0

SK9822 failure rates?
Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.

SK9822 failure rates?

by davelasers on Fri May 14, 2021 11:46 am

Hello, we purchased 500 SK9822 from Adafruit recently and had 280 used in 20 assembled boards from a large board assembler. Of these 20 boards, only 5 had 100% working LEDs and we are unsure how many have failed since they are daisy chained 14 in a row, so once we hit a failure we can't check the subsequent LEDS.

The question I have is this a component (sourcing) issue or a potential assembler issue? The SK9822 seems to be one of some number of clones of these dotstars. Image included here has a different layout than the one on adafruit (you can see the RED colored die is not in the middle and the die size and pad locations are different, plus the board has slightly different pads as well).

dotstar.JPG
Adafruit image compared to die received for SK9822
dotstar.JPG (197.71 KiB) Viewed 98 times


Can we rely on this part? We need 99.99% yield on components and in the best case we are only at 95% right now.

Advice?

davelasers
 
Posts: 2
Joined: Fri May 14, 2021 11:30 am

Re: SK9822 failure rates?

by adafruit_support_mike on Fri May 14, 2021 11:15 pm

That would be a fabrication issue. We use the same parts for our own boards, and our defect rate is well below 0.1%.

The parts are sensitive to damage in the reflow oven, and you have to find the heating/cooling profile that works for them. That varies from one oven to the next, and with the number and size of the boards being processed. Part of our development process includes tuning the reflow parameters for a new board.

There are also issues with moisture sensitivity. You might need to bake them before assembly if they've been exposed to open air for too long.

Talk to your assembly house and see if they're willing to tweak the reflow profile, and if they have any suggestions/requirements for a testing process. You'll need to work closely with them, and will need to do test runs to find a suitable process.

adafruit_support_mike
 
Posts: 62799
Joined: Thu Feb 11, 2010 2:51 pm

Re: SK9822 failure rates?

by davelasers on Sat May 15, 2021 11:39 am

adafruit_support_mike wrote:That would be a fabrication issue. We use the same parts for our own boards, and our defect rate is well below 0.1%.


Thank you. I did speak with them and they confirmed they used a standard heating profile that was in excess of what was recommend for these components. They confirmed they followed the moisture sensitivity requirements. We are going to have the boards reworked with components we just re-ordered. Will confirm success at that point.

davelasers
 
Posts: 2
Joined: Fri May 14, 2021 11:30 am

Please be positive and constructive with your questions and comments.