Current draw from controller output with & without power injection

Discussion in 'RGB Lights - Intelligent Pixels and 3-Channel RGB' started by nutz4lights, Jan 13, 2014.

  1. nutz4lights

    nutz4lights Full Time Elf

    Joined:
    Dec 12, 2012
    Messages:
    305
    Likes Received:
    0
    Location:
    Melbourne, Florida
    Hey all,

    This has been on my mind so I figured I would pose the question here. I plan on making some measurements later today, but figured I'd get the thread started. This is based on my recent discovery that the Technicolor 12V pixels actually draw 0.6W per pixel even though the spec sheet states 0.3W. My original plan was to run 150 of these from a single output of a p12s injecting power with a "tee" setup between pixel 100 & 101. I figured, 0.3W per pixel is 30W for 100 @ 12V = 2.5A out of the controller and then another 1/2 that (50 pixels) injected at pixel 100/101. Now that I know these are 0.6W per pixel, that means 60W going out of the controller, which is 5A (a perfectly fine value on my p12S, but not quite so good on an e6804 which is fused at 5A)...

    So, I got to thinking... what really happens to the power draw out of a controller when you inject power? On my megatree this year, I ran a 50 pixel string up and another 50 pixel string down, supplying power and data at pixel 1 and then just injecting power at pixel 100. I never bothered to measure the current draw on that power injection line, or the line out of the controller for that matter. So let's say I would have done that megatree without power injection. Those were 5V 2811 pixels rated at 0.3W per pixel. 100 pixels off one output would have been 30W and subsequently 6A of power draw. Six outputs would have been 36A on one bank of the p12s controller which would have blown the 30A fuse. That didn't happen, so obviously, by injecting power at pixel 100, I also have pulled some of the current load off of the controller. I am passing current into the 100 pixel string from the pixel 1 direction out of the controller and then from the pixel 100 direction direct from the power supply.

    My curiosity is, do the two legs supply equal current? That would be 50 pixels @ 0.3W per pixel = 15W which is 3A at 5V... and for 6 outputs on the p12s that would have totaled 18A... much more reasonable, and another 18A direct out of the power supply. To be completely fair, I have measured the current draw on those 5V pixels already and only saw 1.4A of current draw for 50 (didn't measure the 100 pixel setup).

    Now back to the more interesting scenario (currently). What happens to power draw when you inject between pixel 100 & 101 of a 150 pixel string? If the current draw on a 100 pixel string flows current back into the string from the pixel 100 direction, then theoretically injecting in between pixel 100 & 101 of a 150 pixel string would probably flow current back into the 100 pixel section of the string AND into the pixel 101-150 section of the string on that injection line. So, the controller would output 50 x 0.6W = 30W which at 12V is 2.5A (or closer to 3A using my 10.5V setting on the power supply) and then the injection line would be responsible for providing power to 100 pixels (100-50 & 101-150). That injection line should be carrying 100 x 0.6W = 60W which is 5A at 12V (or closer to 6A using my 10.5V setting on the power supply).

    So, does this all make sense? Like I said, I will try to get some good measurements here today and post an update. I am curious what others thoughts are on the subject. I don't know that I've ever seen a post covering this topic before... feel free to link one here if you know of one.

    Thanks for looking.

    -Louie
     
  2. lithgowlights

    lithgowlights Senior Elf

    Joined:
    May 6, 2010
    Messages:
    742
    Likes Received:
    16
    Location:
    Lithgow, NSW
    Find Me On:
    The specs say 0.3w as that's what the LED nominally draws, but the catch with 12V pixels is they draw close to 0.02A at 12V per color (usually around 0.015A in reality) so that's often closer to 0.55w drawn from the supply (the difference is dissipated in the controlling chip).

    I'll leave it to the other experts to tell you how they fuse and inject power
     
  3. randallr

    randallr New Elf

    Joined:
    Sep 3, 2012
    Messages:
    31
    Likes Received:
    0
    Location:
    Andrews, Texas
    I had a P2, with each output driving 1/2 of 24x50 pixel tree. All of the 100 ct pixel strings went up and down, and I had power injected at the first pixel of every string (which is also the end of the last one) Four 350w power supplies each injected at 3 points. I had one power supply fail. (noticed strange behavior on several strings). When under load one PS was dropping way low.


    It's now my understanding that, to add to the confusion, the power supply with the highest voltage will carry the load. Theoretically, if one PS was at 12.2v, and the others were at 12.0v, the 12.2v one would be at maximum load, and thus would be more likely to fail.


    In reality, voltage drop along the pixel strings may negate this issue.
     
  4. AAH

    AAH I love blinky lights :) Community Project Designer

    Joined:
    Dec 27, 2010
    Messages:
    2,507
    Likes Received:
    1
    Location:
    Eaglehawk
    Find Me On:
    I'll help to clear a few things up. A fair portion of the pixel chips are constant current devices so that as long as they get sufficient voltage they will all draw the same current for the same "colour". 18.5mA is what the spec is for the 2811 chipset and 20mA is 2801 I think. This means that with 12V powering the pixel and a coloured led on to 100% it will be drawing 18.5mA or 0.222W (Power equals Volts times Current). Each of the 3 colours will draw 18.5mA when on white so the individual LED containing a red, green and blue will be drawing 0.666W from the power supply. I reality it is likely to be between 0.5W and 0.6W depending on how much power injection is done along a given string.

