2811 MODULE WIRING SETUP AND PRETESTING OPTIONS

Discussion in 'RGB Lights - Intelligent Pixels and 3-Channel RGB' started by talk2coxie, Jul 8, 2013.

  1. talk2coxie

    talk2coxie Full Time Elf

    Joined:
    Dec 29, 2010
    Messages:
    137
    Likes Received:
    0
    Location:
    Heathcote NSW
    Find Me On:
    Hi Guys,

    Like many others I have bought some 12V 2811 modules for candy canes.

    http://www.aliexpress.com/item/LED-digital-module-WS2811IC-5050-3LEDS-DC12V-input-waterproof-20pcs-a-string-256-gray-scale/578798417.html

    I have also soldered waterproof pigtails as per sketch below.

    https://dl.dropboxusercontent.com/u/43251051/CANDY%20CANE%20POWER%20INJECTION%20V1.2.jpg

    A few questions

    1 - Any unforseen issues with the way have I wired up these modules?
    2 - Is there anyway to pre test the wiring and soldering from the 4 pin connectors using a multimeter before powering up?
    3 - If so what do I look for and how can I do it?

    I already have the candy canes completely built and am hoping to avoid having to replace any damaged modules should I not have soldered the waterproof connectors on correctly.
     
  2. ecbailey

    ecbailey Full Time Elf

    Joined:
    Dec 26, 2011
    Messages:
    127
    Likes Received:
    0
    Location:
    Armadale, Victoria
    As the V+ and V- (ground) have been wired in parallel, checking the resistance (ohms) between the V+ points or the V- points with a multimeter should give a low or zero resistance reading. There should be substantially more resistance between a V+ and V- point, if not infinite.


    I'd just plug them in one by one and give it a shot. Whilst I haven't used pixels myself, I think that if something is wrong, it should only take out a single pixel, not the whole lot.
     
  3. DeeJai

    DeeJai Is that Magic Smoke?!?

    Joined:
    Oct 18, 2011
    Messages:
    1,007
    Likes Received:
    0
    Location:
    No longer in Radelaide
    the only pretesting you will sucessfully accomplish is to ensure there is a circuit through each the +ve and -ve tracks. so using your multimeter, attach it to the +v at the very start of your run, and the +ve at the end, see if you get a throughput. do the same for the ground.


    How many canes are you making, containing how many pixel modules each and how spaced out are the cances? Im not sure the power injection will be required as you have it setup, (although, depending on # of pixels and distances that could change)

    I tend to inject every 50 pixels or so, and looking at your setup, that would be every 5 canes (assuming 9 pixels each) You may be better off just running a seperate injection cable from the controller to the 6th cane and adding an injection port there. Just a thought


    For my canes, I also have both of the plugs coming from the same point (at the bottom of the cane), this prevents the need to have unsightly cable drops from the top.
     
  4. Boof63

    Boof63 Full Time Elf

    Joined:
    Nov 7, 2011
    Messages:
    486
    Likes Received:
    7
    Location:
    Kadina
    I used same Modules for 8 canes x 8 modules, Only needed to inject power at end of the run , where it joined and ran into next element.
    cheers Boof63
     
  5. OP
    OP
    talk2coxie

    talk2coxie Full Time Elf

    Joined:
    Dec 29, 2010
    Messages:
    137
    Likes Received:
    0
    Location:
    Heathcote NSW
    Find Me On:
    Thanks guys, seeing there was no real way to test I've just been plugging in and finding out my issues.

    I've only tested a few elements so far but there seems to be an issue elsewhere.

    Matt was a huge help in chat a few nights ago but I think I'll need to drop into chat and hope someone with a similar setup or experience with a PIXad8 and 2811 modules can have a look through my setup...... I now have team viewer installed which Matt helped me do as well.

    Cheers, Coxie
     
  6. OP
    OP
    talk2coxie

    talk2coxie Full Time Elf

    Joined:
    Dec 29, 2010
    Messages:
    137
    Likes Received:
    0
    Location:
    Heathcote NSW
    Find Me On:
    I have been having issues with the 2811 modules behaving properly.

    I think it is narrowed down to the Pixad8 setup somehow.

    A good string of 9 modules worked with 30cm of cable from the controller but any further length data cable the light did not behave properly.

    With the working short cable modules I then connected via daisy chain another 10 modules. All 19 modules worked nicely when applying a chase using da_131 and also worked nicely with single colours red, green, and blue.

    Once two colours (or white) were applied to the 19 modules, they went all different colours and stayed on.

    I have tried several speed options for the data ranging from 1000 to 3200 with no change to the above testing.

    Not sure if anyone else has seen similar issues?

    I have also attempted using some 12V 2811 strip but had absolutely no luck with that at all.

    Any suggestions to fix are welcomed.
     
  7. kane

    kane Dedicated Elf

    Joined:
    Dec 23, 2010
    Messages:
    1,019
    Likes Received:
    0
    Location:
    Trigg (Northern suburbs of Perth)
    Find Me On:
    How old is the PixAD8 ? Is it one of the first ones?

    This is going back a bit, but there some issues with the buffer chip used on the first batch, which didn't affect 6803 and 2801, but did affect tm1804 chips (which I use quite a bit, and are similar to the 2811)..

    I had to replace the buffer chips with a different model, and this fixed a bunch of crazy issues, and also gave me a lot more distance to the first pixel.
     
  8. OP
    OP
    talk2coxie

    talk2coxie Full Time Elf

    Joined:
    Dec 29, 2010
    Messages:
    137
    Likes Received:
    0
    Location:
    Heathcote NSW
    Find Me On:
    Hi Kane,

    The controllers are 2 years old and Eddy was good enough to upgrade them to support the 2811 and has the 2.7 firmware.

