Wednesday 26 February 2014

Keeley Katana Boost Layout Rev 2

alphadog808 was having problems with his Katana Boost build and I noticed one of the cap values was incorrect.  I also noticed that I had no motivation to fix it because it was an old pig ugly layout so thought I'd do an updated version and add it as a new thread to keep the versions separated. 

A bit smaller than my last layout and I've also included trimmers to replace the JFETs drain resistors to make it easy to bias and fine tune to your own liking.

Info about the original:

As far as guitar clean boosts go, the Katana is KING TONE. It's the pedal to have for the finishing touch on your tone.

Want a mirror image of your tone, only louder? This is it. The Katana is also a fat, harmonically rich boost, just pull out on the volume knob and stretch your imagination with overdriven tone! You can keep it on all the time to push your amp harder, or as an occasional boost to cut through the mix.

Obsessed or Compelled with the Fattest, Fullest Drive? Well, Before other people started using Blue LED's and non-diode clipping stages, there was the TMB and Katana.... Cascading FET stages do the tube-amp-thing like nothing else!

The Katana circuit is Robert Keeley's first guitar effect design for Keeley Electronics. It was originally designed for Legendary Tones for the Time Machine Boost but then became an independent product under the Keeley name after the Compressor came out. Robert's very first effect pedal design (1999) was a digitally controlled Analog Delay complete with built in digital tuner. It worked perfectly and can even be seen at the shop from time to time. Delay times could be input via a keypad to enter the amount of delay in milliseconds or by potentiometers. The Katana is no less stunning in its beautiful simplicity and stunning sound quality.









156 comments:

  1. Mark, thanks so much! I changed the 150nf to 15nf and the pedal now sound great!

    As for the pedal making a popping sound when the boost is engaged, I put a 75k resistor inline with the 10u electro(for the boost) and it did quiet the pop. However, there is a tone change. I'm not sure if it's a volume drop or if the tone is truly changing. It's hard for me to verify as I'm manually going back and forth. Would you happen to know what the resistor is doing besides stopping the pop?

    Thanks for the new layout; I'd make this one, if I hadn't just built the older rev.

    ReplyDelete
    Replies
    1. Have you tried a new 10u cap? I've found some of them seem more prone to popping than others.

      Delete
    2. Hi Mark!
      Shouldn't Volume 3 go to Sw5 ?

      Delete
    3. Shit how did I do that?!
      I'll correct it, thanks for the heads up

      Delete
    4. IvIark, hmm, nope haven't tried a new 10u cap. Maybe if I try using a ceramic or poly cap that might help?

      Delete
  2. Hello Mark,
    Dummy enough to ask this question...
    Is this a true bypass one ? (as switch is DPDT)...
    Thanks a million for this amazing blog.
    Greg

    ReplyDelete
    Replies
    1. Yes every layout on the site assumes true bypass switching unless noted in the layout. There are probably only 5 or 6 that switch is a less usual way like buffered bypass. So the switch is wired as per the Offboard page in the menu.

      Delete
  3. Man, I just placed an order from Tayda before I saw this...too bad otherwise I'd have ordered the right IC, transistors and diodes!

    ReplyDelete
    Replies
    1. I did the same thing. Then I realized Tayda doesn't have the 2N5484's anyway, and I felt a little better.

      Delete
  4. Hi Mark, Can i change the jfet with 2N5487 and MAX7660CPA Charge pump?

    ReplyDelete
    Replies
    1. Yes you should be able to use pretty much any JFET because you can bias them with the trimmers, although that doesn't mean they will all sound the same as they obviously have different characteristics.

      I can't find the datasheet for the MAX7660CPA but if they follow the same numbering convention as the ICL versions then it may have a lower frequency oscillator which will cause high pitched whining. Personally I would look for a xxx7660"S"CPA which should have a boost function which increasing the oscillator frequency. ICL7660SCPA can be found very cheaply and in quantity on eBay.

      Delete
    2. Yup. S stands for Super. You'll want those :)
      +m

      Delete
  5. what about the push pull??

    ReplyDelete
    Replies
    1. The switch connections are shown on the layout. Use a push pull if you like, or a toggle if you prefer. As long as it's 2 position DPDT.

      Delete
    2. sorry im beginner, i dont see the layout showing any connection to toggle switch, please advise.

      Delete
    3. Everything marked Sw# goes to the lug shown on the switch on the layout. So lugs 1, 2, 4 and 5 need connections

      Delete
    4. I see, i thought the sw# are meant to connect to the footswtich, and I have to use a dpdt footswtich for this~!

      thanks~!

      Delete
    5. No the stomp is wired as per usual with the usual offboard wiring layout using the input wire and volume 2.

      Delete
    6. sorry, one more question, where so i connect LED+ to? it seem that the offboard wiring layout do not have that.

      Delete
    7. To the LED anode, then the cathode goes back to the stomp for switching

      Delete
    8. thank you for your patient, i guess I am very amateur,have a nice day

      Delete
    9. oh and I assume i do not need 2k2 resistor for this board?

      Delete
    10. No you don't. Where I have space I always put the LED resistor on the board because it simplifies and tidies up the wiring if you don't have to solder a resistor hanging in space somewhere offboard.

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

      Delete
  6. Hi lvlark, just assemble it, there are no signal coming from audio, although when i activate the footswitch there are light from Led, may i know where should i start checking??

    ReplyDelete
    Replies
    1. sorry i mean there are no sound when activate the footswitch , but sound only when bypass it

      Delete
    2. If any pedal doesn't work the first thing you always need to do is check voltages of the active elements, and so any ICs or transistors. If you can post the DC voltages you've getting between all IC and transistor pins and ground that will give us a big clue as to what your problem is.

