MBI5026GN Question
  • Hello, 

    I was wondering if someone on the forum might have experience in designing and building with the MBI5026GN chips sold in the shop.  I have run into an issue where I seem to be burning out individual output channels on a few of my chips.  What I mean by this is that the affected output channel appears to suffer from some current leakage that allows the connected LED to dimly illuminate.  It's definitely not stuck fully "on", as the LED does not illuminate with full brightness until it is latched to ON.  

    As background:

    I am working on a final project for my ECET associate degree program in which I have used the 5026 chips as the basis of a control board for a 8x8 RGB LED matrix.  The board uses 12 of the chips, 4 for each color bank.  It's been a long semester of ups and downs, but I'm at the point where I've built my matrix which works properly, built the control board (old school wire wrapping is fun!), and have software written to run the board from an Arduino Uno.  On the software side, I have to give much credit to some 8x8x8 cube designers whose control program served as the jumping off point for my code.. I'm much more of a hardware person, so having a template to work off of helped. 

    I'm about to order some more of the chips from the ESM store, but want to try to ensure I don't continue to "blow" the chips out.  The chips I have now have been with me through the design and testing process, and some were definitely in circuits that weren't properly designed at earlier phases. Last night, once my project was fully working, I swapped in all known good chips, hoping that maybe I wouldn't lose any more now that I seem to have a stable design, but after an hour or so I lost another output to the leakage problem.  

    I'm running the board and LEDs at 5V, and using an R-ext of 860Ω, which yields an output current to the LEDs on the red channels of 20.1ma.  The blue and green banks are running at slightly lower currents due to the higher LED voltage drops. Given the information on the data sheet, I don't seem to be exceeding any of the specifications currently.

    At any rate, thanks for reading my wordy post. I know it's not possible to troubleshoot based on a post like this, but I was hoping that maybe a more experienced user may have encountered this failure condition before and have some advice.  If not, it's not a huge deal, but I would prefer to be able to hang this on the wall after my presentation and enjoy years of pretty colors given the amount of time and energy I've put into the design and assembly of the project. 

    Thanks for reading!

    Mike

  • I don't think that I've seen this particular symptom, but it is possible to damage the ICs through a few different ways. The most common way is overvoltage on the power supply pin. Make sure that your 5 V power supply really is 5 V (not an unregulated "5 V" power supply).  Also make sure that you have adequate capacitors on the 5 V supply, perhaps near the ICs, so that suddenly switching off current to the LEDs does not cause a spike in the 5 V power.

    Second, we have seen a symptom that _looks_ like the one that you're describing. If you have an LED line turned "on" in the chip but are not providing current that can flow through it, it can exhibit "ghosting" for a moment after that. If you have any high-side control, it's important to only control the LEDs to be on when there is a path for the current to flow.
  • Thanks for your reply.  I'm pretty confident of the power supply I've been using.  It's one of our lab bench HP power supplies, and I periodically verify the voltage needle on the supply with a DMM.  I'm running one 0.1uF capactor per IC, placed directly between the Vin pin of each IC and ground, but I haven't acutally monitored the behavior of the voltage on a scope while the board was in operation.  I think I'll put that on my to do list for my next visit to the lab. 

    There have definitely been a few times where my connection for the 5V that runs from the control board to the matrix has come loose while the ICs have been running. One of my goals this weekend is to switch the connector out so that it has a 'catch" to keep it from coming loose. So it's possible some damage was done at the various times I've lost that connection in the past. 

    There were also some shorts and missed connections on the control board when I first got it all wired, that took a few days to track down, so that might not have helped matters. 

    At any rate, I do appreciate your response. If I figure out the cause, I'll come back and update the post. 

    I really enjoy looking through your site and checking out your kits and projects.  Wish I was local so I could stop by the store. 

    Thanks again!

    Mike