Results 1 to 5 of 5

Thread: Reversing injection polarity consequences

  1. #1
    Join Date
    Nov 2016
    Location
    North Vancouver, BC, Canada
    Posts
    347
    Post Thanks / Like

    Default Reversing injection polarity consequences

    A little annoyed here, due to my schedule, I've only been able to work on the display when it's pitch black out. I have 8 mini trees all connected through one output on the e682. The first 2 trees are powered by the controller, but between the 2nd and 3rd tree, I kept the negative and signal running through from the 3rd to 8th straight through, but cut out the power wire after the second tree powered by the controller. This is because I was going to connect a seperate power supply for power injection at the start of 3, 5 and 7. These are 5v pixels ws2811. My understanding was that I need to do this to ensure that the V+ line from the two power supplies NEVER connect, but leaving the v- and data in place straight through so the power injection connects the v- between controllers. Anyways, working in the dark in freezing conditions, I somehow managed to reverse the power injection wires for the 3 points after the first 2 trees powered by the controller. No reaction, smoke or signs of anything whatsoever, but when running the show, only those first 2 trees are lighting up whatsoever. This is only a 5v 22amp max PSU. I shut it down, and reversed the power injection wires to correct polarity, but still, no activity on the last 6 mini trees, while the first two powered by the controller work fine still. So #1) are those pixels fried as a result? I thought that would be the case, but it occurs to me, with the negative v- common through all the trees, while reversed, i would have been feeding v+ into the v- line of the first two trees, and yet, the first two trees still work fine. Also, it's a hard cutoff. The last pixel on the second tree is fully lit up and functional, the first pixel on the 3rd tree with the power injection is fully dark. I've double checked patching, all seems well, and even tried reversing the string order to see if it was patching, and it wasn't. In that scenario, testing fx on tree 7 and 8 now lights up, and trees 1-2 do not, which is the reverse of what happens. So patching appears to be ok. So what's going on here?

  2. #2
    Join Date
    Oct 2018
    Location
    Sherwood Park, Alberta, Canada
    Posts
    43
    Post Thanks / Like

    Default Re: Reversing injection polarity consequences

    If your using power from the same power supply for power injection you don't need to cut the V+. The V+ only needs to be cut if your using a second or third And so on power supply.

    Do you have inline fuses on your power injection lines?

    And you might have poped the fuse that's your PSU. Did you confirm you have power from your second power supply? What kind of power supply are you using?
    Last edited by Jyman; 12-06-2018 at 05:11 PM.

  3. #3
    Join Date
    Nov 2016
    Location
    North Vancouver, BC, Canada
    Posts
    347
    Post Thanks / Like

    Default Re: Reversing injection polarity consequences

    Quote Originally Posted by Jyman View Post
    If your using power from the same power supply for power injection you don't need to cut the V+. The V+ only needs to be cut if your using a second or third And so on power supply.

    Do you have inline fuses on your power injection lines?

    And you might have poped the fuse that's your PSU. Did you confirm you have power from your second power supply? What kind of power supply are you using?
    yes, it's a second power supply, hence the cut V+. it's actually a pc power supply, and it fused through a distribution block. It does stay on during testing for an extended period. Or at least it did. There's no light to show power is live on this unit and it was dark out. So I'll be checking this as a first consideration when I get home (writing this from work). With these 5v power and pixels, does reversing that injection fry the pixels normally? I had thought I read something like that, but again, like I said, technically, the v+ from the injection would have been tied to the v- from the 2 working trees, feeding it, and those are still working. Don't know if that means anything though

  4. #4
    Join Date
    Nov 2016
    Location
    North Vancouver, BC, Canada
    Posts
    347
    Post Thanks / Like

    Default Re: Reversing injection polarity consequences

    Quote Originally Posted by Jyman View Post
    If your using power from the same power supply for power injection you don't need to cut the V+. The V+ only needs to be cut if your using a second or third And so on power supply.

    Do you have inline fuses on your power injection lines?