      Delete
  7. hello my voltages are

    GND 9.35
    Trigger 8.83
    output 9.13
    reset 8.66
    vcc 9.35
    discharge 8.75
    threshold 8.67
    control voltage 8.62

    seem ok?

    ReplyDelete
    Replies
    1. NO! for starters, GND supposed to be ZERO volts. the output shouldn't have any DC voltage on it either. IMO, you must have some major errors in your build.

      Delete
    2. Hi John there are something i dont understand, the GND (first top pin on the left), which is next connect with 9v wire, why will there be no voltage ?

      Delete
    3. I think you have a wrong chip. Pin 3 is ground on ICL7660S...
      http://www.intersil.com/content/dam/Intersil/documents/icl7/icl7660s-a.pdf
      Wrong chip would also eplain why have near supply voltages on all pins.
      +m

      Delete
    4. Hi Mirosol

      I had upload some photos maybe you guys will have more clues,
      Thanks for helping and patients.

      top view : https://www.dropbox.com/s/cee7pot0vlonbce/Photo%2023-3-14%2012%2052%2004%20am.jpg

      bottom view : https://www.dropbox.com/s/15av5xb6p9jn8oi/Photo%2023-3-14%2012%2052%2014%20am.jpg

      IC CHIP: https://www.dropbox.com/s/85475uhkepek60y/Photo%2023-3-14%2012%2053%2009%20am.jpg

      thanks again!

      Delete
    5. i'd first check ALL of your cuts for shorts with a meter. many of them look like they're not cut all the way across. especially this one,

      http://johnkvintageguitars.homestead.com/Effects/misc/Photo_23-3-14_12_52_14_am.jpg

      but others look like they may be still shorted too, so check all of them.

      Delete
    6. Just did and still no luck :( may I know how did you guys connect the black and red for multimeters? Does the found connect to pcd

      Delete
    7. All grounds need to be connected together. The the black lead og the meter to those grounds. The red one has two connectors on the meter. Other is for measuring current and the other for volts/ohms/mAs. The second one is the one you need to use.
      +m

      Delete
  8. Does the black connect to pcb or to dc

    ReplyDelete
    Replies
    1. The ground wire? All grounds need to connect together, so ground wire from the board, negative on the adapter, all socket sleeves etc

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

    ReplyDelete
    Replies
    1. or i should put it this way

      IC this way UP

      1=9.35 5=9.35
      2=4.45 6=4.34
      3=0 7=4.93
      4=0.14 8=0.13

      Please help, I now can hear a little signal when I turn the pot and trimmer to max with lots of high frequency noise

      thanks~

      Delete
    2. What are the transistor voltages?

      Delete
    3. I used 2x 2n5484

      the one near LED

      G =0.00
      S = 0.71
      D = 18.25


      the other one

      G 0.00
      S 0.68
      D 12.83

      Thanks

      Delete
    4. Well that proves the IC is working ok then. But both transistors are totally mis-biased, tweak both trimmers until you have 9V on both drains and see how that affects things.

      Delete
    5. Hi lvlark

      I cant get one of the transistor to go 9v? ( the one near SW2, is it trimmer problem, or are they something else i should worry?

      Delete
    6. There must be one or more problem then with the trimmer, the source resistor, a faulty JFET, or the soldering. Check the source resistor is definitely 680 ohms, check your soldering for resistor, trimmer and that JFET, and if all that looks good then you're going to have to start swapping out those components to see what is causing the issue.

      Delete
    7. Will do! Sorry for many message!

      Delete
    8. One of the 220 cap has 18 and other side has 0 , is or weird?

      Delete
    9. I assume you're talking about the cap to the right of Q1, if so that is ok. One side is connected to the drain of Q1 and so you would expect the same DC voltage there that you are applying to the drain. But capacitors block DC (which is the reason they're used here) so no DC voltage at the other end is expected.

      Delete
    10. Hello, I had re assemble all parts with new capacitors,trimmers, resistors and also diodes, It seem to have sound now, but its only when I rotate the trimmer to the extreme anti clockwise, its very LOUD , the pot seem to have zero respond, other than that... its still no audio signal.

      May i also know Volume 2 to 'Output' ... Does it mean footswitch output which is refer to top right corner , not the jack output..right?


      thanks

      Delete
  10. Built the katana, the volume is too low even the pot and trimmer is max, may i know why?

    ReplyDelete
  11. Made it. It's kinda tight, but fits perfectly into a A-box. It sounds like heaven too. http://i57.tinypic.com/2ih6wsl.jpg

    ReplyDelete
    Replies
    1. Haha, what a coincidence, I was just sent a link to that pic on muusikoiden.net. Excellent job, it's a really great looking build!

      Delete
    2. You can take all the credit. There wouldn't be a pedal without your layout. I love the way you make these layouts to be small enough without those "standing" resistors that so many others use.

      Delete
    3. Yeah, Thx to lvlark, i had built 2!

      Delete
    4. here's mine
      1 http://tinypic.com/r/334jlnl/8
      2 http://tinypic.com/r/2ag6xzs/8

      Delete
    5. @salanen - That is an EXCELLENT looking build... clearly this is not your first pony ride... (or is it ".. first rodeo?", perhaps a cooking reference? "... first steamed azna?",".. fricaseed?") sufficiently obscure references, i think ;-)

      It seems to fit that mini-box perfectly - snug rather than crowded.
      Tasty enclosure finish & graphics too... sigh... just can't get 'em to look that good (my board gets a little uglier with every build, lol.. but they sound good anyway - guess that's something ;-)

      Mark - no amount of thanks is sufficient for what you bring to us here.

      Cheers!

      Stephen

      Delete
  12. Is there any chance to put something like master volume here? It's really loud,

    ReplyDelete
    Replies
    1. I'm confused. The volume pot is at the output will take the level down to nothing.

      Delete
  13. Built this with j201 fets. Anyone else try it. When I biased drains to 9 volts would get an oscillation (squeal) when turn up the volume, also with 10k pots could only get biased to as low as 11 or 12 volts. Used 20k pots and got down to 9 volts. Biased back up to 12volts due to squealing and seems to work fine. Anyone else experience this?

    ReplyDelete
    Replies
    1. Swapped the pot for audio taper. Found the linear taper to be too abrupt on low levels.

      Delete
  14. With Vol all the way down there's no audio, as soon as I start turning Vol, signal kicks in at unity. From there it goes up. Up, up, up. This pedal is too loud to be practical, for me. If I just take the Vol pot down to like 100k, that should lower the overall output volume, yeah?

    ReplyDelete
    Replies
    1. No, I suppose that would only decrease the maximum available volume. I'd really like to shift the whole volume range (min to max) down.

      Delete
  15. Use a 250K log to stretch out that 0-unity part

    ReplyDelete
    Replies
    1. Thanks Travis. I did that. The log sweep does help the, but like some others have said, this thing really starts to sound amazing and gritty with the volume around 7.

      As an example: I can have my Blues Jr. set to Volume 1, where the guitar is barely even audible. If I run my guitar through the Katana at Volume 7 or 8, the output is so loud I'm afraid my downstairs neighbors are going to be upset. I understand that a boost should boost, but it's really an amazing difference in volume.

      Delete
  16. I am not sure my post went online. I am building this right now, and I see on the main board Vol 3 goes (right side) to the 220n cap. BUT down below it also says Vol 3 goes to SW5. I am confused... there is only one Vol3, but it can go 2 places (just seems like it would be more clearly marked).

    I am sorry IvIark, I know you have been through this rev several times now.

    Also - best sub for 2n5484? J201? or something else? Thank you!!!

    ReplyDelete
    Replies
    1. Both need to connect to Volume 3. So you can run 2 wires to it, or daisy chain. Do it any way you like as long as Volume 3, the point on the board and Sw5 are all connected to each other.

      Delete
    2. Thank you, I had the lead from the board to SW5 and the lead from Vol3 to SW5 so I think am good. Can you please explain the LED? I see LED+ comes off the CLR (4k7) but where do I terminate the - (gnd) end of the LED?

      Delete
  17. Now I have the |>| working - good. I also show 18.80 volts at the Tran sockets, the problem is my trim pots won't reduce that to less than about 18.60. Just a hair. not sure why.

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

    ReplyDelete
  19. I removed post above because errors were fixed but still no joy. I am using J201s, I see a schematic that says 2n5457 OR MPF102 (which I have, I guess I will try them).

    I am still wondering about the LED - it is dim, but glows brighter when I start to insert the trannies.

    Anyway, I fixed all my problems (wrong connections in the power pump area, diodes & 1M Rs, but that is fixed now - but still no joy.

    ReplyDelete
  20. I found out I had built a good portion of this circuit "incorrectly". I was wide awake and thought I was breezing through it, but I was using pieces placed incorrectly as my guides and hence had a lot of wrong placements - starting over. Just when I thought I was getting decent at this. Very frustrating. I also burnt up some of the last good pieces I had. I guess I will have to wait for new parts to try this one again.

    By the way - the instructions on the LED could be better - (where to connect the cathode), and also the volume 3 going two places is confusing. I also doesn't help that the trannies and Diodes are unobtainium now. I'm just saying it would help to clarify this piece that a lot of people seem to have struggled with.

    I tore my old one apart last night - but now I am wishing I had just replaced the diodes and tried a few things at a time.

    ReplyDelete
    Replies
    1. If you follow the offboard wiring shown here, the LED should be fairly clear. You have the required current limiting resistor on the board and the marking shown for the anode. You know that the other side of the LED has to be switched by the stomp in order to turn the LED on and off, and so that fits in with the offboard wiring diagram. And if there are two required connections to Volume lug 3, then I can't reasonably do much more than show you were to connect them either graphically or in the notes as I have done.

      I'm not going to start adding offboard wiring to every layout, and won't hand hold everyone for the entire build, but all the required blocks are here and most people seem to follow it ok. Plenty of people seem to have managed to build this successfully. For anyone unhappy unfortunately I can't offer a refund for the absolutely nothing you paid for the information.

      Delete
    2. I'm afraid Mpaul that if you're gettin' understanding problems with such an easy layout, you're burning some of your parts or calling "unobtainum" a diode like the 1n5817 (if I'm not wrong I have about 300 and did cost $2,5), I'm afraid that the problem is between the screen and the chair, not in the layout.

      Seriously, this is a 15 minutes build for an experienced builder, and not more than half an hour for a newbie, it the newbie pays a little attention to notes and has a minimum knowledge base. If you have no idea about how to wire a stompswitch or a single led, blame it on yourself, not on the layout.

      J.

      Delete
    3. Was going to answer this, but I will pass. I will just say you should not assume I have not tried to figure this stuff out on my own. I have. Sometimes a noob misses the correct answer because he does not recognize it when he sees it. That is what the phrase "A noob doesn't know what he doesn't know" means.

      What is not known to the people who already know how to do this stuff is that there are many different versions of what is correct out there, including three different "offboard wiring diagrams". And although I love this web site, I have to honestly say this particular offboard wiring diagram has not worked for me, and I cannot say why (I am sure it is me) - but even when I follow it to what I think is a "t" I still get hum.

      Yet, when I consult a different one I get immediate returns, it works perfectly every time.

      Once again, I know its me (hey, I ain't stupid but thanks for suggesting it anyway), but the reason WHY they have not worked for me is what I am trying to figure out.

      I just figured out the LED thing. Hey, someone gave me bad information. But I just tried to do it in this circuit (tie the anode to + and cathode to ground) and the light still just stays on. It's something else in my switch, I assume.

