@joegrand tweeted, “Practicing BGA reballing with the @dangerousproto BGA Reballing Practice Kit.”
More resources and instructions available here.
Get your own kit for $79 at Seeed.
It’s well-known that buying Flash storage devices from cheap online retailers is fraught with danger. Stories abound of multi-gigabyte drives that turn out to be multi-megabyte ones engineered to falsely report their capacity. So when [Jason Gin] found a source of 64GB Toshiba eMMC chips for only $6 per device he bought a few, but was prepared for disappointment.
To test them, he decided to use an SD card interface. There are minor differences between eMMC and SD, but the interfaces are electrically the same and in most cases an SD controller will happily do business with an eMMC. It was not however an easy task to connect the two — these eMMCs were in BGA packages, hardly the easiest ones to work with. He resorted to dead-bug soldering the relevant interface wires to SD lines, and connecting up his computer.
His first attempt was something of a failure, wiring the chip to the PCB of a cheap USB-to-SD adaptor. This did not put him off though, he followed it up by cracking open a very old 2GB SD card that contained a PCB instead of being potted, and soldering his eMMC in place of its Flash and controller. This even fit in the original SD housing, and met with success when plugged into more reliable SD card readers. He was thus able to confirm the capacity of his chips.
His blog post is worth a read for more than just the very fine soldering involved. He takes us through some of the intricacies of SD interfacing, as well as talking at length about the decoupling and termination required to make a reliable connection. We particularly like his use of an area of unconnected BGA balls as prototyping space for decouplers.
If you marvel at the exceptional dexterity required for hand BGA work, we’ve a couple of other treats for you. There is this TI infra-red sensor BGA soldered to a piece of stripboard, and this wafer-level chip package soldered to an SOIC prototyping board.
Making stuff is hard, especially when you are making lots of stuff. The OpenMV Cam project knows this, because it has hit a problem while putting together their cheap machine vision module. The problem is with the BGA solder balls that connect the image sensor to the main board.
We’ve covered this intriguing project before: the aim is to build a small, cheap module that can run image processing algorithms to easily give robots sight. The sensor is a Ball Grid Array (BGA) package, which means there are a grid of small solder balls on the back that form the electrical connections. It seems that some of these solder balls are oxidized, preventing them from melting and fusing properly with the board. This is called a head-in-pillow defect, because the ball behaves like your head when you lie down in bed. Your head squishes the pillow, but doesn’t merge into it. There are 38 balls on the OV26040 image sensor and even a single bad link means a failure.
The makers of the project have tried a number of solutions, but it seems that they may have to remake the ball links on the back of each sensor. That’s an expensive process: they say it will cost $7 for each, more than the actual sensor cost initially.
A few people have been posting suggestions in the comments for the project, including using solvents and changing the way the sensors are processed before mounting. We’d like to see them overcome this hurdle. Anybody have any suggestions to quickly and cost effectively move the manufacturing process forward?
Thanks to [eDgE] for the tip.