Saturday 18 February 2012

Split 'n' Blend

I like these little "utility boxes"



(Please take note! This circuit will not work as a blend for pedals that invert the signal polarity (or phase). For a few examples of such pedals; the likes of Dyna Compressor/Ross Comp and their derivatives and Marshall Bluesbreaker among others. The symptom of inverted signal polarity (phase) will be significant drop in level when the blend pot is set near its middle position. This is caused by two polarity phases cancelling each other out. If you need a way to control the phase, then you need to address this accordingly.)

166 comments:

  1. Could you advise me on the what caps I should use ? Im not sure what those yellow ones are.
    Overall it looks just like what I need for my setup .

    ReplyDelete
  2. Any caps with a 5mm pitch would be perfect, or even a 2.5mm pitch and you can bend the legs out. You can get a kit of useful values from here:

    http://www.bitsbox.co.uk/polyfilm.html

    ReplyDelete
    Replies
    1. thanx 4 that quick response ! as you can probably tell this is my introduction into the world of home builds :)
      Caps in particular are so varied in appearance that I thought it best to identify your choice . Ive noticed that some people stick some massive looking caps into tiny pedal circuits but is the main choice here fundimentaly one of size or audio effect ? so many caps to choose from all look so different , Values being the same .

      Delete
    2. Don't worry so much about the colours or dimensions, the vast majority of the time you would choose small and compact caps. With that in mind, for picofarad values, get yourself a collection of ceramics, nanofarad a selection of polyester film (mylar) and everything above 1uF look at electrolytics. You can get all of these is value packs from Futurlec here:

      http://www.futurlec.com/ValuePacks.shtml

      Which is a good way to start your stash for not a lot of cash.

      Delete
    3. Youve been a great help Ivlark .

      Delete
    4. Hi again lvlark .
      Ive been having some fun with this circuit for some time now ...is there any way I can bring the gain up a bit by chainging a resistor value somewhere ? The only reason I ask is because I added a bypass swich which has revealed a slight drop in signal when the circuit is engaged . If yours works at unity gain what should I be looking at chainging ?

      Delete
    5. The JFET's in this are source followers which are at most unity gain. If you've lost some signal you'll need some sort of gain recover stage like a clean boost (take your pick) to get it back up. If you like using JFETs, then maybe something like the AMZ JFET booster:

      http://www.muzique.com/tech/bipolar-j.htm

      Delete
    6. OK in that case I think I know what I have to do with a gain stage between (in) and (board) also i'd like to bring down the dry signal a notch before the pot because dry is slightly louder than wet (maybe this is because of the cheap parts i was using ?!)

      Delete
    7. what are the 3 yellow 100n?

      Delete
  3. How would I wire this with a 3PDT switch?

    ReplyDelete
    Replies
    1. Well the send and return goes to the external effect so you don't have to worry about that. If you switch the input and output wires as with any other effect then you bypass the whole thing, like this:

      http://1.bp.blogspot.com/-k5Sh68yVU18/TzGRFKbiALI/AAAAAAAAAlk/CbfaaduUjYQ/s1600/!Offboard+wiring.png

      Delete
    2. thanks for the quick response! I'm new at this so bare with me hear but I just wanted to make sure, for the jacks would I be soldering to the tip or the sleeve?

      Delete
    3. The wires from the boards go to the tips, then all the sleeves are connected up together with all other points of ground.

      Delete
  4. verified. if not already..

    ReplyDelete
  5. what's that for??

    ReplyDelete
    Replies
    1. Putting an effect in parallel with your dry signal.

      Delete
  6. I thought about something like this yesterday. I don't see any reason why this couldn't be used with analog spring reverb tank.. But i guess it would need pre and post gain stages to recover levels after the tank. So 2 LPB1 circuits in series with send and return wires could make this analog spring reverb..

    There are a few projects still going on, but i'll promise to take this "slight mod" on before fall. I'm visioning a 19'' rack unit with two or three tanks switchable :)
    +m

    ReplyDelete
  7. just built this one and it seems to work well but loses a little level and mids/mid bass when used with my P bass. also didn't work all that well with my lovepedal silicon fuzzmaster since it actually changed the sound of the pedal due it it's lower output impedance, but IME,fuzztone/overdrives are extremely sensitive as to what level and impedance that their seeing at the input.

    ReplyDelete
  8. We don't need to ground the other open terminals of the In/Out/Send/Return sockets? or we leave it like this only?