      Maybe I burnt out a diode - no wait, those were fresh. Maybe I burnt out a pot, but other people said they had the same range problems I had. Can I assume any 7660S is the same? - I don't know, it seemed to be working.

      Okay - enough beating this dead horse. Thanks for the help, and the other comments as well.



      Delete
  21. Sorry - guess I got on your wrong side, so first of all, I apologize. I didn't mean to sound like I was ranting, and I appreciate all the help you have given me so far a lot.

    I am a NOOB, and the problem with us is that we don't know what we don't know. "You know that the other side of the LED has to be switched by the stomp in order to turn the LED on and off, and so that fits in with the offboard wiring diagram." Actually, I know that logically that is supposed to happen, but for some reason I can't make it happen. For some reason I have never been able to get a circuit where the LED is between the power source and CLR to work. Why? I have no idea. Maybe there is no clear guide out there on how to make it work. I thought I found one and posted, and someone told me it was wrong. (here: http://guitar-fx-layouts.42897.x6.nabble.com/3PDT-true-bypass-wiring-pcb-s-tp6496p12544.html)

    This is one of those circuits - your wiring guide shows the CLR the between power source and the LED. This circuit has it backwards (I think). I am sorry, but that has thrown me off course every time I have encountered it, and the ONLY fix I have ever found is to turn the +9 - CLR - LED connection around as shown in your wiring guide. And I don't know why. Hard to believe I am the first person who has ever had this problem, though, since it happens to me EVERY time I encounter it. (I am laughing at myself here, not mad, just frustrated as he||.

    Also, I am trying to use these foot switch PCBs I just bought from a member here, and for some reason I cannot make those work, either, because the LED is "wired backwards" - it just doesn't work for me, I don't know why.

    With this circuit, my LED stays on, it does not switch off. And when I plug this in, and I cannot get the FETs to read below 15v no matter what I do. And I have a very loud ground hum.

    This is not your fault, IvIark, but people get frustrated when they feel like they have done everything shown and it still doesn't work. This isn't my first build from here, it is about my fifth and the rest have been fine, and I have figured out plwnty of problems on my own, but I just rebuilt this one from scratch this morning and I have the exact same problems I had after "fixing" everything last night.

    Once again, I don't know what I don't know. I am using J201s instead of 2N2584s, I am using 5818 diodes. I assume all of those are good subs because people above say they have used them. But my 7660S is marked CPAZ, not ICL. I don't know if that is good (I have been assuming it was. I am getting 18 volts at the Trannies, but I can't reduce them).

    Sorry to go on. But please understand that this is not directed at you personally. A post like mine is meant to go out to people who have had similar problems and may know fixes. If you do not know the answer, I don't expect you to look it up.

    But here are things I don't know - if I can't get either pot to read below 15v, could it be bad pots, bad diodes, a bad IC, what? I changed the diodes, pots, and still no joy. What is the most likely suspect?

    ReplyDelete
  22. Here is where I stand. I changed to foot switch wiring to the one that has worked for me in the past. I now have what I believe are the correct readings for the IC:

    9.53
    4.62
    0.00
    0.14
    0.30
    5.48
    4.91
    9.53

    on the trannys the minimum I can get on the drain is

    14.23 and 15.44 (10k trimmers)
    sources: .27 .25
    gates .01 .01

    The LED is working correctly, bright; on/off

    no overdrive sound at all, just clean guitar with circuit in or out.

    I do hear a pop with the DPDT, and I hear a tone difference.

    That's it - the best I can do. If anyone has any ideas please let me know.

    ReplyDelete
  23. hi there. i think you should just try an other circuit. i know how you feel when you strongly believe that you've tried everything... but for me, this site opened my eyes...here i've found the clearest layouts, and the best guides. so i think you just need more time on this hobby. read as much as you can on every post here, in the forum and everywhere. and for the LED i hope you'll find your mistake cause it's really simple. good luck!

    ReplyDelete
  24. I just finally built this again and got it working. Here is what I have happening now, though; neither J201 will go down below about 14v - the right J201 sounds best when the trim pot is turned all the way down. The second trim pot "motorboats" when attempting to bias by ear - the beats get faster as you go clockwise and then you suddenly get tone. It's a good, fat, loud overdrive as I expected, but I still don't have the recommended 9v bias voltage as mentioned above. I suspect it is because we are using J201s, not the original recommended tranny.

    The 2n5484 is clearly marked "obsolete" on the product sheet: https://www.fairchildsemi.com/pf/2N/2N5484.html (hence my incorrect statement about unobtainium above, I should have said "obsolete").

    And although the J201s cannot bias below about 14v, still, they do work and sound pretty good - but my mistake was trying to bias by DMM to 9v, instead of just doing it by ear.

    Note - these comments are here to help people who try this with J201s and have problems. Yes, I also love this site a lot, but it is not criticizing to add notes when you find solutions to what a few different people have mentioned as common problems in the comments sections. In fact, I think it would be helpful if everyone did their best to be very clear with comments when they figure something out that seems to be a common problem - thanks for listening.

    ReplyDelete
    Replies
    1. Thanks for sharing your findings. No one would ever suggest this type of comment was a criticism, it's precisely the kind of info that we want to share, particularly as some of these through hole transistors become more difficult to find.

