Results 1 to 8 of 8

Thread: Formula for reduction in power with reduction in brightness

  1. #1
    Join Date
    May 2016
    Posts
    254
    Post Thanks / Like

    Default Formula for reduction in power with reduction in brightness

    I'm attempting to clean up my power distribution for next year. I'm revamping the megatree. It's 32x95 360 tree. At 100% brightness on full white, the tree pulls a little over 1700 watts. Originally, I injected power between strands 2/3, 4/5, 6/7, etc.. This year I ran it at 10% and it was still plenty bright, but I kept the same injection scheme. After one fuse on a distro board kept blowing (I think I was using one of my power cables that had a short--the cable was warm), I simply pulled the cable and found that the section of the tree still had plenty of power. That got me thinking about how much power I can eliminate.

    Is there formula for reduction in power needs vs. brightness, or is it assumed that the reduction is power is roughly linear, i.e. a 50% decrease in brightness results in a similar reduction in power needs (ignoring loss due to resistance, for the purposes of discussion)?

  2. #2
    Join Date
    Dec 2011
    Location
    UK S80 postcode
    Posts
    1,404
    Post Thanks / Like

    Default Re: Formula for reduction in power with reduction in brightness

    Hi, what an interesting question. I’m sorry that I’m not going to be able to give you a definitive answer but hopefully I can explain why.

    The brightness of the pixel is determined by the length of time it is switched on for, whenever switched on it runs at 100% brightness. For a WS2811/12 pixel it draws 18.5mA per colour or 55.5mA when displaying full white.

    If you set the pixel to 50% brightness it draws 55.5mA for 50% of the duration and 0mA for the other 50% of the duration. So yes the average power draw is linear but the instantaneous current draw can be different.

    I’ve never checked the ‘on’ ‘off’ timing between pixels but I find it hard to imagen that they are all permanently synchronized i.e. all ‘on’ at the same time, in fact because of the way that the data is transmitted I would hazard a guess the they are deliberately out of synchronization to reduce instantaneous demands on the power supplies, this is supported by your findings i.e. you can run much longer strings than expected. This again suggests that the current draw should be fairly linear.

    A thought: If the data deliberately desynchronizes the outputs to minimise load, any interruption of the data then relies on the internal clocks of the pixels to maintain the desynchronization. If there was a long data outage with pixel clock variations the load could become unpredictable and in excess of expectations, the perfect storm scenario could be 100% pixel load for 10% duration and 0% pixel load for 90% duration.

  3. #3
    Join Date
    Jan 2014
    Posts
    82
    Post Thanks / Like

    Default Re: Formula for reduction in power with reduction in brightness

    Last edited by dennismc; 01-04-2019 at 06:25 PM.

  4. #4
    Join Date
    May 2016
    Posts
    254
    Post Thanks / Like

    Default Re: Formula for reduction in power with reduction in brightness

    Well, that makes things a whole lot more interesting. If data is time-shifted, then would it indicate that the pixel is taking advantage of our persistence of vision (I mean more than we already build into sequences when we select a frame rate)? Theoretically, I suppose that it would be possible to time-shift enough pixels so that only a certain percentage of the pixels are pulling a full current at any moment, so the instantaneous current draw at full brightness might actually be a fraction of 100x55ma (in a 100-pixel strand), but because we can only see so many FPS, the switching is imperceptible. Sort of like timing of the data stream is analogous to the signal/scan frequency to an old CRT.

    If power is on 50% then off 50% for a given pixel, then the trick is determining the number of pixels that are on at any given moment to calculate the instantaneous current draw across the prop, correct? That would give you an accurate formula.

    I think the issue from a viewing standpoint is whether there is enough current such that the human eye perceives that pixels are lighted when they should be lit (and at the proper color). So long as that condition is satisfied by use of the average draw figure you've cited, then I think that's what's important to me. I'll mess with the tree when I'm taking it down tomorrow and see how many pixels I can push at various brightnesses. That'll get a few data points.
    Last edited by Kensington; 01-04-2019 at 06:49 PM.

  5. #5
    Join Date
    Jan 2008
    Location
    Kennesaw, GA
    Posts
    926
    Post Thanks / Like

    Default Re: Formula for reduction in power with reduction in brightness

    I would not waste too many brain cells thinking about this. To me it appears to be fairly linear. I can easily run 240 12v pixels at 10% and the display looks terrific. A few years ago I built a little meter and put a male and female pigtail on it so I could insert it in front of any string in my show. So when I say it is fairly linear I have actually watched on my meter while I changed the brightness. However I do not size my power supplies for 10% brightness and you just know that something will get forgotten at some point and pull way to much power.
    Attached Images Attached Images
    Jim Nealand
    Kennesaw, Ga

  6. #6
    Join Date
    May 2016
    Posts
    254
    Post Thanks / Like

    Default Re: Formula for reduction in power with reduction in brightness

    Interesting results from tonight’s test. I could only push about 150 pixels directly off the F16v3 at 10%, but adding power at the end of 10 strands and between pixels 190 and 191 and at one other point powered 10x95 nicely. So I can cut my cable overhead by nearly half.

  7. #7
    Join Date
    Dec 2018
    Location
    Winnipeg, MB
    Posts
    25
    Post Thanks / Like

    Default Re: Formula for reduction in power with reduction in brightness

    Quote Originally Posted by Kensington View Post
    Interesting results from tonight’s test. I could only push about 150 pixels directly off the F16v3 at 10%, but adding power at the end of 10 strands and between pixels 190 and 191 and at one other point powered 10x95 nicely. So I can cut my cable overhead by nearly half.
    what AWG are your pixel wires, and what is your AWG and length of cabling between PS and first pixel?

  8. #8
    Join Date
    May 2016
    Posts
    254
    Post Thanks / Like

    Default Re: Formula for reduction in power with reduction in brightness

    18g cable, within 4 feet of the power supply. I know I’m losing some to resistance, but I’m intentionally ignoring Ohms law for the moment. And forward voltage drop, for that matter.

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
  •