component aherence force on the solder paste before reflow

Well, I�ve never encountered any problems with it. If you use paste in the specified window according time and environment there should be no problem. If you try to measure this force you would IMO check out if you yourself did not treat the paste as it should be treated. Consider that placed and not yet soldered assemblies are not for "shake, rattle and roll". If you have some kind of problem with missing parts you should check your paste handling process, time after printing and handling, there seem to be most of the problems in that case.

M2C

Wolfgang

reply »

You can test for solder paste tack using IPC-TM-650 method 2.4.44. This will give you an idea of the adhesive strength of your paste prior to reflow.

reply »

We agree with Wolfgang. Adding points along his line of reasoning: * Also consider placed component prescience. Sometimes, a mis-pick results in paste on the nozzle, which leads to the next component being stuck on the nozzle and not placed, which leads to additional mis-picks. * Continuing, people have related experiences of moving components from the placed position during reflow with hyper-active blowers, but we have never seen it done. * Finally, biggest clue to the cause of your missing components is the missing components. It you find a neat little [hopefully little] pile of parts, the cause of your angst is not far up-stream.

Now responding more directly to your question, we use a tack test [probably not unlike the test Bill suggested] to compare the characteristics of solder pastes as an element of a paste qualification, but not as the production process control that you suggest. In this qualification, we perform a �

5.2 Post-Placement Tack Test

5.2.1 Using the stencil described in paragraph 3 above. [NOTE: We use our stencil for a variety of paste tests. For this test, printing an array of approximately 100 parts using production standard pad layout is sufficient.] Print paste on a bare board or laminate.

5.2.2 Place 1206 (or 0805) discrete components in the paste.

5.2.3 Wait one hour.

5.2.4 Gently, invert and support the board above a table.

5.2.5 At one hour intervals, record the number of components that have fallen from the board.

Tell us more about the details and distribution of the situation that makes you question the adhesion properties of your solder paste.

reply »

If you hold your board sideways in a 90 degree angle you apply a force of 1 g to your components. 1 g translates into 32 ft / sec. square. This is not too much acceleration for fearless roller coaster drivers but Melf�s and Tantalums can fall off at 1-2 g. You can create 2-3 g by tapping the edge of the board onto a table. I think these tests are very well reasonable because all chip shooters move the board during placement process and some pick & place machines �shake� the boards in Y-direction. If the axis accelerates at 1 g, it reaches a short distance of 10 mm in ~ 100 ms, at 2 g in ~ 50 ms.

There are a few more �rattles� before the board can relax in the oven. One is created by the air blow of the nozzle, designed to release the component. If you turn up the air that the nozzle whistles, you get an effect similar than holding your thumb on the water hose. It may not effect the component just placed but the neighborhood.

The Z-axis accelerates with 2-4 g and can easily shake the components up if incorrectly programmed.

When the board is done, but you took too much time to refill all the components and the paste is dried out, than the output conveyor may mess up your board as well.

reply »