Printed Circuit Board Assembly & PCB Design SMT Electronics Assembly Manufacturing Forum

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

qfp and soic joint Problem

Jeremy Smith


qfp and soic joint Problem | 11 April, 2000

I am having a problem with some joints on a 100 pin pqfp and few different soics on one particular board. When viewed a 8x with a "Mantis" inspection scope the joints look as though they wetted to the ic leg but when probed the leg will pop loose. When loose the leg still has solder attached to leg as though the pad came up with it but the pad is still on the board. I have seen this problem before on a board that had bad plating and when the legs were picked all of the solder came off leaving a bare copper pad but these pads still have solder on them. We place thousands of similar IC on many different boards but have not had this problem before. The only thing that I can think of that would be the cause is some kind of contamination of the plating but I have no way of proving that. Has any one else had this problem? Any comments or suggestions would be appreciated. I am using Amtech Fresh Mix 559 Noclean solder cream and QUAD ovens.

reply »

Glenn Robertson


Re: qfp and soic joint Problem | 11 April, 2000

Jeremy -

What's the process flow on these modules? Are they wave soldered? What is the board finish (ENG, HASL, etc.)?

Glenn Robertson

reply »



Re: qfp and soic joint Problem | 11 April, 2000

I�m thinking �Poor intermatallic� It could be a few things. Lets start at the beginning. -Is your paste a 63/37 mixture? -Have you monitored you screen printing, Verify Solder deposition, Are you getting enough paste? -You say the solder looks like its wetting to the lead, but is it wetting to the pad? take a bare board screen it and reflow.Inspect the unpopulated pads, verify there is good wetting. - Perform a pull test, Solder some wires to the pads and attach a force guage see how much force is needed to pull the wire loose.See where the joint is broken.

- How many heat cycles does the components see? - The more its heated the more the intermatallic is weakened. -Something that is very difficult to prove or determine is In Melt Stress. This is stress created at the intermatallic caused from the TCE of the Board and the component during cooling.

Just some thoughts!

reply »


Re: qfp and soic joint Problem | 11 April, 2000

Jeremy: Parts assembler make leads of either Alloy 42 (Invar, whatever) or copper. The solderability of these leads are protected with eutectic or non-eutectic tin/lead solder, palladium or some palladium alloy, or gold. Cuppla things to consider:

1. Sometimes, Alloy 42 leads are tough to solder, regardless of the solderability protection. Not all Alloy 42 is the same; while all contain about 42% Ni, the rest is Fe. Ni is solderable, well sort of solderable, but it needs much more thermal energy to go into solution with Sn than does Cu � and Fe is not solderable at all. Remembering back to baby chemistry: Ni, Fe, and Co are magnetic.

Even with perfectly solderable Alloy 42 TSOP components, you will have solder joint reliability problems depending on the application's use environment. Here are a couple of papers that you may find useful: * "Solder Joint Reliability of a Thin Small Outline Package (TSOP)", J. Lau, R. Fornringer, et al, 42nd ECTC conference proceedings, May, 1992. * "Reliability Evaluation of TSOP Solder Joints for PC Card Application", R. Iannuzzelli, et al, Soldering & Surface Mount Technology, No. 23, June 1996.

2. Palladium dissolves into solder more slowly than other coatings Sn/Pb. Check the archives. 3. Gold dissolves into solder more slowly than other coatings Sn/Pb. Check the archives.

Ta Dave F

reply »



Re: qfp and soic joint Problem | 17 April, 2000

All of the above is good stuff.

To add, I have seen this several times if the assembly is wavesoldered. Heat from the wave can partially reflow the top side SM solder joints and weaken them. To test, check joint strength before wave and then after.

Reasons for partial topside reflow; excess preheat, too slow conveyor speed, and poor fab design. Regarding fab design if the designer puts too many vias too near the pads of the the QFP then heat from the wave will move thru the fab and do its thing.

Fixes- Fix the design (can't, already bought all the fabs), shoot the designer (self gratification), mask or shield the bottom side of the fab (near the QFP's) from the wave.

Enjoy, Boca

reply »

ii-feed SMT Intelligent Feeder

SP700avi inline smt screen printer