    And you might have poped the fuse that's your PSU. Did you confirm you have power from your second power supply? What kind of power supply are you using?
    yes, it's a second power supply, hence the cut V+. it's actually a pc power supply, and it fused through a distribution block. It does stay on during testing for an extended period. Or at least it did. There's no light to show power is live on this unit and it was dark out. So I'll be checking this as a first consideration when I get home (writing this from work). With these 5v power and pixels, does reversing that injection fry the pixels normally? I had thought I read something like that, but again, like I said, technically, the v+ from the injection would have been tied to the v- from the 2 working trees, feeding it, and those are still working. Don't know if that means anything though

  5. #5
    Join Date
    Nov 2016
    Location
    North Vancouver, BC, Canada
    Posts
    347
    Post Thanks / Like

    Default Re: Reversing injection polarity consequences

    Quote Originally Posted by Jyman View Post
    If your using power from the same power supply for power injection you don't need to cut the V+. The V+ only needs to be cut if your using a second or third And so on power supply.

    Do you have inline fuses on your power injection lines?

    And you might have poped the fuse that's your PSU. Did you confirm you have power from your second power supply? What kind of power supply are you using?
    So, an update. Good and bad. It looks like the original PC power supply that had been staying on and tested pretty stable outputting 5v is no longer alive. Don't know if that's a result of the crossed wires or what, but there's no power or signs of life from the original PC power supply. So there's problem 1 resolved. I am not able to get a proper meanwell style PSU or knock off without having to wait 10-14 days for delivery, when a huge chunk of the display season is already gone and wasted. So I found another Pc power supply that I knew was working. I soldered together 3 red 5v wires that I cut from the ATX connector. I then soldered those 3 wires to the existing rig in place, which was an inline fuse that connected to a distribution block with 8 possible output points, but I only have 3 wires connected via spade crimps, and each of those outputs are then fused in turn. The other 5 outputs are left empty without anything connected or any fuses. So basically, I left everything as is, but cut out the old PSU on the other side of the inline fuse, and reconnected 3 joined red 5V+ wires to that. I then snipped 3 black ground wires from the ATX connector, and connected each to an individual wire connected to the injection negative points. Jumpered the green socket to black to trick it to start up. When plugged in, the led on the supply is on, fan runs, and without any further modifications or additions, stays on seemingly. All good right? Not exactly...

    The first 2 trees are powered from the controller if you recall, then the red power wire is cut in between the 2nd and 3rd tree, as power injection points from the pc supply take over powering the last 6 trees @ 35 pixels per tree. The green data and negative white wire from the pixels remain connected from start to finish. When I power up the PC supply, there's definitely power getting there, but it's messed up. The mini trees that are powered by the pc supply above via injection points (trees 3 to 8) aren't acting right. The injection points are at 3, 5 and 7. Right away when the pc power supply is plugged in, trees 5 through 8 all light up a fairly bright white. This is despite the fact that there is no data signal or output being fed to the trees. I've also tried unplugging the controller power to ensure that no data or other factors were involved, and no change. No matter what, as soon as I plug in the pc power supply, the last 4 trees instantly light up white and stay on. Trying to play a sequence or shut down vixen, reboot controller seems to have no impact on them. They just stay on white and ignore any data being sent from the controller. This is ONLY the last 4 trees of the 6 powered by the pc power supply. Tree 3 has about 10 pixels on the side closest to trees 1 and 2 (the ones that are powered from the controller and work perfectly) that works perfectly, and then the remaining 25 pixels on tree 3 will only light up red when the tree is supposed to be triggered, no matter what the color should be. And not all 25 activate at the same time, probably half spread out over the 25 pixels light up red everytime the tree should trigger. Tree 4 has 1 single pixel at the very end, pixel 35 that's lit red and stays on. No other pixels on that tree seem to react when it should be lit. Then the very next pixel after that solo red one is the first pixel of those trees that are permanently turned on white. Sorry for the probably dense explanation, but I'm at a loss as to what is happening here. The pc supply worked perfectly with no signs of issues before this. what's particularly odd to me, is that there is that first tree in the power injection section, tree 3 with 1/3rd of it working perfectly, then the rest of it showing only red random pixels, followed by the next tree that's all off, no lights except for the one red pixel at the end, and THEN the 4 bright white perma lit trees. The signal and direction being messed up, then basically dead, then suddenly bright white full power doesn't make sense to me

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
  •