Effective Low-Mid Volume Solder Flux Removal

[MATLABdude]

I've recently accepted a short-term contract position as basically an electronics technician / solder monkey and we're using rosin-core silver/tin/copper solder for our boards for RoHS compliance. Up to this point, I've only done low-volume production where it's been more or less acceptable to finish off flux removal (after soaking / brushing) by hosing down boards with enough flux remover (MG Chemicals Flux Remover) or Superclean to ensure that no residue or re-deposit is left behind.

However, now that I'm expected to produce in moderate volume (dozens of boards) I'm wondering if there's an alternative to going through cans of this stuff, or if that's just the cost of production? Should we just be changing the primary flux remover wash more often to dissolve the amount of dissolved Rosin? Should we switch over to a water-soluble rosin, and just put boards through the dish washer? A no-clean formulation, and just leave be?

Some of the best results I've achieved thus far have been to immediately taking the boards from the wash / scrub pan, and washing the whole thing under water (thankfully, no cardboard speakers or transformers), and then using the air compressor to clean off the water. However, I'm a little leary of leaving moisture under components / inside connectors, and was pretty paranoid about drying off the boards.

That all said, I was wondering if anybody had any suggestions for effective low-mid volume defluxing of boards?

 

[mdjensen22]

If the boards need to look pretty (no flux residue), then I would recommend a water soluble flux first pending the board can be washed and a no-clean flux secondly.

Generally speaking, rosin core would only be used if you don't care about the 'tree sap' look since the cleaning process is more time consuming/costly than water soluble.

Many places have a 'PCB washer' attached to their wave solder if they use water soluble flux. After washing the boards, they are typically placed in an oven for drying.

Edit...
Some things to note:
No clean flux residue can have undesired effects on low-voltage/high-frequency analog circuitry (personal experience). I did a design that had to measure uV ranges at the 100KHz ranges - the manufacturing and touch-up processes needed to be monitored carefully as to not alter the performance (prior to conformal coating, of course).

I would recommend deionized (DI) water for cleaning (depending on job requirements)http://www.circuitnet.com/articles/article_67321.shtml
http://www.finishing.com/Library/flux.html

 

[dlgoff]

It was a long time ago when I was up against the same situation (project used 50K ICs). Since we were a small company, we purchased a solder pot as opposed to a wave solder machine. I would pick up the stuffed boards with a homemade sheet metal tong, press the solder side on a resin flux soaked pad in a large baking dish (making sure flux came up through the plated through holes), used a piece of cardboard to skim the dross from the melted solders surface, ran the boards while pressing them into the melted solder (making sure that solder wicks to the top of the through-holes; practice makes perfect), and immediately into a tub of http://en.wikipedia.org/wiki/1,1,1-Trichloroethane" . Once I had a full tub, which removed most of the flux, I placed them in a ultrasonic cleaner with 1,1,1-Trichloroethane for time to remove any remaining flux. The Trichloroethane tends to leave some residue so from there they went into a tub of hot soapy water followed by a clean water rinse. The boards solder side came out looking like a mirror.

 

[MATLABdude]

Thanks for the suggestions!

I ended up hand-soldering a bunch of PCBs with water-soluble flux lead-free solder (downside to water soluble flux: your hands feel really funky after you've been handling soldered boards). I washed the boards in Sparkleen (the detergent I had on hand, though I suspect nearly anything would've worked) and hot tap water, rinsed thoroughly with deionized, and dried using nitrogen / CDA (clean dry air). Some of the connectors we used (screw-down style) were a little hard to dry, but I believe I got all the moisture.

Bright, shiny (and, most importantly, flux-free) PCBs!

 

[yungman]

Methanol might work. I don't like water solder, they eat up tips. Compress air will clean up nicely.

 

[MATLABdude]

Methanol might work. I don't like water solder, they eat up tips. Compress air will clean up nicely.

You got that right: in two weeks of soldering, my (new) tip looks worse than the tip I'd been using for the previous two years! The scorched flux is quite persistent (but I haven't really tried any of the more aggressive cleaning methods yet, as I'm able to still get solder wetting).

Are you suggesting methanol to displace the water and make drying easier?

 

[yungman]

You got that right: in two weeks of soldering, my (new) tip looks worse than the tip I'd been using for the previous two years! The scorched flux is quite persistent (but I haven't really tried any of the more aggressive cleaning methods yet, as I'm able to still get solder wetting).

Are you suggesting methanol to displace the water and make drying easier?

Methanol is very good in cleaning conventional flux, if you use compressed air, you don't really worry who dry easier. You want to get the liquid off the board before it dry and leave stuff behind. Those flux residue can cause leakage in circuits if the impedance of the circuit is very high or the circuit is working in HV environment.

I forgot, those water solder smells terrible too!

 

[jim hardy]

in early 1980's i visited a local elctronics factory.
The engineer giving the tour showed us his PCB wash station.
It was five ordinary Sears household dishwashers using everyday dishwasher detergent.
He said it did a fine job and gave him no troubles with hazmat solvents to dispose of.
They dried boards in a warm oven with fans.

 

[cmb]

I washed the boards in Sparkleen (the detergent I had on hand, though I suspect nearly anything would've worked) and hot tap water, rinsed thoroughly with deionized, and dried using nitrogen / CDA (clean dry air). Some of the connectors we used (screw-down style) were a little hard to dry, but I believe I got all the moisture.

Components that have nooks-and-crannies that may be difficult to dry tend to come with some recommendation of mounting, rinsing flow/time, and drying flow/temp/time, esp things like the larger surface mount capacitors and resonators where you need a little underside clearance to ensure good wash/dry to avoid problematic failures from later humidity that might form low impedance paths underneath, if not cleaned sufficiently.

I'd recommend time in ultrasonic/purified water bath, if there are any doubts at all on the 'rinsing process'. Sounds too uncontrolled at the moment to be sure.

 

[yungman]

When I was working for the mass spectrometer company, we had a ultra sound freon bath washer because not only we need to clean PCB with very high impedance and 10+KV circuits, we need to clean stainless steel components that goes into ultra high vacuum. The tank was design in the way that it is quite deep with cooler coil at the top opening so when freon evaporate from the bottom, it is condensed back when it try to rise and escape from the top opening. We hook the pcb and dipped into the dirty freon at the bottom and let the ultra sound shake loose all the dirt, then we slowly pull it pcb up, let the droplets from the recondensed freon from the top wash it clean on the way up. The bath at the bottom was very dirty, but the recondensed freon is pure and clean. When you pull it out, the board is shinny clean! I don't know much about these equipments but it did not look expensive.

If you are doing it for manufacturing, you might want to consider this. It is very fast. You can dip a few boards at a time using a metal basket and the process is just 20 seconds, no scrubbing.