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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

soldering issues



soldering issues | 6 May, 2002

Hello all, I am having soldering issues with an Everlight component it is a TM4201/TR2 (Right angle IRDA with a 1mm pitch). The problem I have is that after reflow the solder joints look perfect, but during test we are losing 7 micro amps as soon as I hand solder the components they pass test. One issue I was concerned with is that there is a dpak cap on the other side of the board directly under the IRDA and the board is only .060 thick. The dpak gets soldered first then the board is flipped and the IRDA is soldered on the other side. I was wondering if the dpak could be acting as a heat sink and stealing some of the heat from the IRDA thus causing a small fracture of some sort that is not visible. I have already run a profiler thriugh the oven to check if the heat mathched with the component specs and it does, at the back of the component but I could not get a probe under the component itself.

I know this is a long shot but we have exhausted all other routes to this problem besides looking at the design of the product. any help would be greatly appreciated hopfully their is something we have overlooked. Also we are using a convection relow oven without nitrogen and water soluable paste. We have tried to rewash the boards without reflowing the components and that does not help.

Thank you in advance Jacob

This message was posted via the Electronics Forum @

reply »



soldering issues | 6 May, 2002

Would the lead finish on this part happen to be Paladium? I had a problem similar to this some time ago and we found the lead finish was not tin/lead over copper but paladium over nickel. We had to increase the reflow to 225 deg. C peak temp with the TAL at about 70 sec. to "burn through" the lead finish.

All I could think of


reply »


soldering issues | 9 May, 2002

Russ makes a good point. While your profile probably meets your paste supplier recommendations, it may not be proper for the component. * Paste suppliers offer general recommendations for tin / lead interconnects. You need to modify that for board design, component finish, etc. * Check with your component supplier for recommendations.

Probably Russ is correct about palladium. That you get a good solder connection when hand soldering says that his line of thought is spot on. * TI was the first to use Pd on a large scale. Check their site for suggestions on soldering Pd leads. * If you peak at ~210�C, you need to either slow-down your conveyor or increase your peak. * Doug Romm at TI says �Generally, there is no "clear-cut" way to identify Pd versus Sn/Pb. Pd does "look" different than Sn/Pb, but you have to look at a large number of units to be able to recognize the difference.� * Pd dissolution rate into solder uin/sec 215�C|| 0.7 250�C|| 2.8 These dissolution rates give an indication of the time necessary to dissolve the palladium into the solder before a solder connection can be made to the nickel. * Search the fine SMTnet Archives for more on Pd.

Taking a slightly different tact than [slitting hairs with] Russ� �225 deg. C peak temp with the TAL at about 70 sec.�, we like to see at least 5 sec at liquidous plus 25�C. The liquidous that we�d be talking about, in this case, is Pd in to Sn/Pb, not Sn/Pd in Sn/Pd.

reply »

Fluid Dispensers

Metcal soldering rework