    ReplyDelete
  9. think this would work well with the ts808 to be able to blend dry/wet mix?
    it's not supposed to change the sound really, or?

    ReplyDelete
    Replies
    1. also, what are the 245, are they A or C?

      Delete
    2. They're just unity gain buffers so it won't really matter, justuse whicheverones you can get cheaapest! :o)

      Delete
  10. I made this today and I also found a big volume drop in the center position.. It's a shame as I would of quite liked it otherwise.

    ReplyDelete
    Replies
    1. I'm pretty sure that one buffer before the out and one before the return will solve this.

      Delete
  11. i've built 6 of these now. one as a stand alone unit, and the others built-in to some of my pedals. i've used J201's, 2n5457's and BF245C's and all of those transistors sound great in them (but all of them have a slightly different tone). IME, out of the three, BF245C's seem to add a little warmth and mid bass the most, and the 2N5457's are the most transparent, but the J201's are just as nice in their own way.

    ReplyDelete
  12. This comment has been removed by the author.

    ReplyDelete
  13. Just built one and it works great. I get unity volume right through the entire throw of the pot too. I used 2n5458s.

    http://i203.photobucket.com/albums/aa227/axe_34/IMG_6081_zpsc82d46b5.jpg

    http://i203.photobucket.com/albums/aa227/axe_34/IMG_6082_zps412ac1e7.jpg

    ReplyDelete
    Replies
    1. which paint do you use? And how do you do those black letters?

      Delete
  14. Would it make sense to build it in a wah/volume pedale enclosure to use it to smoothly fade in and out effects that are in the loop ?

    ReplyDelete
  15. Could this be used to blend to signals from a stereo bass? As in only connecting cables to input, return and output.

    Thanks.

    ReplyDelete
  16. If I want to build this into a box together wih an effect, do I just connect in and out to the jacks and send to input of the effect and return to the output? Seems to be a useful feature to lots of pedals.

    ReplyDelete
  17. Hi, does Split'n'Blend disconnect loop when the effects inside loop are nonactive?

    What happend when effects inside loop are active, do i hear a clean guitar with sound from loop or just from loop?

    Sorry for my english...

    ReplyDelete
    Replies
    1. No. Depends on the setting of the "blend pot", it blends from the (now buffered) input signal to the loop signal. If the pot is set at noon you hear the input signal and the loop signal....

      Delete
  18. I built this a while back, and it works great, but it really darkens my signal. I used j201's for my fets, but everything else is whats on the layout. Is there any way to preserve the high end? I put a boost circuit in front of it and turned it into a dual pedal, which slightly helps the problem.

    ReplyDelete
  19. I'm confused (big surprise). I'd like to add this to the DBA Robot. Do I have this right?

    IN --> 3PDT IN Lug
    OUT --> 3PDT OUT Lug
    SEND --> FX INPUT ???
    RECEIVE --> FX OUTPUT ???

    ReplyDelete
    Replies
    1. You got it, that's right. You may want to put a boost circuit like a LPB in front of it to preserve the high end a bit more. I use it right now for my bass rig to blend in effects, and preserve the low end.

      Delete
    2. I am here because I had this exact idea. did it work?!

      Delete
  20. I've got plenty of 25 k Pots but no 100k pots.
    Can I just use those instead without altering the circuit?

    ReplyDelete
    Replies
    1. There's no reason you can't use 25k pots for this, I'd stock up on 100k pots though, almost every circuit uses one.

      Delete
    2. Yeah I know and I already ordered a couple of them at Banzai but they seem to have run out of em...

      Delete
    3. could you give me the full schematics of this blend ?
      Thanks

      Delete
  21. If I want to remove the pot from the vero and make it a 50/50 blend, can I just skip the pot and wire the return/output together or do I need mixing resistors? Two 50k resistors in this case?

    ReplyDelete
    Replies
    1. I just used an internal trimpot to blend the signal to 50/50, which works fine. I'm getting a little noise from this circuit though. I built it in an A enclosure. The noise is only there when daisy-chaining the effect, when I use an isolated ground from a pedal powersupply it's quiet. Does anyone know what the problem could be?

      Delete
    2. noise bleeding from another effect pedal through the non-isolated daisy-chain? just an idea, i've read it in some other places :D

      Delete
  22. this thing is really amazing, you can easily use an LPB one between the board and the output to mix 3, to boost your signal to your needs, if your effect is too strong :) it just happens to need an isolated power supply because the external effect bleeds through the power feeds into the mix :)

    ReplyDelete
  23. Hi all. I've built this and am not getting any signal. I've been over the layout quite a few times. I tried all the usual suspects, swapped out the bf245s a few times, and checked for solder bridges. I installed it in an enclosure with a rat, and put the send to input of the rat, and the return to the output of the rat coming from the volume pot. What could it possibly be? I'm stumped.

