Problems seen with some of the pad detector assemblies. We suspect overheating occurres ar Pace.

(1)

A surface mount cap (yellow) is soldered to a bridge piece (not shown). A kapton tab (brown) with 5 copper/gold strips (white) is glued to the to of the cap. Wires (blue) are soldered from the tab to the cap ends. Wire bonds (black) are bonded to the top half of the strips, and covered with encapsulant (dotted line).
(2)

This is an assembly which appears not to be overheated. Note the appearance of the hand-soldered connections.
(3)

This assembly has discolored PC board (light brown), but the solder has not slumped. The wire bonds are OK.
(4)

This is an assembly that has discolored (dark brown and black) PC board on the back. Note the solder has re-melted an slumped down. Also, one of the wire bonds (under the encapsulant) between the tab and the MCM is broken. This could have happened as the cap is no longer firmly attached to the base when the solder is in a liquid state. Movement of the cap floating in solder can break the wirebonds.


Using the little oven, I did a few small tests with small (1x5cm) samples of the pc board, on which I put a small amount of wadded-up solder and flux paste. The solder is from the same roll we use for the soldering, and the flux paste is also the same.

Sample 1 was in the oven for 15 minutes. At the start the temperature was 360 °F (xxx °C), and at the end of the period it was 330 °F (xxx °C). The pcboard had turned a slightly deeper yellow, but I would rate this as 'no change'. Solder not melted

Sample 2 was in the oven for 20 minutes. At the start the temperature was 380 °F (193 °C), and at the end of the period it was 370 °F (xxx °C). The pcboard had turned golden brown, and the solder had melted into a ball.

Last update 22 Nov 2000 - HvH
back