Home Forums GuitarPCB Build Support SWAH troubleshooting

Viewing 17 posts - 1 through 17 (of 21 total)
  • Author
    Posts
  • #14225
    feedback
    Participant

    HI All,

    Sorry for the essay, trying to provide as much info as possible. Happy to post any more readings or pictures.

    Bribes of eternal thanks, beer and/or a donation to charity will be offered.

    Now the problem is exactly  as referenced in this earlier thread

    https://guitarpcb.com/forums/topic/swah-no-effect-sounding/

    Im seeing the exact same issue. The pedal functions, bypass works and is clean, the effect engages and sounds clean, but the wah is not sweeping, either through settings or picking attack. I realise the design is tried and tested, and the problem lies with a stupid build error on my part or component failure or choice.

    The Sens pot appears to not work at all. Ive traced the connections on the schematic and Sens has continuity with everything it should, and also provides the right range of resistance values. With SENS at any setting, the BIAS pot will sweep the filter over the whole frequency range manually.

    This seems to contradict whats on this page: “<b>Bias</b> – Controls the filter resonance frequency. When Sensitivity is turned fully off the Bias can be used as a sweepable filter.”

    http://guitarkitbuilder.com/build-mad-professor-snow-white-autowah-non-clone/

    Im an experienced builder, and have built entire synths from schematics on stripboard but this has me stumped. Im completely open to the idea ive made a very silly mistake somewhere, its certainly not the first time. Im quite decent at troubleshooting, but I dont have any engineering background so I dont know how to test the correct ranges of things like transistors, or know what level of voltage is correct for op amps etc.

    Ive tried the following, and the behaviour has stayed exactly the same throughout:

    (All ICS and Trannies were purchased from https://www.pedalpartsaustralia.com/)

    • Replaced IC1 and IC2 with multiple others
    • Replaced Q1, Q2, Q3 with multiple others
    • Reflowed every connection with liquid flux 3 times.
    • Replaced SENS pot.
    • Checked continuity on schematic between most connections, and everything connected to SENS.
    • Replaced battery, reads 8.8v

    Heres my voltages and pictures. The PCB is a little worse for wear in places after extensive component replacement, but the  behaviour is exactly the same as the initial build. I lifted a pad on the SENS pot, and have directly connected it to R17, and checked continuity with IC2 Pin 1. The burns on the caps are post troubleshooting, but the behaviour was exactly the same when they were pristine.

    Im assuming the issue is with power due to the low reading on Q1. Im not familiar with the VA, VB, VC designation which is making troubleshooting harder. Any clarification about what values I should be expecting at what points is very helpful.

    Ive also had trouble in the past with out of spec 2n5457s, but these are from a reputable dealer that have worked in the past.

    All values are with knobs fully left, any numbers after slashes are values with knobs fully right. I havent add the right side values when they are the same as left.

    IC1 LM13700

    1. 1.31
    2. 0
    3. 3.39
    4. 3.35
    5. 4.5
    6. 0
    7. 4.5
    8. 3.3
    9. 3.3
    10. 4.5
    11. 6.6
    12. 4.5
    13. 3.3
    14. 3.3
    15. 0
    16. 1.3

    IC2 TL074

    1. 3.41
    2. 3.398
    3. 3.39
    4. 6.77
    5. 3.31
    6. 3.31
    7. 3.38
    8. 3.15
    9. 3.3
    10. 3.3
    11. 0
    12. 3.3
    13. 3.3
    14. 1.6

    Q1 2N5457

    1D 0.03

    2S 0.03

    3G 0.59

    Q2 BC550

    1C 4.95

    2B 1.03/0

    3E 1.13/0

    Q3 BC550

    1C 4.95

    2B 1.13/0

    3E 1.3

    1n34

    1.6

    1.07/0

    #14226
    Billy
    Moderator

    I’ve had a quick look I’m at work at the moment

    All your values that I can see look correct can’t see all of what makes up R27 or the majority of the capacitors or TR1 value which should be a 503 code