    ReplyDelete
    Replies
    1. What are the transistor voltages?

      Delete
    2. on Q1: G=8.68, S=8.62, D=9.2, Q2: G=8.68, S=8.62, D=9.45

      Delete
    3. There's little to no current flowing, thus the high voltages. Check all the ground/earth connections for starters

      G

      Delete
    4. checked for cold solder joints/tiny bridges.nothing. next idea?

      Delete
  24. This comment has been removed by the author.

    ReplyDelete
    Replies
    1. This comment has been removed by the author.

      Delete
  25. Is there any way of adding power supply filtering to this circuit to remove the interference caused by other daisy chained effects - or must it have it's own P/S?

    ReplyDelete
  26. I have a Line6 M5 stomp box modeler that I mostly use for time-based effects, and wanted to keep my core signal intact. I built it using MPF102 JFets I bought t the local radio shack. First off I made the rookie mistake of using a metal DC power jack. Since I didn't have any plastic ones, I put a switch in the hole and hung a battery clip off it. I'll get around to ordering a proper jack soon.

    First impressions: I'm stoked! Exactly what I wanted, and I can't believe how much more "pro" my rig sounds just with keeping my pure signal out of any A/D or D/A converters. My only complaint is the volume drop in the middle. When I engage the M5 I notice it right away, and when I bypass I get my signal back. So that's a little weird. Hopefully adding a recovery buffer will help. I'm also going to add a 3PDT switch to bypass the wet signal. I'll post a pic once I have my jack installed. Currently it looks a bit disheveled with the battery hanging out the back. Not bad for my first build all in all. :)

    ReplyDelete
    Replies
    1. I think that might be down to the source follower / buffer ..Ive found that simple neat circuits often perform slightly under unity including most buffer circuits . you can either incorporate a clean boost circuit in the box or stick one in the loop to bring things back up to unity . I dont think adding more buffers will cure it more likely to cause even more volume drop .

      Delete
  27. Neat circuit, very good for running two amps side-by-side. It doesn't sound exactly like I had imagined. There's a bit of volume drop on all settings which I don't like.

    I'm going to tweak it to get unity gain and then use it solely for running multiple amps.

    ReplyDelete
  28. Just finished this and it works beautifully. At least for pedals with their own volume control, easy to match the levels when you've found the right blend setting. For those of you trying it with 2n5457's, the layout has the transistors facing the "other" way (they're drawn to represent a different type). Newb mistake.

    ReplyDelete
  29. as in: a mistake that I made..

    ReplyDelete
  30. I'm building this with J-201's - am I right in thinking that I need to reverse the pinout relative to those in the schematic?

    ReplyDelete
  31. Will putting 10K trimpots instead of those 4k7 resistors...solve anything with that volume drop?
    I would like to build this for myself, because i HATE my 6534+'s serial loop...can't do anything with it...

    ReplyDelete
  32. a user on the madbeanpedals forum said that he used 2n5458 and had no volume drop in any position. worth a try i think

    ReplyDelete
  33. Hey!
    Could I use the yellow and grey wire without the pot for signal split? The input will be the "send" from another effect.

    ReplyDelete
  34. If I put delay pedal on send-return, is the signal parallel to input-output?

    ReplyDelete
    Replies
    1. Yes. So you could run a digital pedal 100% wet and the unconverted dry signal will route around the side in parallel with it. That is basically what I do with an Eventide Eclipse and TC G-Force in the loop of my amp but using a more advanced mixer.

      Delete
  35. Hi Would like to build this but with a true bypass switch Can anyone see a problem with this layout?
    http://i912.photobucket.com/albums/ac323/Bat_fastard/splitblendwithtruebypass_zps449b7e14.png

    ReplyDelete
  36. So, is that correct? Split 'n' Blend with 3PDT?
    http://s29.postimg.org/72qigfirb/Sn_B_3_PDT.jpg

    ReplyDelete
    Replies
    1. One more question - what is the right position of a S'n'B? Version1 or Version2?
      http://s9.postimg.org/xxe6sxflr/ver1.jpg
      http://s21.postimg.org/54q0aufev/ver2.jpg

      Delete
    2. Either way is fine, it just depends on the usual positioning of the effects in your chain. If you have overdrive, distortion or fuzz effects permanently in your chain, and want to put a delay pedal in the blend loop, then use version 2. If you have a delay pedal permanently in your chain and want overdrive, distortion or fuzz effects in the blend loop, then use version 1. Or you can have permanent chain effects before AND after if you want.

