This is a little embarrassing asking this question, as I think I am profiling our oven incorrectly.
I have been using the profile based on the solder paste specification, however we have been getting a large BGA chip coming more unreliable, and breaking off the board after shipment. It's a clean break, and often there is clean hold pads under the part after it's broken off, so the solder does not look like its adhering to the pads correctly, even though the product tests ok.
We have run our profile specification window 6 degrees hotter in the reflow part of the profile, above what the solder paste specification has, and it seems to have fixed the issue.
However I assumed if I probed the parts, and they achieved the temps based on the solder paste I am using, it should result in good joints. Am I wrong here, as that does not seem the case, and there must be another reason the joints are not as good.
I even purchased two profilers to ensure we had something to compare against, and they are within 1 dec/c of each other, and we are using high temperature thermocouples. The paste is Koki which has given very good results in the past.
However for our most complex products, it seems we need to juggle the profile to get lower failure rates, independently of the paste specification.
Is this normal, am I doing this right, as if this is normal then it would be good to know what would cause a profile that's done according to the paste specification not to solder correctly. I seem to be missing some kind of knowledge here, which is freaking me out a little. One of the limits of learning all this by myself I guess. It leave gaps sometimes.
Any advice would be great.