Leds lighting
  • throppthropp
    Hi guys, I'm having some issues at step 11 on digital control board.

    Ok so, it suggests that while testing it we should be able to see leds 3 and 7
    light up, and stay that way, and while rotating the selector watching other
    leds form patterns [??].

    1. What I had at first was no response on positions 3 and 7 but, while pushing
    buttons and rotating the selector, I got some response from all the others
    except 3, 7 and 8.

    2. After I swapped the power adaptor endpoint module to a larger one, I got signal
    from position 7, but no responce from 3 and 8. Still the buttons respond lighting
    up other leds and I can hear a few subtle sounds and lfo's, specifically from
    position 6. But, then again, only when I turn the mixer channel volume all the way up.

    3. It's not really clear if it's necessary to add those 2 little jumper wires
    on each side of the analogue ports [marked R1 and R2]. I've only soldered
    them.. maybe that wasn't even necassary, or it doesn't matter at all [?!?]; it's not clear.

    4. One last, extremely important, thing I noticed is that on the digital board, that guy has
    installed the IC socket [labeled IC4] with the notch indicating rightwards, while it is printed
    the other way around.. The correct thing, if we are supposed to strictly
    follow the board printed indications would be placing it the opposite way, as I did. I'm really
    surprised by the fact no one ever commented on that.

    5. Oh and yes, I certainly didn't go wrong placing the infamous shift registers IC2 and IC3
    [74hc165 and 74hc595] on their respective places.

    Looking forward to read you guys !! Thanks
  • fcd72fcd72

    3. you MUST solder the bridges or put a Pot in, or theres no connection to the audio jacks, hence no sound.

    While you are at it check that theres a Jumper on the Pole Selector Pins….

    4. The orientation of the Socket is meaningless unless you put the IC in the right way.

  • pichenettespichenettes

    1. 2. Must be that the LEDs are wired backwards ; or that some pads on the 74hc595 are badly soldered.

    3. If you do not wire pots, solder bridges. If you wire pots, you do not need them (or else pots won’t do anything).

    4. Not a big deal… What matters is the orientation of the IC itself, not the socket.

  • throppthropp
    ok I did some resoldering, but now while I'm testing, leds 7 and 8 are on by default not 3 and 7.
    I noticed that IC1 (lt1054) on filter board gets really hot after a whilet, is it normal ?
  • pichenettespichenettes

    Have you checked for shorts between GND and the +5V / – 5V rails?

  • throppthropp
    the only short I can see but I guess it is printed that way is between sw and 1N4004 diode. Where should I be looking ?
    Do you mean the regulators ?
  • pichenettespichenettes

    Use a meter in continuity testing mode and check if there is any short circuit between GND and points at +5V / -5V. Have you performed the voltage checks in the assembly instructions?

  • throppthropp
    I ran the continuity test and when I probe pin marked with + sign on 7905 as well as on pin marked with - on IC 7805, I get continuity sound...not good news is it ? or is
    it irrelevant ? In case it is bad news, any thoughts on how to fix it, because I can spot any shorts with naked eye.
  • throppthropp
    sorry I meant I CAN'T spot, or cold joints whatsoever
  • pichenettespichenettes

    There shouldn’t be any short between the +5V rail and ground, and the -5V rail and ground. Does your meter just beep or does it show a resistance value on its screen?

  • throppthropp
    Yep, 003..!
  • throppthropp
    It does both, it beeps and it countdowns till it stabilizes on value 003.
  • pichenettespichenettes

    Then you have a short between GND and the supply rail(s) unfortunately. You have to look for solder blobs adjacent to both a +/- 5V pin and ground. Or for ICs inserted with the wrong orientation.

  • throppthropp
    I checked LT7905 and LT 7805 datasheets, I was mistaken about what I said above, I beeping sounds out of the ground pins which is normal.
    I also probed again using continuity test, the values suggested on step 4 of the SMR4 board, everything is clear. No "illegal" beeping sounds. Can think of anything else going wrong such as malfunctioning part, maybe a specific IC could be held responsible for such behaviour ?
  • throppthropp
    hello, are you here ?
    I will work on the shruti this afternoon. Can you please answer me because I'm lost, the shruti is almost finish and I can't go any further without help.
  • throppthropp
    Could it be a broken LT1054 ?
  • 6581punk6581punk
    LT1054 is for providing a negative voltage from a positive one, mainly for filter boards than use an OP amp as they require positive and negative voltages. If there was a problem with power to the digital board then it wouldn't function at all.