      Tell me which pedals you're using, which you want to blend, and I can tell you the way I would do it.

      Delete
  37. Hello,
    I want to had a serial/parallel footswitch but I can't figure how to do it...
    Please, can someone help me?
    Thx

    ReplyDelete
    Replies
    1. I'm still working on the Serial/Parallel Switch.
      Do you think this can work?
      http://www.hostingpics.net/viewer.php?id=561282photo.jpg
      Thank You!

      Delete
  38. Hi, I want to mod this slightly to use in my bass rig. I want to be able to have a 100% dry signal and the use the pot to blend in the loop over the top and keep the dry 100%. Is this as simple as making a link from the buffer just before the pot to the out jack, or can I bypass the buffer and link the in to the out jack?

    ReplyDelete
    Replies
    1. I'd re-jig it slightly. Take the light blue wire from the board and lug 2 straight to the output jack, take Return to lug 3, and ground lug 1. Then your dry signal will be buffered and taken straight to the output so unaffected by the pot, and the pot will be acting as a volume control over the effect in the loop and mixing the level you choose back in with your buffered dry signal. So in the far clockwise position the full effect signal is passed to the output jack, in the far counter clockwise position the entire wet signal is dumped to ground.

      That would be a better way of doing it, as the blend control as it is won't silence the wet signal, just put 100K of resistance between it and the output.

      You can't bypass the buffer. If you do there is nothing to split the signal in the first place. If you mean could you still use the split but daisy chain out of the input straight to the output instead of passing the dry signal through the buffer and coming out of the light blue wire, then it really wouldn't make much sense. You'd get none of the advantages provided by the buffer and load down your signal immediately with the connections and passive components hanging off it, without having the benefit of anything to prevent that happening. It's a bit like buying a car and asking if you can set it going and walk alongside it :o)

      Delete
    2. That makes a lot more sense cheers, and thanks for the super fast reply!!

      Delete
    3. Hi, I have one problem that I can't get my head around. After wiring the blend up as your describe I have found that the pot is turning off the buffered signal as well as the loop signal when turned all the way off. Then when you begin to blend in the loop, the buffered signal cuts back and is then unaffected by the pot. Any ideas?

      Delete
    4. It would appear that when the return is dumping to ground so is the output which is connected to the buffered signal. But only when in the clockwise position. It is also causing some volume loss in the buffered signal.

      Delete
    5. Hi, any more advice anyone can give me on this? thanks

      Delete
    6. I've wired mine as mentioned above and I'm having the same issue with the pot cutting both signals out when turned completely off. Any ideas?

      Delete
  39. where is the best place to add a booster? last, after output? and with 2x3pdt so you can run it anyway you want it (4 ways, incl full by pass)

    ReplyDelete
  40. Would it be possible to introduce a footswitch before the pot to give the option of two pots and therefore two settings so that I could leave my Holy Grail on full and have it awash with reverb for single notes replacing chords but instantly tame it where clarity is more important? I think it's one knob is a blend anyway because on full you lose your dry signal.

    Would it need to be a 3PDT switch?



    ReplyDelete

  41. Hi

    Sorry to pester with the same/similar question. I've been playing with this circuit and have come to a halt.

    I'm trying to create a blend pedal where you can set two levels of blend and switch between them via a 3PDT and have wired that circuit into a standard 3PDT true bypass wiring.

    I've connected a 3PDT switch between two identical 100k lin pots as follows:

    3PDTSwitch - - -
    3 2 1
    - - -
    Centre switch terminal 1: connects Lug 1 on the pots to the tip of the return Jack
    Centre switch terminal 2: connects Lug 2 on the pots to the terminal on the bypass switch that is normally used for the output of the circuit board.
    Centre switch terminal 3: connects Lug 3 on the pots to the grey wire on the diagram (board output, I'm assuming)

    From then on I've wired the Bypass switch as usual and made a common ground strip of stripboard connecting the grounds from all four rings of the jacks, the board and the power socket.

    Connected up, I get the bypass signal but nothing when the bypass switch is engaged.

    Tests:

    I've followed the wiring from each lug of both pots, through the pot-select switch to their intended terminal points with a multimeter

    I've tested the circuit with one pot on a breadboard and it works

    I've followed the connections for the bypass switch with a multimeter and and they also seem sound.