    Your voltages are off the voltage designations are

    VA 9v

    VB half power supply around 4.5v

    VC 5v

    We can see IC3 works with Q2/3 collectors at 4.95v so VC is ok

    I’d check your power circuit first check what voltage you get on the 9v pad then through D4 the anode should be power supply then a little less at the cathode side then follow the schematic to the junction of R29 and 30 the voltage divider which should be half of your power supply (below) as you say Q1 drain voltage is way off hence why I’d start at the power supply circuit to see if your problem begins here, Q1 drain you’re looking for power supply less the voltage dropped through D4 around 0.7 to 1v

    Here’s what your voltages should be around

    Proper Q voltages for comparison (all pots full CCW)

    Q1: (7.27 – 3.71 – 3.23 – DSG)  Drain Source Gate not CBE

    Q2: (5.05 – 3.75 – 2.90 – CBE)

    Q3: (5.05 – 2.90 – 2.86 – CBE)

    What is the input impedance of your meter that would let us know the accuracy of your meter readings

     

    #14236
    feedback
    Participant

    Many thanks Billy this was very helpful. Knowing the power scheme has made troubleshooting much faster. Can you please confirm that the BIAS will only sweep the frequencies when Sens is at zero, and not when engaged?

    Ive been trying to google how to tell the input impedence of my meter but not finding anything specific. At max ohms (2m) touching the probes together gives a reading of zero.

    Initial probing shows that VA isnt connected to the pad on pin 1 of Q1 at all. Ive made jumpered connections and Drain is now at supply voltage (about 7v), but Im getting some inconsistent results on different points of the schematic. And the sens knob still doesnt appear to work. Also the Decay knob doesnt appear to change anything either but I dont know if thats another symptom of the same issue.

    Even with jumpered VA connections Im only getting 1.1V on Gate of Q1, and still getting low readings for Q2 and Q3.

    I have a feeling that the pads may be a little less robust than usual on this particular one board, or I was unprecedently rough with my component stuffing. Ive built over 50 pcbs at this point and have seen a couple of boards where the pads werent properly adhered to the traces or planes. Ive certainly lifted some pads and traces when doing aggressive troubleshooting, but not in the initial construction like on this board. Im ordering another and trying again and putting it down to experience.

    On the new board Ill build out the power rails first and make a post with readings to make sure im on the right track.

    Thanks again for your time Billy.

    #14240
    Billy
    Moderator

    You should find your meter impedance from the make and model basically you want a 10M ohm impedance for accurate readings

    BIAS: Controls the filter resonance frequency. When Sensitivity is turned fully off the Bias can be used as a sweepable filter.

    There’s some information here on the old site which runs through the circuit with the block diagram (below)

    https://guitarpcb.webs.com/<wbr />apps/forums/topics/show/<wbr />8230458-swah-super-white

    If you click on a link that doesn’t work just add .webs to the url eg guitarpcb.webs.com

    I wouldn’t give up with this pcb you can always bypass traces if pads are lifted by connecting components directly with jumpers as you have with Q1 the pads are also double sided so you can solder component side also if needed

     

    #14243
    feedback
    Participant

    Cheers, the block diagram is super useful.

    As far as the BIAS goes, I was just looking for a confirmation that it only sweeps when Sens is at zero, and doesnt sweep at any other Sens value. It specifcally states it does sweep at Sens zero, but not specifically when Sens is not zero, if you get what I mean.

    If true its just a very easy way to check if a change Ive made is actually making the Sens knob works, as BIAS currently sweeps at any SENS position.

    Ive spent a long time probing this board now, and ive identified more than one pad not attached to a trace, power rail or ground plane that should be. Ive made 5 jumpers already and have to assume that any pad could be suspect. Ill keep tinkering with when i have time though for my own education, im learning a lot about pedal craft here.

    Imn using a cheap multimeter, sounds like its  time to save for a more accurate one. Having it top out at 2m has been fine for synths but it seems pedals commonly have values far above 2m..

    Thanks again for the prompt support, its much appreciated.

    #14254
    feedback
    Participant

    I can confirm the boards ground plane is faulty. Unless Im reading the schematic wrong (which is entirely possible).

    Drain of Q1 has a high impedence connection to ground as does VA, pin 14 of IC2 and several other pads. This is making jumpering impossible. Unless youve got some other tips for this situation, its  time to move on to the next one until the replacement comes. Ill keep you posted.