    Are all the joints on the 595N chip okay? no shorts? sometimes you can't see the short as it is very slight, especially on the black boards.

    I suppose you could also measure the output at pins 2 and 6 of the 595N chip to see if it is outputting anything, if it isn't then there's a problem around there.
  • throppthropp
    Thank you for providing an answer! I was thinking the same about LT1054, but was annoyed by the fact that it raises some high temperatures
    really fast.

    Till now, I have checked every each one of the joints many times; I have even resoldered some suspicious ones. Always using a magnifier!

    So, I checked pins 2 and 6 of the 595 shift register and I get an output, that means things are clear down there. But, I'm guessing it's got
    to be a "routing" problem 'cause, testing it at step 11 now, I'm getting LEDs 7 and 8 on by default and if I press button 3 or 5 then it switches
    to 3 and 7..
  • 6581punk6581punk
    Well if all the joints are okay, the LEDs are 100% correctly orientated then the only thing left to blame would be the 595 chip itself.

    Are you 100% sure the LEDs are correctly orientated? you can usually see a little triangle at the side of an LED and they should all point the same way. They're diodes and will pass current one direction only. So you should get little resistance one way and high resistance the opposite direction. This should be fairly consistent with all the LEDs.
  • throppthropp
    Hmm, I hope 595 is to blame. Much cheaper than LT1054!! But I'm not sure.

    And yes, I checked the orientation of every LED; because,apart from the triangle thing, when I use the trigger
    or push the buttons, each position corresponds by lighting up the respective LED.

    Just boot test fails by starting with LEDs 7 & 8 and not 3 & 7..

    Is there something that could interfere in such a manner? Apart from cold joints or stray blobs, which I've checked
    for a million times already.

    Thanks again for helping me out, I'm pretty lost at this point.
  • 6581punk6581punk
    Well, it's a digital process. The processor sends serial bits to the chip and that then converts it to a parallel output.

    Digital stuff doesn't tend misbehave unless the chip is faulty or there's a problem with the circuit, short or open circuit etc.

    When you said you get an output at pins 2 and 6 what sort of voltage was that? compare the output to a LED that works, pin 1 for instance.
  • throppthropp
    You're my hero for helping that much!

    Ok so, on my multimeter - on 200m setting - I get 1.3 out of pin 1, 1.1 out of pin 2 and 0.6 out of pin 6.

    Is this normal?
  • 6581punk6581punk
    200m? is that a milli volts setting? (ie. 0.2v max).

    You want to measure the voltage output of those pins, so place the black lead on gnd and then the red lead on the pin you are testing.

    I would expect about 2v or so at the LED after the resistor. Seems to be the usual requirement for common LED colours.

    I'm trying my best, but remote diagnostics is tricky with software, never mind circuits :)
  • throppthropp
    Hi there again (and again and again..!).
    Now I can confidently state that I've checked everything and I'm getting the right values.
    All LEDs read about 2 and so on.. I just can't make out why I'm getting positions 7 & 8
    lighting up by default, and not 3 & 7. Doesn't it seem like it's got sth to do with the shift
    registers?
  • throppthropp
    By the way Pichenettes, it seems like you are the main guy who runs this company, so I really don't understand why you don't pay no more attention to my question ?! Maybe I'm mistaken...
  • fcd72fcd72

    No need to get upset, thropp…

    You bought a KIT, that means you are by yourself responsible for the correct assembly, see the Building Instructions.

    Personally even if i know a bit about the Shruthi here and there i cant follow the conversation quite well.

    Youll have to be a bit more precise, “All LEDs read about 2 and so on” is hard to follow, my LEDs dont show any numbers, and if you measure with an instrumen then its 2 of what? Ampere? Volts? Farad? Relative Moon Humidity? I guess it means you measure 2 Volts at the Point between the LED and the Current Limiting Resistor 220R, right? But i had to invest 6 Minutes of my time to get a brief glimps of what your problem is, time i could have spent with my Son brushing his first new Teeth. See what i man?

    Precise Questions and precise Observations lead to precise Answers, with the information you provided i cant tell more than its some problem around the 595, all else would be wild guessing. Here are many many nice People, all willing to help if only they could decipher what people mean – and we all have no time to crawl in someone leses head to understand. So try to speak the language used here and useful help will result.

  • roschrosch

    do you actually have the right voltages at the power supply now? (-5V, GND, +5V)
    test it without the digital board connected first. if the analog board says ok you can connect the digital one and check the power rails again.
    it’s always the first guess or hope that there may be a broken IC, and sometimes there is. but most of the time it’s bad soldering, or wrong connections. or missing parts, like jumpers.