     
  5. fasteddy

    fasteddy I have C.L.A.P Global Moderator Generous Elf

    Joined:
    Apr 26, 2010
    Messages:
    6,471
    Likes Received:
    16
    Location:
    Albion Park NSW
    Find Me On:
    These threads may also help in providing further info on voltage and current

    http://auschristmaslighting.com/forums/index.php/topic,3668.0.html

    http://auschristmaslighting.com/forums/index.php/topic,2595.0.html


    You will find that rated and real measured currents are different due to variations in materials and certain design specs and nearly always the rated current is higher than the actual current

    Now with power injection the reason why we say to not connect the +V together between different power supplies is due to the variations in powers supplies and you will find one power supply taking on more load than the other.
    When injecting power using the same power supply then the current will flow down the least point of resistance, so depending on the circuit reistance (cable size) then this will determine the level of current flow through each cable
    Now to add to this, the cable itself is a load as it itself has a resistance so you could have 2 seperate power supplies both connected together through a cable and the cable will act as the load due to the reistance, but for this to work you need a decent amount of resistance and the reason why we say to keep power supplies seperatly connected.
     
  6. OP
    OP
    nutz4lights

    nutz4lights Full Time Elf

    Joined:
    Dec 12, 2012
    Messages:
    305
    Likes Received:
    0
    Location:
    Melbourne, Florida
    Thanks for the replies guys. I'm very comfortable with the voltage and current aspects of the pixels, just not the injection. I don't know if I've seen any numbers posted for this type of setup before, so I went and measured them tonight.

    Single 12V power supply / 350W, but with the voltage lowered to 10.50V to hopefully prevent the Technicolor string LDO's from going south. This power supply was hooked up to a p12s with 12AWG wire. I took three strings of Technicolor 50 pixel strings with the stock connectors on the strings and went to work. The p12s is connected to the first Technicolor string with a 1.5m length of Ray Wu 3-core cable/connector. This will be a white light test with all RGB run full on.

    The following "tee" adapters from Ray Wu were used in this as well since I plan on using them in the actual display:
    http://www.aliexpress.com/item/3-core-T-type-waterproof-splitter-BLACK-color-the-male-connect-s-diameter-13-5mm/1019809897.html

    SINGLE STRING: multimeter was place in-line between the p12s output and pixel #1 which showed a measured current of 2.70A, which is 54mA per pixel & 0.57W.

    TWO STRINGS: two strings were connected to each other and the "tee" splitter was connected at pixel #100 with power injection going from the power supply to the "tee" adapter with a 1.5m 3-core Ray Wu cable. Current was measured both between the p12s and pixel #1 & also between the power supply and pixel #100. p12s > pixel #1 current had decreased to 2.37A and the power supply > pixel #100 current was 2.67A. Those correspond to 47mA and 53mA per pixel, or 0.49W and 0.56W.

    THREE STRINGS: a third string was connected to the above, as expected, using the "tee" adapter as it was intended. The "tee" adapter allows the data to continue on past pixel #100 to provide data to pixel #101-150 while allowing for power injection as in the "two string" scenario above. Current was measured both between the p12s and pixel #1 & also between the power supply and pixel #100. p12s > pixel #1 current increased slightly to 2.46A and the power supply > pixel #100 current was 5.07A. Those correspond to 49mA and 51mA per pixel, or 0.51W and 0.53W.

    So, I have my answer which is that, when using the same power supply for power injection as was used to power the string through the p12s, the current flow remains fairly isolated from each other on the p12s line to the pixel string and the power injection line to the pixel string power injection point. Yes the current on the line from the p12s to pixel #1 decreased 10%-ish, which doesn't seem unreasonable, even when there were 100 pixels placed after that first string.

    Hope this information is useful to somebody else out there.

    -Louie
     
  7. fasteddy

    fasteddy I have C.L.A.P Global Moderator Generous Elf

    Joined:
    Apr 26, 2010
    Messages:
    6,471
    Likes Received:
    16
    Location:
    Albion Park NSW
    Find Me On:
    You will find less current will flow down the string that has the longer supply cable due to the additional resistance of that cable and you will find additional current on the shorter cable due to less resistance.
     
  8. multicast

    multicast Senior Elf Generous Elf

    Joined:
    Jul 13, 2013
    Messages:
    711
    Likes Received:
    5
    Kirtoffs current Laws can help you solve the problem on paper.
     
  9. AAH

    AAH I love blinky lights :) Community Project Designer

    Joined:
    Dec 27, 2010
    Messages:
    2,507
    Likes Received:
    1
    Location:
    Eaglehawk
    Find Me On:
    Murphys Law overrules all other Laws :p
     
  10. fasteddy

    fasteddy I have C.L.A.P Global Moderator Generous Elf

    Joined:
    Apr 26, 2010
    Messages:
    6,471
    Likes Received:
    16
    Location:
    Albion Park NSW
    Find Me On:
  11. OP
    OP
    nutz4lights

    nutz4lights Full Time Elf

    Joined:
    Dec 12, 2012
    Messages:
    305
    Likes Received:
    0
    Location:
    Melbourne, Florida
    Ya I think I got a C grade when I took AC Circuit Theory in undergrad... good thing I was a chemistry major and it wasn't that important to learn that stuff! :p
     

Share This Page