     

    #14260
    Barry
    Keymaster

    If it helps I just checked the continuity of every pad across the board which is from the same lot as yours.

    The only pads connecting with ground are the ones that are supposed to and none of the ones you mentioned are connected.

    These boards are all fully checked from the factory as well as they probe every board for potential issues.

    I would mention that I mod my boards a lot trying out new ideas, values and potential mods and the two things that keep me from lifting a pad is resisting the temptation of using too much heat in tandem with repeated applications of the conical tip to the pad while trying to work a lead out and I use a Soldapullt III PT-409 desoldering pump which will remove all or most solder in a single try. It is so much better than all of the knock-offs.

    #14376
    feedback
    Participant

    Thanks again Barry. Its good to have that confirmation.

    Im at a bit of a loss here then. Id be more than happy to say that Ive butchered the board if thats what happened, but Ive built a lot of  pcbs successfully with this exact same setup, same iron, same solder. Basically several modular synths worth. I tend to use flux and wick rather than pump as its less likely to lift a trace with a cheap manual pump.  The only time Ive had a problem was when I was just beginning to use an iron or it was a board that later turned out to be verified defective where several pads had been joined to a plane either by a manufacturing fault or by fracturing the planes from heavy things stacked on top of them in shipment.

    I know you dont know my experience from a can of beans, and im not suggesting there is any kind of widespread manufacturing problem with these, just that this particular instance has been quite confusing. Im also not looking for a refund or a replacement from the reseller, im going to get a new board and try again and see what happens. If something similar happens then Ill have to revisit everything I thought I knew about soldering! I did butcher the board eventually, but i was very careful as always with the initial build, and the exact same symptoms persisted from the very first test before any later desoldering was done.

    For reference, heres the sort of denser synth pcb Ive built over 20 of successfully at this point, with absolutely no lifted traces, pads or problems with ground planes. This is much denser board, many more pads and smaller. I do SMD 0402/0603/0805 double sided boards by hand with a standard iron with this exact same setup and havent experienced these issues before apart from when I was learning which was many boards ago. If I was using too much heat as a general practice then I would be killing multiple sensitive components, pads and boards on these quite often.

    Board example.

    I do have a variable temperature iron, and Ill check that its not broken and pumping out max heat all the time.

    I do really appreciate the support and the time youve taken to help me through this.

    If you have a pretty definitive list of correct test point voltages for this circuit I can refer to next time while building, id be happy to write it up into a guide doc once im done.

    Can I pay pal you or a charity a couple of beers for going above and beyond with a painful build and crochety customer?

     

    #14377
    feedback
    Participant

    Just wanted to add the other interesting point that my issue was exactly as described in the earlier post I referenced, so I know this exact same set of symptoms has happened to someone else. That thread was never resolved and  I did reach out out to that forum member to see what the washup was but havent heard back.

     

    #14380
    Barry
    Keymaster

    The fact the v4 board tested fine is all I can tell you undeniably from my shop.

    As for my other advice I am not saying that you exactly lifted pads with heat, only that is a common that happens.

    The Soldapullt is a recommendation because I love it. $25 or less most shops.

    The other thread you refer to having the same exact issue had very different voltage readings so the cause is not the same.

    His are in black     vs.     yours in violet.

    Q1: 6.93, 4.17, 3.2   –   Q1: 0.03, 0.03, 0.59

    Q2: 4.98, 4.02, 3.58   –   Q2: 4.95, 1.03, 1.13

    Q3: 4.98, 3.94, 3.33   –   Q3: 4.95, 1.13, 1.30

     

    TL074: Pin 14: 4.35   –   1.6

     

    I only listed the obvious ones that would make a difference for brevity.

     

    The reason for pointing out the voltage readings here is so that people reading in the future won’t try to lump them together. Just so we are clear I am not trying to point at anything other than the difference.

    Forums are great for troubleshooting and we love to help but unfortunately most people do not bother to show off their successful builds and even worse can disappear forever when they fix a troubleshoot and don’t bother to let us know the results.

    Again anything is possible certainly and this is simply the best info I can give from my end.

