Soldering back a new M92T36

Maybe I m too silly, but I have problems to get my M92T36 solder right in place. It s the second time on this one mainboard and have to replace the M92T36 due to testing and creating shorts, that blew the M92T36. And this is the second time I have trouble to place the M92T36 back.

Removing was no problem. I added flux and heat the ic with 420°C and 12% airflow with a Weller WHTA. The tip has a 6 mm and 45° bent outlet. I place the tip aprox. 30 mm above the ic. And it came off in maybe 30 seconds. I cleaned the flux and add new one and retined the pads. Than cleaned again the flux and add new before placing the out off the blister new M92T36. I add heat till I could see the tin is liquid and than give the ic a little nudge with my tweezers. But instead of jumping in the correct position the ic stays where it is nudged to. I tried severell times than I removed the ic completely and retined the pads and the pads on the ic and start a new attempt. This time during soldering I pushed with my tweezers from the top and squeezed some tin from under it and then gave it a tiny nudge, but also here: no self alignment.

Any ideas, what I m missing…

Best advice i can give on this subject.

I’ve mentioned this a couple of times here before but if you’ve got your “leaded” solder off ebay or amazon, then it’s likely not actually leaded, typically the sellers seem to be selling a rebranded unleaded or a terrible alloyed leaded that melts at similar temps to unleaded.

Would reccomend 63/37 chipquick leaded solder.

If i ever bother re-tinning the IC itself i typically just do the outer pads and avoid the inner ground/sink.

i’ll add the absolute bare minimum flux during the initial heat up process and when i feel i’ve got everything up to reflow temps (which you can verify by adding leaded solder to a surrounding component, for example a capacitors end-caps) then i’ll add more flux during, this ensures the chip doesn’t run away at the beginning, too much and it will float off too little and it won’t pull itself into place.

If too much solder is placed on the center ground/sink it will prevent the outer pads on the IC from making contact on the board pads in turn preventing self alignment, general rule here is less height in the middle compared to the outer pads. As the center is higher mass it will naturally pull it down to whatever lower level of solder applied at which point the outer pad rows will do the work of alignment, so basically just use less solder in the middle :smiley:

I know other people techniques are to bring to reflow temp and push down and clean up after with an iron… i don’t like doing it this way and don’t feel it’s neccessary.

1 Like

Thanks for the advice. I will try it tomorrow.
The tin I use is from STANNOL and marked with S-Sn60Pb39Cu1.

No worries,

Interesting composition, not far off the old 60/40 alloy.

The melting temperatur is discribted from 183°C-190°C. So it might be slightly more than Sn63Pb37. In Germany I can t get Sn63Pb37 no more from my usual supplieres. And from AliExpress I m not sure what I will get.

Personally, from aliexpress the advertised 63/37 behaved more like unleaded.

i got the chipquick stuff from mouser… TME might be an option otherwise

tinning the inner (ground) pad helps a lot to keep the chip on its place, so it doesn’t move everywhere when heating…

I’m a novice here but I can never get it to jump into place either.

I actually don’t even use flux at the initial stage on putting back because it moves around too much. I just plant it where it needs to be n double check with a scope. Heat it up with chip on its location, when the tin melt I stop the heat. Flux it up and run the soldering iron around all the contacts. Clean flux. And new flux and heat up the whole chip until tiny balls around each contact forms up then stop the heat.

This is prob not the right way but it works for me. I tried many following all the videos n tutorial but always end up going back to measure n touch up the joints over n over