  • throppthropp
    @fcd72:
    If you were willing to understand, maybe you would've spend less time just by taking time to read the whole post and make out
    where "it reads 2.." corresponds; it doesn't take a lot to realise that it actually corresponds to volts; not relative moon humidity,
    not vegetable or mambo jumbo! Either way, you didn't have to reply just to mock me, as I'm looking for some sort of help, and
    that didn't help a bit. But, no hard feelings, you could have as well noticed that I'm keeping it very gentle towards every help
    I'm getting. I'm just looking to get the job done.

    @rosch:
    Yes, I get all the right values for voltages, even run through some datasheets to find out which pins output what. I also took really
    good care of soldering and orientation, jumper and stuff. I'm getting to that it's got to be a broken IC, specifically a shift register,
    but that's just a wild guess. I'm counting on that someone will verify that sooner or later just by the behaviour (LEDs 7 & 8 on, bla bla..
    everything I've been mentioning above). Or maybe I'll just go buy a couple of shift registers, if nothing..
  • fcd72fcd72

    @thropp
    if its a problem for you just actually write you measured 2V (1 character) instead of “read about 2” and expect people to read 2 Pages just to understand what you did it will be hard to solve your problem. I replied because Olivier already diagnosed your problem and hinted you precisely to the Solution. So “willing to understand” is a bit more a problem on your side – no offense intended, just my 1,7 cents.

  • throppthropp
    Not the whole post, just the above comment by user 6581punk saying "I would expect about 2v or so at the LED after the resistor"..
    so simple. And yet, you keep this up.
    And please, in case I didn't get it (as the guy who kept helping me after Olivier's suggested solution), where's the precise solution?
    Because you can see my replying saying no, I have no shorts 'cause I'm reading the right values, just a post bellow.
    Thanks anyway.
  • pichenettespichenettes

    First, check very carefully the soldering of the encoder and the CPU pins it is connected to.

    Then, boot the unit once with the encoder depressed and see if it makes the “wrong” LED pattern disappear.

    It looks like you had booted the unit with the encoder depressed at least once. This has the effect of putting the unit into “preset mode” in which it boots from the patch presets page (LEDS 7 & 8) instead of the filter page (LEDS 3 & 7).

    Also, have you tried hooking up a MIDI source and see if you get sounds out of the unit (or at least on pin OSC> of the control board?). That’s a very reliable way of checking if there is any serious problem with the control board :)

    Another way to confirm that the 74hc595 and MCU are fine is to boot the unit with S6 pressed. If leds 1, 3, 5, 7 light up you are good to go (just make sure you don’t send SysEx to the unit in this mode).

    I apologize for not answering to this message, but it is just not right to expect me to troubleshoot everyone’s build on this forum. First, I have a very lousy experience with troubleshooting Shruthi-1s (built 100+, only 2 minor problems), and then my approach to troubleshooting (scope + logic analyzer + writing custom test routines for the MCU + optionally opening the spice model) doesn’t translate well to people with less equipments. That’s why I count on you people to help each other, because you have a good feeling for the kind of mistakes you make and can think from the other builders’ perspective. I’m more of a designer/engineer so it’s hard to see a broken kit from the same angle as you!

    Last thing – 3 years+ doing electronics and buying heaps of parts, I’ve never seen a faulty IC. Guys are ST, TI, Atmel… rule, we are the ones who suck :) Think of 74xxx or 45xx as screws or nails. Malfunctioning screws or nails?

  • 6581punk6581punk
    I've run a bunch of 74 series chips with the connections totally wrong, they sort of did something and got very hot. I didn't damage them.

    Both the 165 and 595 chips point in the same direction.
  • throppthropp
    OMG!! Genial! That did the trick, both booting with the encoder depressed and booting with S6 depressed
    replied exectly as you stated! Now, I'm almost dumbfounded, I don't know how to thank you.

    That's what I was relying on, and sorry if I put some stress on your schedule, but as you can see
    you provided a very efficient troubleshooting step just from line one, so I was really hoping to get
    a reply by the guy behind shruthi, and that seems to be you!

    Thanks everyone for all the patience and valuable information, now I can proceed and enjoy my shruthi!
    ps.I never knew about that fact behind shift registers, put like that, yes, of course you can't have malfunctioning
    screws; only obviously broken!

    Thanks a million!
Add a Comment