As I don’t have any more ideas I’d like to ask around here.
Got a broken Switch “does not turn on”.
Identified and changed a shorted M92. The Switch started booting again up to the switch Logo.
I reflowed and somehow f*cked the wifi chip bga array (shorted after reflow because I knocked it to hard). So I removed, cleaned and resoldered it and the short was gone. But the unit still stops at Switch Logo.
I finally replaced the wifi chip with a harvested one but no luck (still stuck).
Someone here suggested that missing Caps left to the CPU shield could cause this issue - and indeed one little cap was missing there - replaced again with a harvested one. But still stuck.
As a last resort I changed the BQ. Shortly after that the MAX89??? On the back “blew” and brought the nearby coil to boiling-temperature (I posted on that in another thread). I replaced both and am now where I started “stuck at the switch Logo”.
Hopefully any of you has an idea how I could proceed?
Honestly as I’m just refining my toolset I used a more practical approach like in this video:
https[:]//youtu[.]be/QAv1UKDi_eg (please remove square brackets - still can’t post links :-/)
Summary:
He applies leaded solder by hand with an iron on each pad. Then he smoothes it down mechanically and reheats the blobs with hot air so they “ball up” again. This can be done over and over until all balls have a uniform shape.
This was for the first (original) IC.
The second harvested one I soldered on directly after removal from the donor board without further actions.
Do you think that gently pushing it down while soldering could increase the connection possibility?
Then that’s your problem bud, you have to reball properly - you can’t really ever get away this “manual” approach or by just directly soldering a pulled IC and relying on the original solder.
I would never reccommend the manual approach and particularly on IC’s such as the wifi IC or similar as the coating on the uderside is so thin and ICs such as this are not rated for so many rework operations
Okay, now it’s done. Pulled a pad on the IC upon cleaning.
On gbatemp someone mentioned that the switch would boot fine with the BCM removed. (https[:]//gbatemp[.]net/threads/replace-ic-wifi-bcm4356xkubg-with-the-new-one.544895/#post-9194624 - please remove square brackets)
So I tried that and still no luck. I’m running out of ideas here. Don’t know how much more effort to put into this :-/
you don’t need to wick the IC when using stencil and paste, just tin it prior, just make sure to set your iron to around 250 to 280C - at about 400C or greater you’ll strip the coating on the underside of the IC and pads will merge, then it’s dead.
As I mentioned you cannot use the “manual” methods shown in that video, the ball size is too small and there is too many, you may be able to get away with the “manual” methods on smaller ICs such as the Fuel gauge or the Max IC you mentioned earlier but you put the SoC at risk of instant death if you get “high” voltage on low voltage LL I/O
I have not tried this but I have my doubts that it would boot without the IC, and even if it does it could be completely dependent on OS version
“How?”
Probably to hot iron and yes - used wick :-(.
It was the harvested one that still had unleaded solder with it. Wanted to do it “the right way”. Ripped a pad on one corner of the ic.
I don’t have another harvested one. Is it confirmed that new ones won’t work due to different firmware?
Thanks for your help so far. I will keep your suggestions in mind for further repair attempts.