Maintenance notice: These forum archives are read-only, and will be removed shortly. Please visit our forums at their new location, https://www.evilmadscientist.com/forums/.

One very bright unresponsive row on Peggy 2

patholoraptorpatholoraptor
edited June 2012 in LED Matrix Kits
Hello,

I have half the LEDs soldered in to my Peggy 2, and it seems to be working well.  I can upload and run sketches, and everything seems to work, with the exception of row 2.   It's not that Row 2 is just a little brighter than the other rows, something that I understand can go away after completion of construction.  Rather, Row 2 stays entirely, brightly lit during running of the sketches with no discernable response at all to the sketch.  

I understand that the transistors control the rows, and I have checked the solder joints and don't see anything obvious.  Can anyone point me in a potential direction for troubleshooting?  Thanks in advance,

p

Comments

  • Windell OskayWindell Oskay
    It is likely that you have a short circuit somewhere in the control section that is holding the base of one of the transistors low-- or just connecting it to a neighboring signal.  This could be either at the transistor itself (which you have already checked) or at the pins of U2, the LED driver that controls the second row.   If you have access to a multimeter, you might be able to use that to help locate the issue.  
  • patholoraptorpatholoraptor
    edited June 2012
    Thanks Windell.  I am having some difficulty locating the short.  Also, in looking at the circuit diagram, it's actually row 1 (the second row of lights), not row 2 that has the issue.  A few things I discovered in working on this:
    • Using the continuity tester on my multimeter, when powered on, the + and - leads of the LEDS on Row 1 are continuous.  This is not true of any other row.
    • Row 1 is not ALWAYS all lit.  When the device is powered on, an apparently random pattern of LEDS in Row 1 sometimes light (usually in blocks of three or four) with a few LEDS in between not lit
    • My earlier statement that the LEDS show no response at all to the sketch may be incorrect.  I can see them flickering slightly but they are so bright it's hard to be sure.
    I don't know if this is possible to debug remotely, but if the above gives any indication as to where the short could be, could you please let me know? Thanks in advance. 

  • Windell OskayWindell Oskay
    >Using the continuity tester on my multimeter, when powered on, the + and - leads of the LEDS on Row 1 are continuous. 

    I'm not sure what you mean by "continuous" here.  The square-hole sides are connected together in every row, and the round-hole sides are connected together in every column.

    >Row 1 is not ALWAYS all lit.  When the device is powered on, an apparently random pattern of LEDS in Row 1 sometimes light (usually in blocks of three or four) with a few LEDS in between not lit  My earlier statement that the LEDS show no response at all to the sketch may be incorrect.  I can see them flickering slightly but they are so bright it's hard to be sure.

    Have you finished building the Peggy yet, or are you still trying to debug it while building?  If there's a pattern in the behavior, it is likely to reveal itself more clearly when all of the LEDs are in.  And, if something is flaky, try wiggling the different components to see if the transistor or one of the chips has an intermittent connection.

  • patholoraptorpatholoraptor
    Thanks Windell.  I am in fact still working on soldering in all of the LEDS; I have about half in at the moment.   I have narrowed down the issue more by running various sketches.  The behavior seems to be that if ANY LED in a given column is lit, the Row 1 LED from that same column will also be lit.  

    I will take your advice and continue to work on getting all the LEDs soldered, and see whether that makes anything more clear.  

    Thanks again.
  • patholoraptorpatholoraptor
    Almost done, and I have confirmed the above behavior--all LEDs work, the sketches run as they should, but Row 1 LEDs light when any LED from the same column is lit.  Row 1 is very bright compared to the others.  Checked transistor Q1, and I see nothing visibly wrong with it.  Checked the soldering on U2 and it looks OK.  Could I have a damaged transistor, or should I look elsewhere?  Thanks!

  • Windell OskayWindell Oskay
    If I understand correctly, whenever an LED lights anywhere, the LED in row 1 but in the same column as that LED lights up as well.  That suggests that the transistor on row 1 is stuck in the "on" position.  Using your multimeter, check to see if the base pin of Q1 (the left-most pin, connected to RB1) is stuck in the low-voltage position.  If so, it's likely that the transistor is OK, but is receiving a bad signal.  Check to see if that pin is electrically connected to nearby things.  Then, check the other end of RB1 to see if it might be shorted to pins 2 or 20 of U2.  It *should* be connected to pin 3 of U2, and that pin alone.
  • patholoraptorpatholoraptor
    In response to your suggestions:
    • The base pin of Q1 (left-most) is not connected to anything nearby (except RB1) that I can detect using the continuity function on my multimeter.
    • The other end of RB1 is connected to pin 3 of U2, and only pin 3.  Definitely not connected to pins 2 or 20.
    • Forgive me, but I'm not sure I understand what "low-voltage position" means.  I am not sure how to test this. 
    I discovered one other interesting thing using my multimeter:  the positive pin of every LED in row 1 is in continuity with the emitter pin (right-most) of EVERY transistor.  Does this mean I have a problem "downstream" of the transistor, or could the issue still be with the transistor?  Thanks!
  • Windell OskayWindell Oskay
    By "low-voltage position" I just meant that it was held electrically low (close to ground).

    > the positive pin of every LED in row 1 is in continuity with the emitter pin (right-most) of EVERY transistor. 

    This should not be the case; compare with the other rows.  The emitter of the transistors is connected to the positive power supply ("Vcc"), and Vcc should not (directly) make it into the LED matrix anywhere.  The likely issue is that the collector (middle pin) of Q1 is connected to the emitter.  Check on the top and bottom sides of the PCB for a short at Q1.  If you cannot find an external short, try replacing that transistor. 
  • patholoraptorpatholoraptor
    edited June 2012
    Regarding your suggestion about Q1, I double checked and I can find no evidence of a short between the emitter and collector pins of Q1.  I replaced the transistor once before trying to debug, but I will try it again in case somehow it turns out to be the culprit.    Edit:  I am an idiot.  I located the short between the collector and emitter of Q1, as you said I would.  The story of how I am only finding it now is too embarassing to publicly relate :)  Anyway, problem solved!  You are the man.

Sign In or Register to comment.