Flux Selection for Wave Soldering Process

Six Sigma – iSixSigma Forums Operations Manufacturing Flux Selection for Wave Soldering Process

Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
  • #239485


    Hello all

    Do you have any experience of selecting right flux for wave soldering process? We have different flux suppliers and we need to evaluate them and select one flux that we will use.

    There should be possibility to produce specific board that can help us to evaluate the best flux.

    Is there any experience or link where I can get more info how to evaluate different flux for wave process?

    There is high influence of many inputs – if we run several batches to evaluate different flux.

    Thank you for recommendations.


    Mike Carnell

    @tomasSpisar My training in soldering was from Ralph Woodcock (I think I got that correct) and was in the 1980’s. Here is the basic piece of information. A good solder joint only requires 3 things 1. heat 2. intimate contact 3. clean solder able surface. So what is flux for? Flux cleans a dirty surface. Theoretically you don’t need it if your boards are built and handled properly.

    We were in a government facility so we could only use RMA at the strongest. So everybody did stupid stuff like drive the specific gravity of the flux up trying to throw more solids on the boards. Nobody thought to keep them clean in the first place. Basically in your situation you can use a less active flux more successfully if your boards are cleaner. The advantage is a less active flux will have less ionic contamination and the board will less susceptible to dendritic growth. With todays lines and spaces I would assume you can short a board in the blink of an eye.

    Part of what gave use issues with high specific gravity was the wicking action of the flux was affected by the high solids so fluxing the topside of the board was difficult. I would assume there is little to no through hole technology today so probably not an issue.

    We did all of our testing on a group of prototype boards. We started by scrubbing them clean which was relatively easy in a big tank of tricolor 111. You probably don’t have access to such a benign chemical cleaner. Ultimately they were run through a cleanliness tester to assure they ere all at the came level of cleanliness. We set up treatments using a group screen design which was a fractional factorial with 4 groups of 4. The top group was then put in a full factorial design.

    You might be able to find a write up on this in one of Mikel Harry’s old books. I have no idea what the title of the article was but the title references Mil-P -28809 (something like that) It was also published in an Industrial Engineering book from Jay Heiser.

    Bottom line is the less contamination on the PCB the less you need from your flux so keep the boards clean and avoid the ionic contamination issue.

    I did this work with a bunch of people and magically the group got bigger once the results showed about an 80% reduction in defects. This was also the same thing that taught me to not trust brain storming. We had this down to 16 variable which is way to many for a normal DOE. We had close to 30 experts rank order those variables for the group screen design. The one factor that drove the whole thing was I think #12. 30 people “experts” could not even get it into the top 10. To a person when the final report was published they would stop us in the halls and tell us “If that was all you wanted to know you should have asked us.” Be polite ask for their input and then do whatever you think is right. Do not let them intimidate you.

    Good luck,

Viewing 2 posts - 1 through 2 (of 2 total)

You must be logged in to reply to this topic.