    Interesting to hear similar issues, thanks for your reply :)
     
  9. kane

    kane Dedicated Elf

    Joined:
    Dec 23, 2010
    Messages:
    1,019
    Likes Received:
    0
    Location:
    Trigg (Northern suburbs of Perth)
    Find Me On:
    About two years ago sounds about when they were first released.. When FE upgraded your board, he wouldn't have touched the buffer chips - he would have just flashed the firmware.

    Most people didn't have any issues with them, but that was mainly because at the time, most people were using 6803 and 2801 - I was one of the only silly ones playing with 1804. They definitely helped me at the time, so worth investigating... Ed@j1sys changed the buffer chips on later orders to address the issue

    I do have some spare of the buffer chips at home somewhere, and can send them to you if you like. Otherwise, these are the ones I bought - $0.60 each, and I'm pretty sure you only need 2? RS components do free next day delivery, so for a couple bucks, it's probably worth a try! I probably won't get a chance to get to the post office for another week, so ordering from RS (or elsewhere) may be a good idea.

    Link to RS Components
    (RS Stock Number: 126-7399)

    Cheers

    Kane
     
  10. OP
    OP
    talk2coxie

    talk2coxie Full Time Elf

    Joined:
    Dec 29, 2010
    Messages:
    137
    Likes Received:
    0
    Location:
    Heathcote NSW
    Find Me On:
    I'm not really sure what this tells me but I also have some 6803 strips which the PIXad8 ran without issue in last years show

    http://www.aliexpress.com/item/5m-LED-Digital-strip-with-controller-with-6803-IC-30pcs-5050-SMD-per-meter-DC12V-input/306377966.html

    The exact same communication error appears with this strip as well.

    I setup the PIXad8 for the 6803 strip (5 Metres) through the browser and with 30cm of cable between the controller and first pixel in the strip. The strip performs fine through a chase and each colour when testing and also tests well with 2 or more colours using da_E131 for testing.

    I then used a 1.2M extension cable and the strip performs well with single colours and a chase sequence with da_E131 but with 2 or more colours the strip has random colours across the 5M strip.

    Am I right thinking it is an issue with data signal strength or noise? Any suggestions to test to fix or confirm?
     
  11. kane

    kane Dedicated Elf

    Joined:
    Dec 23, 2010
    Messages:
    1,019
    Likes Received:
    0
    Location:
    Trigg (Northern suburbs of Perth)
    Find Me On:
    The fact you're getting the same issue with 6803 probably would probably rule out issues with the buffer chips (although for a couple bucks, I would swap them over to rule that out completely)

    Have you tried different outputs on the pixad8, in case it's isolated to one of the outputs?

    Also, the cable that doesn't work - what sort of cable is it, and are any of the cores (especially data or clock) doubled up? Have you tried a different cable?

    And I'm sure this isn't the case (given it works for the shorter cable), but do make sure you have nothing else on the network that could be sending e1.31 data.. I had wierd issues when testing last year when LSP was loaded in the background (not even outputting), where it was sending random e1.31 packets and causing these sorts of issues.

    Cheers

    Kane
     
  12. OP
    OP
    talk2coxie

    talk2coxie Full Time Elf

    Joined:
    Dec 29, 2010
    Messages:
    137
    Likes Received:
    0
    Location:
    Heathcote NSW
    Find Me On:
    Hi Kane,

    Have you tried different outputs on the pixad8, in case it's isolated to one of the outputs?
    Yes, I have tried the "good" strings and strips in the outputs which were giving me bad results. The issues followed the "bad" results. eg, output #3 worked with the strip going straight into the controller but not with the 1.2M extension, same result in output #4.

    Also, the cable that doesn't work - what sort of cable is it, and are any of the cores (especially data or clock) doubled up? Have you tried a different cable?
    Thought the extension cable might be the problem so made up another one of them too. The extension cable is 4 core 14/20 alarm cable but haven't tried different rated cable. I have some cat5 and thicker gauge speaker wire to test if that is worthwhile?

    And I'm sure this isn't the case (given it works for the shorter cable), but do make sure you have nothing else on the network that could be sending e1.31 data.. I had wierd issues when testing last year when LSP was loaded in the background (not even outputting), where it was sending random e1.31 packets and causing these sorts of issues.
    Still a valid point, but I don't have any other software running which outputs e1.31 to my knowledge. I have had outlook and windows explorer open but not sure if they would interfere?

    More than happy to hear all suggestions and test/rule out whatever I can.
     
  13. kane

    kane Dedicated Elf

    Joined:
    Dec 23, 2010
    Messages:
    1,019
    Likes Received:
    0
    Location:
    Trigg (Northern suburbs of Perth)
    Find Me On:
    About the only thing you haven't been able to isolate is whether there's something funny going on with the PixAd8 itself. If you had a second one, you could see if you get the same results with that..

    In terms of cable - I can't see that being an issue - there could have been an issue with one, but if you've tried another, then that's unlikely to be the issue.

    You'd know if you had any other programs loaded that output e1.31 - certainly outlook and explorer don't.. (the only things that would are things like LSP, LOR, LFP, etc, plus any test utilities like da_131.. Basically, to be sure, you'd only want one of them loaded up at a time.. But I don't think this is your issue.

    You said Eddy updated the firmware - perhaps there's a big with the newer firmware?? I'm not sure if it'd be a straightforward process to downgrade the firmware to rule this out?
     

Share This Page