with only half an opamp: check this : http://geofex.com/Article_Folders/polarity_reverser/polarity_reverser.htm
MA: i know you're in the midsts of a board overhaul/redesign but did you make any decisions with this design? Definitely interested to see what comes out of this.
Quote from: Marshall Arts on November 07, 2018, 07:33:14 PM
Quote from: madbean on November 07, 2018, 04:56:35 PM
[...]Anyway, here's how I would do it.
(http://www.madbeanpedals.com/forum/index.php?action=dlattach;topic=28875.0;attach=19390)
Not sure, if I fully understood this: spillover mode, bypass, lower Opamp used... How does the spillover from the return jack get to the output jack?
The reason I used both halves of the op-amp is for lower noise. When you use one op-amp and select between inverted and non-inverted side, you have to use much higher resistor values on the non-inverted side to have a properly high input impedance. Using the first stage as a buffer let's you make those resistors 10k instead of 220k or 470k. So, there's not much advantage the other way unless you really want to use as few parts as possible.
The way the spillover gets to the output jack is through mixing the effect output with the clean buffered guitar, both of which go to the output jacks in spillover mode. Setting up the return signal with the mix pot accomplishes two things: give you a level control for the spillover volume and it acts as a pulldown resistor for C2 and C3.
Thanks, Brian. But shouldn't Swirch TB/Tails Lug B2 be connected to Switch Phase, Lug 2?
Quote from: Marshall Arts on November 12, 2018, 05:16:39 AM
Thanks, Brian. But shouldn't Swirch TB/Tails Lug B2 be connected to Switch Phase, Lug 2?
Yup, you are absolutely right. Funny how another pair of eyes can make all the difference! Now that I look at it again, I see another problem. The Mix pot is a load on the RTN path when the spillover mode
isn't used and would act as a second volume pot. So, maybe it would be better to keep that as just a pulldown resistor. I got too busy being clever to see what I was doing :) Replacing that with a 1M resistor would ensure it has no effect on the RTN output.
You could get fancier and move the phase switch to the RTN path instead and then take advantage of the low impedance output of the effect to do the phase reversal like you mentioned. I think like the example below. Also, when the mix knob is off you have just buffered guitar signal in tails bypass.
(Sorry if I'm taking over the thread. This kind of interests me!)
That's it. I would breadboard that, omitting the potentiometer. It's great, that you are taking this over, I was hoping for that, actually. We all can learn a lot from you ;-). Furthermore, this is YOUR forum ;-)
BTW - It must have been crazy early in the US when you replied. Do you sleep... at all?
Quote from: Marshall Arts on November 12, 2018, 11:04:14 AM
That's it. I would breadboard that, omitting the potentiometer. It's great, that you are taking this over, I was hoping for that, actually. We all can learn a lot from you ;-). Furthermore, this is YOUR forum ;-)
BTW - It must have been crazy early in the US when you replied. Do you sleep... at all?
My daughter had to leave for a school trip at 5:30am this morning so I was up pretty early. Not sure what to do with her gone a whole week!
Quote from: madbean on November 12, 2018, 12:59:15 PM
My daughter had to leave for a school trip at 5:30am this morning so I was up pretty early. Not sure what to do with her gone a whole week!
I know how that feels ;-)
Anyway, I thought about how to embedd that in the redesign of my looper and thought it would be fancy, if the complete circuit resides on a small daughter board held in place with the jack on the bottom side of the PCB. This would save some space on the main relay board, which I could really use... All the supply voltages are available within the looper, so no need for extra components for power supply on this board (just some more offboard wiring...)
(https://i.postimg.cc/y6pMVVBF/181112-Layout-Variable-Bypass.jpg)
(https://i.postimg.cc/C52XJ4Jv/181112-Schematic-Variable-Bypass.jpg)
I will add a second gnd pad for the I/O, as I will use shielded wire for that.
Quote from: Marshall Arts on November 12, 2018, 03:44:01 PM
Quote from: madbean on November 12, 2018, 12:59:15 PM
My daughter had to leave for a school trip at 5:30am this morning so I was up pretty early. Not sure what to do with her gone a whole week!
I know how that feels ;-)
Anyway, I thought about how to embedd that in the redesign of my looper and thought it would be fancy, if the complete circuit resides on a small daughter board held in place with the jack on the bottom side of the PCB. This would save some space on the main relay board, which I could really use... All the supply voltages are available within the looper, so no need for extra components for power supply on this board (just some more offboard wiring...)
(https://i.postimg.cc/y6pMVVBF/181112-Layout-Variable-Bypass.jpg)
(https://i.postimg.cc/C52XJ4Jv/181112-Schematic-Variable-Bypass.jpg)
I will add a second gnd pad for the I/O, as I will use shielded wire for that.
Were you planning on fabbing the board? If so, I'd be very interested in getting ahold of one if possible.
I'm kinda doing the same thing but as an add on for existing 3PDT switches or relay bypass. I don't work well with a breadboard so I may just do a test shot as an etch.
Brian, how are you getting those 5-sided tabs (ie: VB). I thought I had it figured out within Eagle but was wrong.
Yes, I will get them fabbed - will just take some months, as I will order that with the yet-to-be-designed looper boards.