      I was lucky with my timing and spent a few years stocking up on some of these parts that are now becoming hard to find, but if you are looking for something in particular you can find a lot of these on aliexpress. I've bought from them a few times now and it seems very similar to eBay for buyer protection, so you can always get your money back if they don't work in a circuit and so you suspect you've been sent a dodgy part. But everything I've bought from there so far has been fine and have been received fairly quickly as well, although admittedly that will vary from seller to seller.

      http://www.aliexpress.com/wholesale?SearchText=2n5484&catId

      Delete
    2. dude, just because a part is marked "obsolete" doesn't mean you can't get them, you just have to look other places the the typical big box companies like mouser, you might have to look on ebay. think about it, we have an entire thread about Ge transistors, those have been obsolete for over 20 years, yet we miraculously are able to find them, and in most cases don't cost an arm and a leg. here's a link for 50 of them for $25, which would be $0.50 each, which is not uncommon.

      http://www.ebay.com/itm/50PCS-2N5484-Manu-NS-Encapsulation-TO-92-N-Channel-JFETs-/131214015382?pt=LH_DefaultDomain_0&hash=item1e8cf6fb96

      part of the problem you're having is you seem to just be doing, and not understanding what you should be doing. for instance, if there is a part that is obsolete and you can't get it or it costs a fortune look to find a substitute. i typed into google "2n5484 substitute," and there is a link to Geofex about matching FETs that on the top shows a chart comparing different FETs. it clearly shows that the maximum voltage for 2n5484 is 3.0V, while the j201 is 1.5V.

      http://www.geofex.com/article_folders/fetmatch/fetmatch.htm

      from that info alone, i would say that you can NOT substitute a j201 for a 2n5484, that is you're biggest problem with your build. if you don't have the right parts you're build will not work right. its like this, you can't take a ferrari on a rock climbing trail, or a car build for drag racing can not succeed at the 24hr of Le Man race.

      can you read schematics? if you can't you should really learn how, as building something without reading and having the schematic is just asking for issues, cause you can't understand what you're doing in a build. i think you need to do a little more research on a builds and make sure you have the right parts BEFORE starting a build.

      the same goes for your issues with the offboard. if you don't know what you're doing and why, you're going to fail. i think you need to do some more learning about how effects work, what parts do, and how they work together before picking up that iron. cause it keeps seeming you have major issues with minor situations. especially if you can't understand the notes on the layout telling you where the wires go.....

      Delete
  25. Thanks Johns, I appreciate your comments. This is a learning process for me as I have said before. You just taught me something I did not know about J201s. I was not alone in lacking this knowledge -- if you scroll up you see many other comments asking about using J201s. I wish this answer had come sooner, but at least it is here now.

    I can read a schematic, but I am new to it, so as far as using that to calculate voltages at a given point, etc., I'm not really there yet.. I just started a thread in "open chat" asking people today "which original parts can you substitute, and which should I really try to find?"

    I never took electronics in high school, or college or night school, but I am pretty good at learning things when I apply myself and when I get the help of people like you, Mark, Mirasol and other people here. I am 60 years old now and I am doing this for the love of seeking great tone and accomplishing things with my mind and bare hands. I don't plan to do it for a living and I could afford to buy these pedals if I wanted to, but I am here for the love of electronics and secrets of tone I have really wanted to know for years now.

    Thanks for your advice and knowledge. I did finally get the circuit to work, which I think says something, and it sounds good. And just for the record, my previous comments in this same board are now pretty old. I just decided to revisit this circuit because I could not get it to work before. I did get all of my grounding and off-board issues worked out.

    Even I can learn a few new things in a month or so.

    ReplyDelete
    Replies
    1. Whoa, that info is for matching fets in builds like phasers, not for substituting them in builds like this, a j201 will work in this. whether it sounds identical is another matter but you will also get variations using different 2n5484`s

      Delete
    2. Yes, matching FETs is a whole different ball game and will determine how well the FETs will work with each other in a given circuit, usually a phaser.

      All JFETs will work in drive pedals, but the way they bias will differ. Because they are extremely inconsistent the bias will differ widely in JFETs of the same type never mind different ones completely. This is why the vast majority of JFET layouts on this site include trimmers so you can set that point individually (excluding muamp layouts which should bias without any further intervention which may be one reason they have become so popular with some manufacturers).

      So any JFET would work, but you may have to change the drain or source resistor may need to be altered so it works correctly. Suitability then is a very subjective thing and what some may like others may not, just like with bipolar transistors.

      The biggest issue with JFETs I think is that they're not understood. The characteristics aren't self explanatory, for instance you hear people say that the J201 is high gain, but there is no characteristic that shows that clearly. I still don't know to interpret the information so I can understand clearly in my own head what is low gain, medium gain or high gain so I have a pecking order of suitability in my own head based on the type of effect. Transconductance is important but it still isn't clear like it is with when you can measure hfe in bipolars.

      So I'd say the same thing, get as many diifferent types as you can and use them in a common JFET drive circuit and see how you think they compare to your own ears. A lot of the dscontinued ones can be bought relatively cheaply in lots of 10, so get a few, try them, and then stock up further on the ones you really seem to get on with.

      Or go SMD with an adapter which may now be the sensible option. One of the reasons I bought the adapters originally because I wanted some 2SK208s which are used in the AMT amp sim pedals and they are SMD only, so going this route will give you options that are unavailable if you stick with through hole.

      Delete
    3. Ahh. I saw that it was about matching jfets, but I thought the chart on the top was informative for comparison sake some of the electrical behavior between some of the different jfets. I could have read it wrong, or taken more or rather inferred some things from that which I shouldn't have.

      FYI, I didn't mean or want to discourage you motter, so please don't take it that way. I type the way I talk. I hope you're able to find the tone you're after, like the rest of us, but I warn you it's a road that seems to never end. The more you build the more you learn and the better each build gets.

      Delete
  26. Even though I usually don't like boosts or overdrives, have to admit that this circuit sounds pretty good an works wonders on solid state amps. Would never use it in a valve amp, but it makes my Valvestate clean chanel work as a "real" amp.

    I've used J201's, tomorrow i'll try 2N5458 to see if there's noticeable differences. By the way, with the J201s, you get the best sound biasing the fets at 10,1 volts.

    When i put the 2Ns, will have to rebias, using the definitive power supply (tests were made using a battery readong 9,1v, my power supplys give 9,6 and 9,4 volts.