    I've tried it with two pedals:

    1. Behringer digital delay - this pedal wouldn't come on as though it wasn't connected from each end and needs to be before switching on

    2. EHX Holy Grail this will switch on regardless of connection and while it worked on the breadboard pretty perfectly, in the whole circuit, nothing.

    I do intend to breadboard everything (I was overzealous and wired an idea up without testing it) but just wanted to check if there was anything I was missing or more likely assuming as I'm relatively inexperienced (built about four pedals).

    I appreciate that this isn't a querey directly about this circuit but any advice would be really welcome.

    ReplyDelete
  42. I added this in to the Keeley 4-knob, and am having an issue when the Blend knob is in the middle - I have barely any output at all. When fully CCW I get the dry signal, and when fully CW I'm getting the signal with the compressor (at least it appears I am).

    I have the input wire coming from the input lug of the 3PDT, the send wire going to the input of the comp, the comp's output (lug 2) to lug 1 on the blend pot, and then lug 2 on the blend pot to the 3PDT switch output lug. I've triple checked the layout and looked for solder issues and everything looks fine, should it be canceling out signal in the middle position or is there another way I should be wiring this up?

    Thank you in advance!

    ReplyDelete
    Replies
    1. These two circuits won't work properly, since the compression is acting as an inverting amplifer and the split/blend isn't. This means that the middle way of your blend pot is cancelling both signals out with the phase of each signal being in reverse. You'll get similar results with Marshall Guv'nor and Split/Blend if you try it. One way to address this would be to have inverting unity amplifier after the return jack and the blend pot's lug 1.

      But true. As is, this split/blend won't work with effects that invert the signal.
      +m

      Delete
    2. This comment has been removed by the author.

      Delete
    3. Dang, I was really hoping this would work to run parallel compression a la the Deep Six Voyager or Wampler Ego. Would it essentially have the same effect as that if I rewired it as Ivlark mentioned above:

      "Take the light blue wire from the board and lug 2 straight to the output jack, take Return to lug 3, and ground lug 1. Then your dry signal will be buffered and taken straight to the output so unaffected by the pot, and the pot will be acting as a volume control over the effect in the loop and mixing the level you choose back in with your buffered dry signal. So in the far clockwise position the full effect signal is passed to the output jack, in the far counter clockwise position the entire wet signal is dumped to ground."

      Although, with this way the uncompressed signal would still be passing through even if the blend was all the way up though, correct? Or would there still be a phase issue and experience cancelation?

      Thanks!

      Delete
  43. Would using a lower resistance pot help with some of the tone drain at center position?

    ReplyDelete
  44. would i be okay using a couple of 2n5088s

    ReplyDelete
    Replies
    1. No, because 2n5088's are not FETs. I would highly suggest using 2n5458's if you build this. 2n5457's have a volume loss when the pot is in the middle, but the 2n5458's do not.

      Delete
  45. i use 2n5458

    https://www.youtube.com/watch?v=9VwiAHGqpuQ

    ReplyDelete
  46. built it, very cool.

    however, a question: i expected no sound with nothing plugged in it and in the 100% wet setting, but if there's no effect in the loop (or even a cable in the send but nothing in the return) it is more or less as loud as the bypassed sound in any setting of the pot. is this normal?

    ReplyDelete
  47. So this won't work with the keeley? I only ask because lvark mentions it in the comments of that pedal. I'd love to add a blend knob to my 4 knob.

    ReplyDelete
    Replies
    1. I will not. I think miro posted an explanation as to why it won't. It's due to the type of compressor it is, ie a Ross style compressor. If you search the forums Gieri posted a layout for a blend that will work with Ross style compressors.

      Delete
  48. I'm building a ROG Ginger on Perfboard which worked perfectly, and I built it into an enclosure with the Split n Blend before it in the same enclosure. It's working, but not properly - for some reason the Bypass doesn't work. I shoulda tested it out of the box more... ugh.


    Anyway.

    Any ideas why the Bypass wouldn't work? I did use an old 3PDT that is honestly probably faulty, but I was thinking my wiring might be bad. I've never had this issue before though...

    Also, when I turn the Split n Blend pot to full clean signal, I get quite a lot of effect bleed. I was thinking this may be due to the fact the Ginger is very loud, but unsure if external issues could be contributing to this. Clues would be great :)))

    ReplyDelete
    Replies
    1. DUH, fixed my problem. Circuit input and board input wires were mixed up.

