Author Topic: Pixel controller output chips  (Read 787 times)

Offline Scottishsurfer

  • Jr. Member
  • **
  • Join Date: Mar 2016
  • Location: Rancho Santa Margarita, CA
  • Posts: 60
  • Kudos: 0
    • My light show website
Pixel controller output chips
« on: February 06, 2017, 11:32:11 PM »
So what exactly do these pixel output chips do on the boards?  I've fried a couple of them by accidentally wiring my pixels incorrectly and they have saved frying my board.  Is that why they are generally replaceable?  So when you are rushing to get a display up and running, and you accidentally mess up your wiring.  (I'm sure I'm the only one who has ever done this :) )


So here's my main question - Do these chips work/don't work?  For example, is it possible to damage a pixel output chip to the point where is works most of the time?  Or when you fry them, they are just dead?


Just wondering if I ever need to "test" my pixel output chips, or as long as they are sending power (tested via multimeter) are they generally ok?


For testing, how would you even test an output chip?


Thanks,
Colin

Offline AAH

  • Hero Member
  • *****
  • Join Date: Jul 2014
  • Location: Australia
  • Posts: 740
  • Kudos: 19
  • Blinky blinky blinky
    • I love blinky lights
Re: Pixel controller output chips
« Reply #1 on: February 06, 2017, 11:55:13 PM »
  The pixel output chips do a number of things. They typically bump the 3.3V logic that a lot of the fast processors run at up to the 5V logic that most pixels work. They provide a fairly high current output that allows fairly capacitive cable to be attached and still get the square wave signal out at the first pixel. Thirdly they provide a buffer between the expensive logic of the microprocessor or other device that generates the pixel data and the outside world. They aren't necessarily designed to allow the 5V or 12V power rails to be connected to the pixel output but between the buffer resistor and the internal protection that the chips have they will often tolerate this abuse and not allow the power to be passed back to the input. The internal protection diodes pretty much become sacrificial with this sort of abuse and the chip will be killed. The buffer resistors which are typically something like 33 ohms to 50 ohms can also be damaged but would usually become open circuit or short circuit.
  Probably about the best method of testing the output is to put a single pixel on maybe 10m of cable and running some data out to it. If the chip is dead or compromised in some way it probably won't drive the full distance of the cable. As an alternative test that doesn't check the chips ability to drive into a load the voltage between ground and the data pin can be measured as the pixels are faded up from 0 to 255. The voltage should change from something like 2V up to maybe 2.5V. The actual voltages will depend on the way in which the data is sent. If it's just 1 packet being sent out at 20mS intervals versus a continuous stream then it will vary. There will always be an increase in the measured DC voltage though.

 

Back to top