    J.

    ReplyDelete
  27. Javi - Hi and How and are you ;-)

    You are correct, the J201s do bias much higher, and I also tried some different Fets and although I could bias them lower, they did not give me the same gain at all. I tried a 2N5457 (I had 1, so I put in the first position) and I also tried a pair of MPF102s. Both biased anywhere from 0 to 18, but neither really worked. The J201s I still cannot get to bias below 14 (on 1) and about 11 (on 2). But the circuit still works.

    I also tried swapping out J201s and found a few "good ones" that made the noise level quieter but gave the playing tone more gain and less grit. A noticable improvement in both areas. Bottom line - fets vary a lot.

    ReplyDelete
  28. Sorry, but "neither worked" I meant "no gain," just clean tone at those bias levels, and also no gain at higher bias levels either. I also have some of the 2ns on order, and will try those.

    ReplyDelete
    Replies
    1. Hi Mr. 7 :P, fine, and you? How your battle with leds and resistors goes?

      Have still not tried it with 2n5457, but I think that tomorrow will give a try with several fets. I ordered just yesterday the push pull 250K pots, so it would be about three weeks more or so until I finish this pedal. Plenty of time to try different things in this effect

      J.

      Delete
  29. Well, tryouts finished :) From worst to best:

    2N5457: Easy to bias at 9 volts. Voltage has to be set really high (about 14volts) in order to get an aceptable sound, but they get really noisy.

    MPF102: Dull sound, no matter how high you set the voltage, empty sound and very little gain.

    J201: Imposible to set the drain voltage under 9,8volts with these trimmers. Anyway, in my pair, the sweet spot is about 10,6 volts. Rich sound, and nice distortion. Work great.

    2N5484: Very responsive to the trimmers (from 18 to 6,7 volts) Sweet spot about 9.5v. Sound really rich, round and powerful, total winner.

    Hope it would be of help.

    J.

    ReplyDelete
  30. Thanks for the input. I am also waiting for 5457 - so good to know they will sound great. This was a learning experience for me with the J201s and I am glad to see you had pretty much the same experiences I had with them and MPF102s. I am still don't know why I cannot bias my first (right-most) J201 below 14. I wish I could get it lower because It does sound best fully counter-clockwise.

    ReplyDelete
    Replies
    1. Just use a 20k or 47k trimmer in this position, you'll be able to lower the voltage some volts .

      J.

      Delete
  31. hello there...just wondering, does the led+ just goto the led for the on off? thnx...

    ReplyDelete
    Replies
    1. yes. the 4k7 is the load resistor for it.

      Delete
    2. It is. I found it easier just to leave it off and wire the box up like any other, but do use that value for the LED (4.7k) which is essentially doubled because of the charge pump.

      Delete
  32. Oh shite, I should have said it is doubled because of the charge pump, but if you wire it up a different way you should use the proper resistor value for whatever voltage you are feeding it. If you are coming right off of the DC input you probably want to go 2.2k (sorry).

    ReplyDelete
  33. Ok...I think I'm going to try to run the led from the board first to see if I like it. If I don't run it from the board do I still need the 4k7 resistor there? Just curious...thnx...

    ReplyDelete
  34. No, all it does is control the voltage to the LED if you choose to use it to show power to the pedal. It does not affect the tone. I used it and it works fine, but it adds and extra wire you need to take to ground from the LED. A lot of people already have a routine of setting up the "on indicator" LED where the resistor is not on the board. It's there and it' up to you whether you want to do it that way, but if you choose to go with another "box wiring configuration" that includes the LED, (such as this "http://tagboardeffects.blogspot.com/2012/02/offboard-wiring.html") then use the value recommended in that diagram.

    ReplyDelete
  35. Circuit Maestros,

    Please help! I'm a noob, plain and simple. I am in desperate need of help getting my pedal to work as it is a gift (thought that counts) for a friend who recently suffered a great loss. I have followed the instructions as best I could, including the offboard wiring. I've also read every comment to try to figure out what I was doing wrong, but to no avail.
    The issue is that there is no output on or off but the LED turns on. The IC also gets VERY hot very quickly. Also, I haven't biased the transistors because either my Multimeter isn't supplying voltage or they aren't getting any because they say 0 V no matter how I adjust the trimmers.
    Is it a component that is bad, a cold joint or something else? One thing I'm not 100% on is that my LED is connected directly to the footswitch lug and the + lead goes straight into the board as the diagram says.
    Any assistance would be greatly appreciated!!!

    Thanks,
    Greg

    ReplyDelete
    Replies
    1. Usually, if an IC or other component is getting really hot there is an error in the wiring related to it getting direct +9 voltage from the power source. Look for a missing link to ground, or perhaps a cut that wasn't made in that neighborhood.

      Delete
  36. I reckon a couple of pictures of the front and back of the board, as well as offboard wiring would be handy to help you out. If the IC is really hot, you may be using the wrong type I think? Biasing shouldn't matter if they're set around the middle, you should still get output.

    ReplyDelete
    Replies
    1. Here are 4 of the board and 1 of the 3DPDT. The footswitch is wired:
      yellow = -LED
      green = input jack hot
      grey = vol 2
      purple (middle) = board input
      purple (right) = output jack hot

      https://dl.dropboxusercontent.com/u/100339873/20150227_170122.jpg

      https://dl.dropboxusercontent.com/u/100339873/20150227_170158.jpg

      https://dl.dropboxusercontent.com/u/100339873/20150227_171935.jpg

      https://dl.dropboxusercontent.com/u/100339873/20150227_171955.jpg

      https://dl.dropboxusercontent.com/u/100339873/20150227_172558.jpg

      Delete
    2. I just realized that I connected the board input to the wrong place on the board cuz I'm an idiot. That's definitely a problem, and may solve the problem, but it could also be one of many, so we'll see. I will fix that and see if it solves the problem - then I'll worry about the biasing hugamaroo.

