Page 1 of 10 123 ... LastLast
Results 1 to 10 of 98

Thread: Sad Trombone heard when lighting up this megatree

  1. #1
    Join Date
    Nov 2016
    Location
    North Vancouver, BC, Canada
    Posts
    752
    Post Thanks / Like

    Default Sad Trombone heard when lighting up this megatree

    Pretty sure I heard the sad trombone sound when I lit this up....wah...wah..wahhhhhhhhhhhhh. Attachment 42196

    16 legs, 4 groups of 4. Each group is connected from the top and zig zags every 50, so 200 pixels per group of 4 legs. Connected to an e682 with 4 outputs. Injected from the same psu at the e682 at pixel 100 and 200. Injection wires are 14-16 AWG less than 3' runs. I ran a 10 awg wire from the psu that powers the e682 to a fuse block to distribute. So basically at the top of leg 2 and the end of string 4 are the injection points for 3 power points for 200 pixels. 50% power ws 2811 5v bullets. This is running test pattern from e682 and if you note, only the first string of each group lights up. Some almost top to bottom, others, only part way. example, first string only lights up like maybe 8 pixels. A bad pixel or connection sometimes...ok fine...But for the exact problem to occur across 16 strings seems unlikely, and if it was pure and simple voltage drop, it should be lighting up again at least somewhat near the injection points, even if it's dim or something. I wondered if this was something to do with the e682 only outputting test pattern for certain pixels on some outputs, so i tried in vixen with freshly patched setup and put on a set level red for every one of the 800 pixels....only the front string of group 3 lit up at all (although the whole thing). The e682 outputs are setup as 200 pixels each, zig zag first at 50, then every 50 after, ws2811. Colors are correct. The only "weird" thing about this setup, is due to varying lengths of the connectors on the e682 (when I had them cut to length of specific layout last year), I had to connect the 4 groups (using ray wu pigtails) to whatever cable could reach, so the outputs aren't exactly one big consecutive block (using outputs 1-2, 3-2, 3-4 and 4-4). I've checked and confirmed that there is 4.85v at the injection points. There is not a single pixel lit or flickering outside of the first 50 on each screen

  2. #2
    Join Date
    Nov 2016
    Location
    North Vancouver, BC, Canada
    Posts
    752
    Post Thanks / Like

    Default Re: Sad Trombone heard when lighting up this megatree

    If for some reason you are seeing invalid attachment I will try to post again, but the gist of the pic was only the first string of each group of 4 had any lights at all, strings 1,5,9,13. And at that, not only one was fully lit

  3. #3
    Join Date
    Nov 2016
    Location
    North Vancouver, BC, Canada
    Posts
    752
    Post Thanks / Like

    Default Re: Sad Trombone heard when lighting up this megatree


  4. #4
    Join Date
    Nov 2016
    Location
    North Vancouver, BC, Canada
    Posts
    752
    Post Thanks / Like

    Default Re: Sad Trombone heard when lighting up this megatree

    Ok string number one may be a bad pixel causing that extreme short light up, disconnection just point about 30% down from the top only putting out 2v, but the rest of them.... The injection point are reading 4.85v roughly

  5. #5
    Join Date
    Nov 2016
    Location
    North Vancouver, BC, Canada
    Posts
    752
    Post Thanks / Like

    Default Re: Sad Trombone heard when lighting up this megatree

    If anyone has any ideas at all, I'm all ears. Kinda stressing out since I'm committed to go for December 1st and this is holding back the rest of the build and test

  6. #6
    Join Date
    Dec 2011
    Posts
    6,072
    Post Thanks / Like

    Default Re: Sad Trombone heard when lighting up this megatree

    Vixen config ?
    e682 config ?
    Pixels wired in the correct input output fashion ?
    Power injection wired correctly ?
    Good solder joins and seal .

    Is this a case of eating an elephant in one bite ?
    Or were small nibbles taken before hand ? hmmmmmmmm
    Power injection should be verified / tested per output and not all at once .

  7. #7
    Join Date
    Jan 2017
    Posts
    439
    Post Thanks / Like

    Default Re: Sad Trombone heard when lighting up this megatree

    You tested all the LEDs before putting them up, correct?

  8. #8
    Join Date
    Nov 2016
    Location
    North Vancouver, BC, Canada
    Posts
    752
    Post Thanks / Like

    Default Re: Sad Trombone heard when lighting up this megatree

    Quote Originally Posted by Bwinter View Post
    You tested all the LEDs before putting them up, correct?
    Almost all, at least a majority of them

  9. #9
    Join Date
    Nov 2016
    Location
    North Vancouver, BC, Canada
    Posts
    752
    Post Thanks / Like

    Default Re: Sad Trombone heard when lighting up this megatree

    Quote Originally Posted by angus40 View Post
    Vixen config ?
    e682 config ?
    Pixels wired in the correct input output fashion ?
    Power injection wired correctly ?
    Good solder joins and seal .

    Is this a case of eating an elephant in one bite ?
    Or were small nibbles taken before hand ? hmmmmmmmm
    Power injection should be verified / tested per output and not all at once .
    Sorry, I'm Not entirely sure what that means lol. Eating an elephant in one bite. Thought I had covered a chunk of the findings, but every bit helps. I already described the e682 config. 4 outputs of 4 strings @50 each. Each string is confirmed to be set at 150 channels, and the universe assignment lines up. Vixen has had patching blown out as well as the controller config and restarted from scratch, and I manually patched each pixel rather than the parent items after re assigning the properties. The mega tree is confirmed to be correct as far as 2400 channels, and I've checked each string is in the right universe corresponding to controller. Also worth mentioning, if I have the controller page open and run any effect on the mega tree in vixen, then refresh the webpage, it shows packets received, so data appears to be transmitting from vixen to the controller. E682 is config to have 200 pixels per output, four strings of 50 zigzag every 50. Top down, bottom to top left to right. Injection, I ran a 10 awg wire from the same 60 amp, 350 watt supply attached to the e682, to a fuse block for distro. 14-16 awg injection wires under 5'. Testing the voltage at the injection points between string 2-3 of each group, and at the end, so power at 0,100, 200, get around 4.85v, so the power injection setup seems fine up to the transition. I'm injecting using the built in wires at the start of the string. I am using the jst connections built in to the string to connect them mostly (no sign of them being a problem, let alone for 12 of 16 strings) . Technically the e682 says that only outputs 3-1 to 3-4 should be used to output test patterns for ws28xx pixels, but 2 of the outputs aren't in that range and output the first string or part of it still. If power is the issue, I would expect to see even a partially lit pixel or something on one of the 2,3 or 4th strings in one of the 4 groups
    Last edited by XmasinVancouver; 11-27-2019 at 04:04 AM.

  10. #10
    Join Date
    Nov 2016
    Location
    North Vancouver, BC, Canada
    Posts
    752
    Post Thanks / Like

    Default Re: Sad Trombone heard when lighting up this megatree

    Quote Originally Posted by angus40 View Post
    Vixen config ?
    e682 config ?
    Pixels wired in the correct input output fashion ?
    Power injection wired correctly ?
    Good solder joins and seal .

    Is this a case of eating an elephant in one bite ?
    Or were small nibbles taken before hand ? hmmmmmmmm
    Power injection should be verified / tested per output and not all at once .
    Regretting my decision to run the controller box up top on the deck thinking everything would be nice and close and avoid long cable runs for other props as well. A factor was also thinking that I didn't want to have to factor in the controller powering part of the strings, then having to cut out and splice a different supply for injection to avoid crossing v+ from different supplies

Page 1 of 10 123 ... LastLast

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •