Both you and a previous poster are correct. Grainy / shiney solder are not a good indicator of a proper connection.
Your quality guy is asking the correct question, though. => "OK. If the connection doesn't look like I expect it to look, how do I know that the metallurgy is correct?"
Comments are: * There are no pictures that will get you though this. * You should expect that gold that is dissolved in Pb/Sn solder to make that look solder grainy / dull, whatever. * Gold embrittlement can occur in connections like this. [You didn't tell us, but we assume the solderability protection on your board is ENIG.] You need to be concerned when the gold is over 3%. [Search the fine SMTnet Archives for background.] * We've heard goofy stuff like drying out the flux before. Listen, if you are getting a smooth solder flow, good fillets, intermetallics at the nickel tin boundry, and all that; your recipe is fine. [You created those solder balls during preheat. It was either too hot or too cool, before going to soak.] * Gold increases the liquidous temperature of Pb/Sn. So you need to sit above liquidous plus 20�C for maybe 5 seconds longer than you would with a HASL board to be sure the gold has dissolved in the solder.
OK, back to the quality guy. The only way you can tell what's really going on in this solder connection is metallurgical analysis. There are good labs listed in the fine SMTnet Archives.
On final note: Get used to this discussion with your quality guy. As bare board and component solderability protection change, in response to the Eurolanders no-lead foolishness, there will be more and more 'exotic' alloys on your board.
reply »