On Brian's drawings... I would mine to look like his drawings (I like the B/W style), so I'd be interested in some information about that as well...
Quote from: Marshall Arts on November 12, 2018, 04:24:32 PM
Yes, I will get them fabbed - will just take some months, as I will order that with the yet-to-be-designed looper boards.
On Brian's drawings... I would mine to look like his drawings (I like the B/W style), so I'd be interested in some information about that as well...
Very cool. Keep me updated if you don't mind?
It looks like Brian's drawings are done in Visio (or something similar) if I'm not mistaken.
I will keep everyone posted about the development in the looper thread, so it would be great, if you could subscribe to that ;-). You can find it here:
http://www.madbeanpedals.com/forum/index.php?topic=28843.0
The Looper-Project will be create some sub-topics - technical discussions about specific issues (like this one here).
I will post these in upcoming, topic-specific threads (rather than bury them in the looper thread), as the discussion will hopefully be fruitful for other projects as well.
Quote from: gordo on November 12, 2018, 03:53:44 PM
Brian, how are you getting those 5-sided tabs (ie: VB). I thought I had it figured out within Eagle but was wrong.
Those are labels for nets. Just draw a net, and click the label button. Then you can choose the little flag (Xref On button) and name the net you want to connect it to.
For the schematics, I just export them as monochrome @ 300DPI then re-size them in Photoshop.
Cheers! Sounds good, thanks!
Awesome!!! Look what I can do!!!!
(https://i.postimg.cc/sXkXMbZm/181112-Spillover-Board.jpg)
So, today I finally had the chance to try different versions on breadboard. It's complicated, I must say...
To get everybody back on track, the last version was this one here, but replacing the mix pot with a 1M Pulldown resistor.
(http://www.madbeanpedals.com/forum/index.php?action=dlattach;topic=28875.0;attach=19414)
So I breadboarded this:
(https://i.postimg.cc/MGSLJ10b/Version1.jpg)
Problem with that version was, that I got a muddy signal in bypass mode. I recognized, that I should add a resistor (R16 in the image below).
(https://i.postimg.cc/SKPzzfZc/Version2.jpg)
Still, the signal was impacted in bypass mode, I guess that the output signal's connection to Vb (top right) and the return jack of the delay has some impact on the output signal (even though it's not that bad, but noticable). The phase switch had no noticable impact on the signal (I would have assumed, that if you reverse the phase of the the spillovers it would cancel out the sound of the guitar in bypass mode to some degree in one position).
So I decided to try another approach and skipped the idea of the phase reverse option: Here it is with inverting op-amps:
(https://i.postimg.cc/rpsrmP98/Version3.jpg)
This one works pretty nice, with R11 it is easy to adjust the volume of the trails in (buffered) bypass mode - 100 k surprisingly created louder trails in bypass than in fx mode. It is a little bit noisy though, so I will try other resistors for R10, R18, R11, R13, R15. Any recommendations?
I tried all possible versions I can think of (I think). Oh how I sometimes wished a breadboard had an "undo" function ;-).
So, this is the version that to my ears has the least impact to the sound:
(https://i.postimg.cc/xjZfK6vp/Version5.jpg)
It has
- an inverting stage as buffer and
- a second inverting stage as a summing amplifier, which is similar to Jack Orman's simple mixer:
(http://www.muzique.com/schem/mixer.gif)
I did not fully understand, why an inverting amplifier is better for summing signals, but my tests tell me that it's true.
I skipped the idea of a phase reversal switch, which is probably overkill (at least with my delays, I did not find one that asked for that feature). I guess that most splitters (for parallel fx-loops) will provide a phase switch anyway, so I leave that up to these devices ;-).
Signal will be always in phase (True bypass, fx - if the delay does not flip the phase, buffered bypass, as it is inverted twice).
Remaining questions for me (more of a general nature...)
- The circuit does not work well without R15 (please tell me why?!)
Input and output coupling caps (C1, C7, C6, C5):
- Do the values in a circuit like this one matter at all? Or could I take 100n non-polarized caps for everything?
- Does the orientation of the polarized electrolytics matter at all (I dont think so, as they pass the AC-signal, but it would be great if somebody could confirm this for me)
If somebody could answer these, I would be really thankful.
Old thread, but things take time. Here is a simplified picture of ONLY the buffered bypass path. It works great in a standalone setup (except for very little additional op-amp noise), unfortunately, I am using this in a microprocessor controlled looper/switcher (http://www.madbeanpedals.com/forum/index.php?topic=28843.0). In this setup, it produces some switching noise, which is ok for most setups but can be annoying on higher gain and delay mix settings. I guess, that the switching noise is fed back to the audio path through the op amps via gnd or power, as the microcontroller also switches a lot of LEDs and relays on during program change (and that creates a peak in the power supply). It's not bad enough for me to worry about it (the looper can be switchted to true bypass which works flawless), I am just posting this for information.... unless somebody finds another bug in the schematic ;-)
(https://i.postimg.cc/MTGqwN7d/190223-Spillover-Board-madbean-review.jpg)
Not getting an image on this. I'm on my laptop so will see if I have similar issues at home (mind bending security here at work)