      Had a phasing issue in the 50/50 section using 2N5457s, but I had a friend make up a small circuit to flip the phase of the effect output. This used a couple of caps, resistors, and one 2N5088. Works a charm. :)

      Delete
    2. Any chance you could send me the circuit he used to flip the phase ? :)

      Delete
    3. I'd be interested in this too

      Delete
  49. Quick general question related to splitters/blenders. Basically I can split the input signal in two directions as long as I put a buffer onto each of those to bump it back up again?

    ReplyDelete
    Replies
    1. Basically. That's pretty much what this circuit is.

      Delete
    2. Thanks Charles. Yeah I got the impression that's all this was doing. I was asking because I actually have two Klone buffers veros sitting here unused and was thinking based on this layout I might be able to put them to use in a similar manner. Might build one of these as well and see how they both go.

      Delete
  50. I've just built this one and I'm getting this weird issue:
    With the pot all the way down I'm getting 100% wet signal, with the pot all the way the other side, it seems that the signal is fed back into the effect over and over. I guess I probably wired something wrong... any hint? :)

    ReplyDelete
    Replies
    1. I guess make sure that you've wired the sends and returns correctly?

      Delete
    2. Send and return have been wired as shown on the schematic (Tip going to the board).
      I checked the board a few times for mistakes but could not find anything...

      Delete
  51. Hi,All!
    As I've understood, there should be out-of phase issues when you use this circuit with pedals like woolly mammoth,

    or other, that inverts signal. Will jfet buffer like one of those

    http://tagboardeffects.blogspot.lt/2014/08/buffers.html resolve the problem? If yes, which one and where to inset

    it? Thx.

    ReplyDelete
  52. in this layout the electrolytic cap is red.
    isn't it aluminum E cap?

    ReplyDelete
    Replies
    1. Yeah it's just a regular electrolytic cap. I think it's just an older graphics pack on the veroboard layout program :)

      Delete
  53. I've got a question about power. Im building this into the same box as a rat, will I be able to power both boards with just 9V eg, by linking the 9V rails on the two boards, or will I need 2x9V (let's say two batteries) with each one powering its own board?

    ReplyDelete
  54. Friends, I just finished that thing, but got a problem. I have dry sound when pot is all way down, but when i turn it up signal disappears. It becomes even if I use just a short cable from send to return. I turn pot all way up - no sound at all. What'd I do wrong?

    ReplyDelete
  55. Hey guys. I have a little problem with this one.
    If i turn my pot up and down i'm always getting 95% clean and 5% effect level or the other way around on max settings.
    Is it normal or is there a way to modify that?
    Useful Information:
    I used 2N5485 instead of 2N5458 mistakenly but it does the same with J201.
    My 100k pot only reads 90k.
    thanks.

    ReplyDelete
    Replies
    1. I'm getting the exact same issue, I added this circuit to a musket fuzz clone. Any help here?

      Delete
    2. I'm getting the exact same issue, I added this circuit to a musket fuzz clone. Any help here?

      Delete
    3. Same here. When it should be totally clean I still have a whisper of fuzz pedal.

      Delete
    4. Impedance mismatch. For starters, try 500K pot.
      +m

      Delete
  56. How would I do this with separate volume controls for each part of the signal? I seem to pretty much be able to only get 100% wet or dry with almost no blend effect. Also on any setting I have a huge treble increase and feedback in the wet signal. Makes my rat sound like a metal zone.

    ReplyDelete
    Replies
    1. Also I have tried it with a clean boost circuit in front, after , and in between, to little or no avail. Could I use any old buffers for this? I am partial to trying a tl072 instead of the two fets

      Delete
  57. Hi, I finished a little angel from here, and I added this split n' blend in the same box, using the 2N5458's which some reported not having volume drop problem, but I'm still having it.
    It's a shame because it makes the split n' blend pointless.
    Any help with this?
    Thanks.

    ReplyDelete
  58. Anyone know whether a pair of 2n5952s would be OK in this circuit? I'm putting one together and have a spare pair from when I made a Phase 45 (they're not matching, hence they're not in the Phase 45).

    ReplyDelete
  59. Turns out that the 2n5952s work really well in this application. I also changed the 2 100nf caps at the Source of the JFETS (the ones going to the 2 4k7Ω resistors)to 1uf (non polarised) to let all of the low end through (I'm a bassist) and used a 4M7Ω for the input resistor as I'd run out of 2M2, but realistically, anything over 1M is good there.

