WTF BGAs are news?
Open NAND Flash Interface Specification from Feb 2008 specifies the pin out for FLASH in BGA ((solder) Ball Grid Array) packages.
Toshiba has produced a tiny but bigger SSD for tablets and ultrathins by adding TLC 3D NAND to its BG1 SSD card product line, and doubling its capacity. The BG1 was introduced in August 2015 as a 256GB NVMe M.2 product. This product had an edge connector and was inserted into a socket. The actual chip was affixed to the M.2 …
With all the discussions about BGA chips, its worth mentioning that many problems were caused by the switch over to lead free,
the workaround is obviously to use 63/37 and some manufacturers now state that the product is "lead reduced" as proportionally it is less than 1% of homogenous volume.
Just for the high reliability parts and everything else is typically SnAgCu or some other unholy mess.
I did come up with an idea to use a high strength neodymium wafer magnet to "push" the In bumped chips onto the base so it should be just as reliable, the idea is that if it does ever break a very simple reflow cycle slightly higher than normal (internal heater?!) will unseat chip without breaking it or the magnet, due to differential effects.
Recall basic physics which shows upper and lower Curie points and also the magnet makes a convenient heatsink plate.
Anyone care to comment on the feasibility of this?