Marc's Public Blog - Electronics Hacking


All | Aquariums | Arduino | Btrfs | Cars | Cats | Clubbing | Computers | Diving | Dreamstate | Edc | Electronics | Exercising | Festivals | Flying | Halloween | Hbot | Hiking | Linux | Linuxha | Monuments | Museums | Oshkosh | Outings | Public | Rc | Sciencemuseums | Solar | Tfsf | Trips



>>> Back to post index <<<

2024/11/20 Festival Dreamstate Hat Built with RGBPanel, Neopixels, and Lipos
π 2024-11-20 01:01 in Clubbing, Electronics, Festivals
Tech specs:
  • 32x16 flexible LED rgbpanel (displaying scrolling Dreamstate Logo)
  • 2 strings of Ray Wu P15 WS2812 both for layout reasons and for backup if one string breaks, the other one will keep working.
  • Lipo battery checker (but the lipos are protected, so they will self shut down)
  • flexible 16x32 RGBpanel
    flexible 16x32 RGBpanel

    found a hat big enough that it mostly fits on it
    found a hat big enough that it mostly fits on it

    of course, it needed a few LEDs :)
    of course, it needed a few LEDs :)

    what it looked inside and it worked with a small 1S 700mAh lipo (although not long enough)
    what it looked inside and it worked with a small 1S 700mAh lipo (although not long enough)

    bigger battery upgrade, and BTW those batteries lie, they are only 1600mAh
    bigger battery upgrade, and BTW those batteries lie, they are only 1600mAh

    testing battery draw, a bit over 0.5A which is a bit too much
    testing battery draw, a bit over 0.5A which is a bit too much

    So, why 3-4V lipos instead of USB battery packs with 5V that is expected by both the panels and neopixels/WS2812? Well, space is limited, USB battery packs are kind of big, and carry extra hardware to recharge that isn't needed inside the hat. There is also a side byproduct that powering both the panels and the LEDs with a lower voltage, limits their brightness and the amount of power (watts) they need, which means that a 16Wh lipo inside a USB battery pack stepped up to 5V, makes everyting brighter, but only lasts about half as long as powering directly from the lipo at lower voltage.

    bigger 4Ah 1S lipo worked for 8H at full power and LEDs worked all the way down to 2.8V!
    bigger 4Ah 1S lipo worked for 8H at full power and LEDs worked all the way down to 2.8V!

    panel worked all the way down to 3.4A before colors went wrong, 16H on a 1.6Ah battery
    panel worked all the way down to 3.4A before colors went wrong, 16H on a 1.6Ah battery

    of course now I had to find a buy a faster 1S lipo charger with the right connectors (had to get JST adapters)
    of course now I had to find a buy a faster 1S lipo charger with the right connectors (had to get JST adapters)

    quick test that a smaller 700mAh battery would last around 5H before things got dim
    quick test that a smaller 700mAh battery would last around 5H before things got dim

    After testing, I was able to confirm that both the $7 amazon controller (SP002E from https://www.amazon.com/gp/product/B09Y8SWJ77 ) and the RGBpanels, work fine with lower voltages (which is not a given since both run a microcontroller that was originally meant to be powered from 5V).
    The RGBPanel controller is a bit more picky about voltage and reboots around 3.5V, while the neopixels tend to drag the battery voltage down, causing the RGBPanel to crash and reboot when a single battery runs both, so I gave it its own lipo.

    Pixels still take 0.3 to 0.5A (I use a potentiometer to dial them down as the cheap controller I put has no dimming control) and the RGBPanel takes less than 0.1A, so that's nice (it actually goes all the way down to 0.05A or just 50mA) when the voltage drops.
    The combined tricks should give around 18H of runtime with the 2 batteries (4Ah and 1.5Ah 1S lipos), which is enough for 17 hours of dreamstate (they are not screwing around this time, 17H !!!)

    About the neopixel controller, I used a cheap $7 amazon controller that only had 3 physical buttons but sadly no dimming control, mostly because I wanted something very small, and while pixelblaze micro is also small, it doesn't have the pattern I want, and I didn't see the point of programming a pattern I already had on the other controller, so I just put a potentiometer to lower the voltage. It's obviously the wrong way to do it, but it works :)

    Here is the end result:




    More pages: November 2024 September 2023 August 2023 June 2023 May 2023 April 2023 January 2023 November 2022 January 2022 May 2020 January 2020 December 2019 April 2019 January 2019 June 2018 May 2018 June 2017 April 2017 May 2016 February 2016 November 2015 October 2015 August 2011 July 2011 May 2011 June 2010

    >>> Back to post index <<<

    Contact Email