    ReplyDelete
  60. Built this according to the layout but I was not pleased by the way the pot acted. I prefer the dry signal to go through all the way always and only mix in the wet. So, board output to lug 2 AND output jack. Then "return" to lug 3. DO NOT attach ground to lug 1, this will, well, ground your output. Also switched all caps to 330n for bass use,

    ReplyDelete
    Replies
    1. believe it or not you don't need to change the input and output caps on this for bass, the stock 100nF is fine and you're bypass signal and engaged dry signal will be exactly the same

      Delete
  61. Built today with a pair of 2N5457's (turned around to keep the pin out correct)

    As many have said, if there is a HUGE volume drop/the bass disappears in the middle position then, chances are, the pedal in the send/return loop is out of phase of the clean signal. I had this issue on some pedal but not others (as expected). To fix this I added an LPB1 circuit between 'return' and the pot on a DPDT switch (the LPB1 inverts the phase) This way, it the volume drops massively in the middle position, flick the switch putting the LPB1 into the wet signal path et voilà!

    ReplyDelete
    Replies
    1. hey man. even easier, just use 2N5458's instead of 2N5457's and you'll have no volume drop in the middle.

      Delete
    2. Nope. You are both wrong. First, the LPB does not invert phase. The volume drop you are experiencing is due to pedal in the loop inverting the phase. If LPB was to invert phase, the drop in middle position (where positive and negative phase cancel each other out - causing the drop in level) would not be present. And no, different JFET type will not affect the phase cancelling at all. Those FETs are just buffers to decouple/couple parts of the signal and act as impedance transformers. As buffers are, you could stick any NPN/N-Channel BJT/JFET/MOSFET in there and there would be no audible difference whatsoever.

      To cure the issue, you'd want to use insulated jack for the return and use DPDT switch to simply swap the + and - leads. Or just not use this particular circuit to blend pedals that invert phase. I do recall i'd drawn a layout for a blender with phase switch, but don't remember where that would be.

      This will get the job done: http://tagboardeffects.blogspot.fi/2015/01/hpf-technology-hpf-pre-2.html
      Just build that between the return jack and split board return wire. Plus you'll get a rather sharp low frequency cutter which may come in handy.

      Either way - Sorry, but both of your solutions are wrong and do not address the issue as all.
      +m

      Delete
    3. Hey Miro. The first time I built one of these years ago in a rat I had a volume drop when the pot was in the middle, but when I switched the FETs to 2N5458 or could be 2N5457 I'll have to check the volume drop disappeared and there was smoother transition and blend. Maybe there was something else going on before the swap, but I do remember that the swap fixed the issues cause I ordered around 500 of the same FETs mostly for use in this blend. Fucking odd.

      Delete
  62. I added a note to the post about the phase inversion issue, as it seems that half of the comments are about this well known issue.
    +m

    ReplyDelete
  63. The issue I had, and still have, is that a rat does not invert signal yet this circuit does not provide anything near a proper clean blend. My solution is to isolate the power supplies and outputs for each half with small 1:1 TRW brand audio transformers. Cleans it up a little but the fact the blend pot causes voltage division cannot be avoided. Still quite lossy. A separate volume control for each side fixes this, but requires the use of an lpb1 or other boost and output isolation to avoid feedback down the wet or dry signal chains. Not really fun after all is said and done.

    ReplyDelete
  64. (For some reason none of my browsers let me click on the "reply", so..)

    Zach: My bet is on that there was something else at play with yours. You could try on different FETs. By electronics theory there should no audible difference.

    apotheodaimon: Sounds like impedance mismatch issue to me. But this circuit will not be the best performing camper for many circuits. I know for a fact that it plays really well with fuzz face style circuits...
    +m

    ReplyDelete
    Replies
    1. When you're right you're right. Unfortunately, there's no way to know what was going on since it was so long ago. My attitude is if it works right now, it sounds good, and most importantly it works right.

      Delete
  65. Hey everybody! I have a Deluxe Electric Mistress that sucks tone, and I thought putting it in the loop of one of these might be a good idea. Any thoughts?

    Thanks!
    Chuck French

    ReplyDelete
  66. Would a 500k Lin pot work on this?
    Thanks,
    -Yari

    ReplyDelete
  67. This comment has been removed by the author.

    ReplyDelete
  68. I want to implement this into a xesign for eurorack, which uses +12v. Do I need to change resistor values?

    ReplyDelete
  69. Gudday all.