      Thanks,
      Greg

      Delete
    3. So now I get output when it's switched off but switching it on there is nothing, the LED turns on and the IC gets really hot almost immediately.

      Delete
    4. Hey Greg - as MotterPaul has stated below, you've actually built the circuit upside down. Please follow his instructions below to get a grasp of it. I did the exact same thing the first time I built on vero! :)

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

    ReplyDelete
  38. The more I look at it, it seems you built it correctly although just "backwards" - I would still suggest starting over but this is not impossible to troubleshoot.

    I mentioned the missing cut. One reason they recommend the process of using a small drill to mark the cuts on the non-copper side is so you can see them in the finished project. It is also a lot easier to build it as you see it - not reversed.

    ReplyDelete
  39. I deleted comment about bottom 10uf - when I look at solder side it appears to be correct, although I could not see the wire on the front angle.

    ReplyDelete
    Replies
    1. So I rebuilt the whole thing (not inverted this time) but it still doesn't work. I suspect because of my offboard wiring. I'm confused about a couple things.
      1.) Does the LED anode connect to the board or to the DC jack like in the offboard wiring diagram on this site?
      2.) When he says volume 2 to output does that mean the output jack or the output lug on the 3PDT switch?

      Delete
    2. The LED is a confusing thing - The offboard wiring diagram on this site has the anode connected to the resistor (CLR) which then connects to positive on the DC jack. But the LED status should not affect whether the circuit works with your guitar. (it is also okay to connect with the CLR on the other side of the LED - but you will get the hang of this)

      Volume 2 to output means that you wire the middle lug of the volume pot (2) to the #7 spot (upper right) of the foot switch, so when the switch is on it connects to the tip of the output jack. (just as the same wiring diagram shows). "Output" on any circuit goes to this same place, whether from a pot or the board.

      By the way - you do not need to wire up footswitches and LEDs to test a circuit. In fact it is easier to test it without that stuff, Just connect the DC + and - to your build, (a breadboard works great) and connect the input and output to tips of two 1/4-inch jacks (guitar input, amp output).

      Here is an example: (different circuit) http://i1304.photobucket.com/albums/s527/Paul_Motter/plexi1_zps1374fd8d.jpg~original

      Also - make sure everything that needs to be grounded is done.

      Delete
    3. Okay I will give that a shot - I'm definitely not a natural at this but I'm going to make this friggn pedal work if it's the last thing I do. Thanks so much for the help!! Much appreciated

      Delete
  40. Thanks for all your help - I appreciate you taking the time to hand hold for a bit. I've had to put the pedal aside for now to save the hair I have left as my last attempts to fix the damned thing (that I really did think were gonna work) were to no avail.

    ReplyDelete
  41. I built keely katana and it works, but every time I turn on footswitch it's making a screaming (high pitch) sound in a constant volume and when I increase treble from my amp its getting worst. Does anyone know the reason and a way to fix it?

    ReplyDelete
    Replies
    1. i have had this problem before with charge pumps. i had it when i used a 1044 instead of the 7660. did you use the 7660?

      Delete
  42. Replies
    1. Oh, and I used the 1N5818 instead of the 1N5817!

      Delete
  43. Hi. Just finished this layout. The signal goes, but there is only little boost. Measured the q1 and q2 drains and i get about 17-18 volts. Tryed to bias with trimmers but with no luck. Where could be the problem? I used 2n5484, and 10k trimmers.

    ReplyDelete
    Replies
    1. Managed to bias the drain sectio to the 9 volts using a 250 k trimmers. The booster works, the overdrive section works too, but when i engage the booster to the clean mode i get very muddy, bassy sound. Where could be the problem? how can i reduce the bass and add more treble and mids to the clean boost mode?

      Delete
  44. It should be clean bypass. I would look at your footswitch wiring.

    ReplyDelete
  45. I built this and it sort of works, but I have a hard time biasing the JFETS. Fired it up first time, biased the 5458's down to 9V, but the level was just barely above unity…

    I can get it louder, but then I read about 2.5-3V?! Could it be there's a problem with my JFETS? I tried 4 different ones in all kind of combinations, no luck— can't remember where I got them from, but could be China.

    ReplyDelete
    Replies
    1. It's strange… the charge pump is definitely working, i get 18V at the trimpots respectively. If I bias to 9V no real volume increase. If I bias to 4,5V (half of 9V instead of 18V) – bingo.

      Checked everything, reflowed joints. Measured the JFETS and selected the two best falling right in the middle of the datasheet spec.

      Anyone got an idea what the problem might be?

      Delete
  46. I have some questions.

    If I look to the only schematic I could find on the web there are some diferences. I refer to this schematic: http://3.bp.blogspot.com/-38lRv_volBE/TzWxOvsMR2I/AAAAAAAAA6o/CJ_Konw23nI/s1600/keeley_katana_briggs.jpg

    1. In the vero there is an extra 10uF that does not appear in the schematic.
    2. Why connect both pin 2 and 3 to V+ on both trimpots instead of only to pin 2. Isn't connecting it to pin 3 redundant?

    ReplyDelete
    Replies
    1. 1. it's just some extra filtering.
      2. there is no difference either way. just imagine V coming into pin2. it will stop at pin 3 anyway cause there is nowhere to go, and the only thing that will matter will be the resistance from pin2 to pin 1 to the rest of the circuit.

      Delete
  47. Built it and functionally it works. But I get a high oscillating sound as soon as I plug in the guitar and amp cables. It's even a very low volume hissing in bypass mode. I will look over my off board wiring carefully, of course.
    I use a ICL7660SCPAZ. When I've used it before there usually is more filtering on the +9V (100 - 47uF) and the high voltage is also more filtered (at least 47uF). Could that be the problem?
    Anyone had this problem with oscillation?