    I would start by confirming the ground plane with a fresh board. I do like good lager and will drink one with you when you are done. I like them especially with a nice Corta Panda on the side.

     

     

    #14389
    Billy
    Moderator

    Q1 seems to be your main problem area

    What voltages do you get on the empty socket of Q1

    I notice you’ve soldered Q2 and 3 into their sockets?  Q1 socket looks different

    Do you have a DC jack wired or is it just the battery

    Probe the +9v and Gnd pads on the pcb what voltage are you getting there, is it still 6.6v

    This is how I’d normally use the schematic to try and highlight where my problem might be yours appears to be working if you have 6.6v going in and your VB is half that, which is correct with your VB points being 3.3v that tells us your voltage divider is working albeit not exactly the voltage we want but still half your power supply, your VC is correct with 4.95v at Q2 collector, so it appears VA just isn’t getting to Q1 drain so I’d check your VA point above R29 and Q1 drain out of socket where you should be getting whatever voltage is going into the pcb

    With your meter impedance it’s not absolutely critical as long as you take it into account

    Some info here from Tonmann on the old site

    https://guitarpcb.webs.com/apps/forums/topics/show/12988309-troubleshooting-opamp-circuits

    Well worth a read as it also outlines troubleshooting op amp circuits

     

    #14479
    Barry
    Keymaster

    Thanks for that info Billy.

    If it helps in regard to the newer v4 Swah board there is a thread of a successful build here:

    While the picture is hard to see you can tell it is a v4 board by the dual 9v & Ground pads in the top right corner which is only on v4.

    • This reply was modified 3 weeks ago by Barry.
    #14710
    dorian
    Participant

    hi,

    I have built the swah, as well, and now i am running in the same problem as “feedback”. My wah ist not making wah.

    I have cross check everything, and meassured all the voltages, which seem quite ok. All are a bit lower, but the divider is working.

    For my debuging, I used a sine wave generator and an osciloscope, so I have some Information about  what is happening in the envelop part, -(the wah part is working, as i can manually wahwah with the trim poti, when increasing the resistor r26, to make the trim poti the main input for the wah wah ic)

    But in the envelop part changes nothing when turning the sens knop. If I increase the sine above 1kHz it start boosting the signal but only when it is at the max setting. The boosting can be shifted with the trim poti, but only down to 1 khz, after that the filter/boost does not work anymore.

    I also tried with the guitar, and only get a clean sound.

     

    Has anybody encounter a similar problem and ny idea how to solve it. Are there any significant messure points to verify where the error is?

    I am happy for any help

     

     

    #14712
    dorian
    Participant

    and can anybody please verify the potis:

    1M reverse log (A) Decay

    100K linear (B) Sens

    Thanky you

     

    #14714
    Barry
    Keymaster

    P3 Decay is a 1M Rev Log C Taper.

    P2 Sensitivity is a 100K B Taper.

    These tapers are based on USA/Asia Manufacturer Tapers and not European if that helps.

     

    #14715
    dorian
    Participant

    Thank you for the reply.

    The potis are correct in the schematics, as suggested. And only in the manual is a twist.

    I am still testing the PCB, but I can not figure out what is wrong. I also checked the PCB lines and solder points, every thing has connection, so the PCB itself has no fault.

    What I see is that the boost is jumping when turning the SENS knob, so it is not slowly increasing or decreasing, but just jumping from boost to non boost state. When using the Trim poti as mentioned above I can drive the boost with theTrim poti to any boost amount.

     

     

    #14716
    Barry
    Keymaster

    I uploaded a new Build Document and made clarifications. I also added some other information

    I hope it will help.

    Make sure you refresh when downloading it since you will probably have a cached version in your browser.

     

    Here is a photo from a different SWAH v4 build that works you can use for reference to compare:

    Sofnwhat used a Buff N Blend in his build but that would not matter as far as the board itself.

    It is from this thread: https://guitarpcb.com/forums/topic/gpcb-mowah-and-swah/

     

    One thing to consider is check the Datasheet for the IC3 after verifying your part and make sure the pins match.

    The board silkscreen is marked correctly but I know some have the opposite pinout so be 100% on it.

Viewing 17 posts - 1 through 17 (of 21 total)
  • You must be logged in to reply to this topic.