    I'm thinking about using this to run two dirt boxes in parallel. Any reason why I shouldn't?

    ReplyDelete
  70. Hi, does anyone have any idea if a similar blend knob could be done using a summing opamp ? Thanks

    ReplyDelete
    Replies
    1. Why not.
      http://tagboardeffects.blogspot.fi/2013/01/2-channel-mixer.html
      Dual linear pot wired in cross for the inputs. I'd pobably still add a dual buffer for the split. Then again, i'm not sure if the complexity will offer much more than the simple split/blend does.

      Also, none of these take the signal phase into account.
      +m

      Delete
    2. Thanks a lot for your comment! I had the same idea (to buffer the input in the case of loading), then again this split-n-blend worked fine so I might as well go with it :)

      Delete
  71. Quick question about the phase issue mentions in the post, would a dan armstrong green ringer octave be okay to use with this circuit? thanks

    ReplyDelete
  72. Would this work as a stereo summing unit (stereo to mono)? If so, how do I make it 3 jacks only (2 input, 1 output) and remove the pots?

    ReplyDelete
    Replies
    1. Hello Robin, Did you ever figure this out? I'm trying to make a 2 input 1 output pedal to mix guitar signals and this was suggested to me. I use a Morley FX Blender that is like this, I just don't use the Send so the Input and Receive are the two inputs. Thanks

      Delete
  73. Built this and work well, just the darn phase issue. Has anyone successfully added the inversion to this design?

    ReplyDelete
  74. Can anyone point us to an IC that isn't out of stock everywhere? Seems like all the JFETs listed in this schematic and comment section are obsolete.

    ReplyDelete
  75. Built this today, trying to add a clean blend to the X-fuzz for use with bass. Used J201s, didn't notice any volume drop in the middle.

    ReplyDelete
  76. So I'm looking to add this into a pedal build so I can blend an otherwise clean FX loop into a distorted tone. I'm a little bit confused as to WHERE to insert the pedal. The best I can figure is the grey wire would run out to the input of the distortion circuit, then the output of the distortion would run into the blend knob's pin 3 - would that give me what I'm expecting?

    ReplyDelete
  77. Hi there

    I built this circuit with 2N5458's and in the dry setting i'm still getting a small bit of bleed from the parallel circuit. Is this normal?

    ReplyDelete
    Replies
    1. Yep. That’s the nature of the beast.

      Delete
  78. Is there any reason not to bump up the 4k7 resistors? I'm getting a 7% voltage drop with them at 4700ohms but playing with a 100k pot there that drop gets smaller and smaller the higher the resistance. At 100k resistance it's down to a 1% drop.

    I wasn't getting the massive drop in the middle due to phase inversion like some, but I was getting a small drop that was still bothering me.

    I also noticed a difference in the drop between 2N5457s and this other random jfet I had lying around, a J310G. The J310G actually had a significantly smaller drop, though it was still a drop. I have no idea if the change in jfet or resistance value will affect the tone or longevity of the parts though.

    ReplyDelete
  79. Hey, i was thinking about to replace the potmeter for a CV expression pedal input. But how should i do that? Is that possible even?

    ReplyDelete
  80. Hi, has anyone tried it with line signals? I'd like to build one for my synths and drum machines, but I've heard that a FET based design can distort fairly easily... Thanks

    ReplyDelete
  81. BOA NOITE. COMO POSSO AGREGAR O CONTROLE DE BLEND NO DYNACOMP?
    GRATO.

    ReplyDelete
    Replies
    1. Agregar quer dizer aplicar este controle ao dynacomp? Para isso basta ligar SEND ao INPUT do Dynacomp, e RETURN ao OUTPUT do Dynacomp. O Input deste circuito será o cabo da guitarra, e o output será o cabo que vai para o Amp/próximo pedal.

      Delete
  82. What are the red squares supposed to represent? Am I to add a conductor to connect them??

    ReplyDelete
    Replies
    1. They're cuts in the track. Have a look at the Build Guide in the menu above, it covers these types of symbols

      Delete
  83. Hey! What is the correct way to use this with 2 pedals in parallel? (like blending 2 dirt pedals). I planned to put one on the loop and other one between light blue wire and blend pot 3. Is this right?

    ReplyDelete
  84. When you use the blend, does it go 50/50 in the middle and 100% of wet or clean on either side, OR does it give 100% clean signal and then introduces the wet effect 0-100% with the knob?

    ReplyDelete
  85. Would this work well with your 4 band gyrator EQ build?

    ReplyDelete