    ReplyDelete
    Replies
    1. Oscillation = high pitch squeal...

      Delete
    2. Hmm, could be a power supply issue or a opamp issue. I'd try powering it isolatedly and also would try other opamps for the voltage doubler.

      Delete
    3. Double check that pin 1 and 8 are connected.

      Delete
    4. Yes, pin 1 and 8 are connected. I've also tried a couple of ICL7660SCPZ (have several) but it still squealing. Think I will try more filtering (47uF) and see if that does the trick.

      Delete
    5. Off course, start by double checking that all the cuts are there and that there are no bad solder joints. Always reflow ALL your soldering before eliminating faulty parts etc.
      A squeal is sometimes also a signal of signal crossing or bad grounding. If you are testing outside of your enclosure, make sure that you have connected the ground of both input and output to the pcb.

      Delete
    6. Checked all the cuts and reflowed all the soldering. The squealing is still there... I think I will try increasing the filtering to 47uF, on both the the 9V nd the resulting 18(ish)V.

      Delete
    7. Good idea. Also note that a squeel and/or motorboating of an effect is often a sign that a capacitor is defective. Did you also double check the rating of the capacitors? I alway use 25V+ rated caps just to be sure that it will not fail at 18V due to the caps.

      Delete
    8. No luck. Changed the filtering of 9V & 18V to 47uF/50V and the 10 uF a new 10uF/50V. I've also tried another ICL7660SCPAZ chip. Still squealing... Besides that, it works as it should. Frustrating...

      Delete
  48. Hi, I believe I am having a similar issue to TDM in one of the posts above. The board I built does seem to function ok. I have an 18V (closer to 17V) rail, and I am able to bias the drains of the JFETs at 9V. However, with the clean boost in operation the output is not noticeably louder even when the volume is set to the highest setting, although I do notice a slight change in tone. Is there some way I can tweak it so that I get more clean boost? Many thanks!

    ReplyDelete
  49. Any time your pedal with a charge pump is squealing the most likely culprit is the charge pump.

    If you tried a few different 7660S, all from the same defective batch for example, it wouldn't help. Make sure you are using a good properly working charge pump first

    ReplyDelete
    Replies
    1. Yes, they are all from the same batch. I will order another batch from a supplier I trust.

      Delete
    2. Tried a 7660S from a new batch - and now it works fine! Avoid 7660S from the batch marked V15084P, I've thrown all my chips from this batch in the garbage (for electronic waste, of course).

      Delete
  50. I used the 5485s from Tayda (instead of the unobtainable 5484s) and they worked pretty good.

    ReplyDelete
  51. Hi

    Does anyone know how Keeley adapt this circuit to make their 1962 overdrive pedal ? Apparently it puts cascading clipping between the two stages ?

    Cheers

    Ian

    ReplyDelete
  52. Hi Guys I’ve built this but seem to get a good amount of oscillation I am using all of the advised parts no surprises and got it in a testing rig I have 9.5v at both drains and on the ice I have:

    1L = 9.1v
    2L = 4.8v
    3L = 0v
    4L = 4.5v
    1R = 9.2v
    2R = 6.6v
    3R = 6.2v
    4R = 4.5v

    Interestingly when I touch the pot chassis it pops away like nobody’s business but has a lot of noise!

    Am I missing something if I have to build it from scratch I will but anything obvious guys?

    Let me know if you want more readings!

    ReplyDelete
  53. Re: the trim pots, what should the "ideal" bias voltage for the JFET's and where do you place the probes for such a measurement?

    ReplyDelete
  54. Is there a circuit that omits the charge pump section?

    ReplyDelete
  55. I’m so glad I found this one. Ive been reading the comments and this one looks like a great platform for experimenting with different jFET’s. I’ve got this on my todo list after a Black Eye, DOD Preamp Overdrive, Proco Solo and the Omnicab sim. Definitely sockening the FET’s and looking at all of the advice and feedback.

    ReplyDelete
  56. I have built this too http://tagboardeffects.blogspot.com/2014/02/keeley-katana-boost-layout-rev-2.html
    And is working but...the dc bias v on the first stage jfet is shiftin down..at first is in the 8,5 v required(aprox) but in the next minutes its beging to shift until 4,5V...why??
    I use two 5458 and I am getting good vcc voltage out of the 7660s (16,5). In the second stage dont have this issue and I have used another 5458 and with the same results...
    any idea its welcomed, thanks in advance!!!

    ReplyDelete
  57. Built this tonight, using 2n5485, both biased to 9.5v. Works well, not sure if it's what I'm after though.

    ReplyDelete
  58. I made this and my LED is always lit regardless which switch position I am in. On effect on position the effect is working and the volume knob is working. On bypass position, there is no sound, and the LED is still lit

    This is my first with a DPDT switch and I'm not really sure what should be done with lugs 3 and 6.

    Hope someone can help

    ReplyDelete
    Replies
    1. the DPDT is not for switching the effect but the extra boost. You should connect a 3PDT for the effect and bypass.

      Delete
    2. So connect theDPDT as mentioned in the layout above and add a 3PDT like this:

      1 4 7
      2 5 8
      3 6 9

      1= Jack in
      2= In on the vero
      3 = Ground
      4= Out on the Vero (down in the street :))
      5 = Jack out
      6 = wire to 1 (Jack In)
      7= LED kathode (- side)
      8= Ground (wire to 3 is often used)
      9= not connected

      if wanted you can switch the row order (so 789,123,456) as long as the pins in the rows remain the same connections as I mentioned above.

      Delete
    3. Thanks for the reply! It makes sense now :) My mistake began with me not really understanding how the real Katana works and I thought it was just a simple clean boost

      Delete