The parts I'm talking about here are the "SEM" series connectors from Samtec. The pins coming out the bottom of the connector are a rectangular post shape. I see no sort of strain relief of the pin iself. However, the connector variety we are using does have locating pins at the back of the connector just outside the pin area. This is supposed to limit the movement of the pin area once placed and soldered down. And, in the front, we have solder posts. One on each end of the connector. These are meant to tie the=20 leading edge of the connector down so that when the mating connector is engaged, you don't have torsional force trying to pry the front of the connector up and away from the board.=20 In summary, there is no obvious strain relief of the leads coming out of the connector, there are locating pins at the back of the connector to limit movement of the pin attachment area, and there are solder posts at the front that hold the connector to the board to keep torsional forces from lifting it during mating connector engagement. Samtec engineers looked at some of the failed units, and they said it looked like the proper solder profile was used, and that the layout was good. But they couldn't explain why the failures happened. They also=20 said that the temp range we are testing to is wide, but the conenctor should be able to survive that range is our unit's operating mode. Regards, Jim > -------- Original Message -------- > Subject: RE: [EE]: SMD soldering question > From: > Date: Thu, May 19, 2011 9:58 am > To: >=20 >=20 > > When these boards go through our production testing, which includes > > several cycles of hot (85C) and cold (-40C) temperatures, we find > > that the solder bond between the PCB pad and the connector pin fails, > > which in turn causes the unit under test to fail. >=20 > My first question would be what sort of strain relief is there in the lea= d. It sounds to me like it is not the solder joint at fault, more that when= the metal lead shrinks as it cools it pulls the lead off the pad. If there= is no loop in the lead to give it some expansion/contraction ability the f= ault you are seeing is exactly what I would expect. >=20 > For space qualified assembly work all leads like you mention should have = some form of Z strain relief in them. The sort of temperature range you are= testing to is the non-operating survival temp range we would use. >=20 >=20 > --=20 > Scanned by iCritical. >=20 > --=20 > http://www.piclist.com PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .