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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

Help with flux residue spec



Help with flux residue spec | 1 November, 1999

I have a thru-hole PCB that malfunctions when "no clean" flux is left on certain parts of the board. I have specified "the PCB shall have no flux residue". Our state side vendors (we have one offshore -they still use freon to clean) are complaining that this specification is very difficult, if not impossible to meet. They are both small companies (20-30 people) and niether has a wash system in place.

Can someone please offer some advice as to what I can expect from good quality mfg houses? Is my spec of "no flux residue" unreasonable? Is there a good way to test, other than visually, for residue? Any help would be greatly appreciated.

Thanks, John

reply »



Re: Help with flux residue spec | 1 November, 1999

No clean flux is really a misnomer. What is often taken to mean no residue in effect means no visible residue. Because you can't see it doesn't mean it isn't here. Many manufacturers have been through this same learning curve. The goal with a no clean is to apply the least amount possible while still attaining acceptable solder results. The "correct"amount of no clean is designed so the solvent evaporates in the preheat stage while the non voalites and activators sublimate or decompose when in contact with the molten solder at the wave. Too much flux and it doesn't all burn off. Too little and it doesn't solder well. The window of opportunity is much smaller with a no clean. The standard band aid for poor solderability is to flood the assembly with more flux. How clean it is after soldering does not come down to visual. Rather the specifiaction to which you build, for example IPC Class 2 will have corresponding cleanliness specs. Doug Pauls & Terry Munson at CSL in Kokomo In phone 765 457 8095 can perform just about any analysis you can ever want on flux residue. Hope this helps you John.

reply »

Doug B.


Re: Help with flux residue spec | 3 November, 1999

Yes this is absolutely true. No-clean flux residue has various affects on the functionality of circuitry especially high impedance circuits. It also affects solderability, and reliability. No-clean flux will always leave a detectable residue. Most flux suppliers will provide a "flux solids deposition" spec. This is given as flux solids per squar inch. A simple test of the flux solids deposition can be done using a standard 8.5" x 11.0" mylar transparency film. Basically, you compare the weight over surface area before and after flux is applied using your normal process setup. The key is that the transparency is completely dry before weighing it so that only the solids are left on the mylar. My rule of thumb is that if you have a solids deposition greater than 1200 u grams/sq. inch, you are apply too much flux. Using an analytical technique like ionic chromatograhy as described in the previous reply is very helpfull in determining if electromigration potential exists. I know this is long but I hope it helps.

reply »



Thanks chartrain and Doug | 4 November, 1999

Thanks folks. The current vendor is doing an alcohol wash, top and bottom, then performing an ionic test on 100 boards. I'm not to confident in the testing because the readings for a "clean" board are 25 ugm/sq in and a "dirty" board are 40 ugm/sq in. Not much of a difference. Their ionic tester is checking for only one type of compound-some type of chloride (Na??) Although I haven't tried it, I suspect a SIR would be better. Better yet, I'd rather have them use a cleanable flux and run it thru a good old fashioned dishwasher!

Thanks Again, John

reply »


Re: Thanks chartrain and Doug | 5 November, 1999

John: Three things:

1 You are correctamundo!!!! What's the difference between slopping alcohol on dirty boards and washing dirty boards? Clean boards!!!! By the time those very nice people get done with their alcohol cleaning (a nip for the board, a nip for me, a nip for the board, two nips for me ... oops sorry there, I got lost in a fantacy) the board you're gonna pay beaucoops of $$$$. 2 Sorry I don't get it!!!! Why are these no-clean boards reading boat-loads of ionics? No-cleans are supposed to be low in ionics. Continuing, ANSI-J-001A, MIL-P-28809, MIL-STD-2000 say LT 10.07 �gm/sg in NaCl equivalent is the limit. No-clean SB LT 3 �mg/sq in in NaCl equivalent. 3 Finally, it would be nice if you could make a piles of good and bad boards, measure the total (ionic/nonionic) contamination on them, and develop a specification for surface contamination that works for YOUR boards.


Dave F

reply »



Wow! Those are low limits. | 5 November, 1999

I think I'll have to get a copy of those specs. This vendor claims that the 40 uG/sq in is IPC class III and is very, very, clean. I would like to show him (or at least upper management)that what he's says just isn't so. Unfortunately, I don't think that NaCl is the problem. Today we tested "clean" boards (that work) from our offshore source. They use rosin based flux and clean with Freon. They are much higher, 2-3 times more, in NaCl. So, it would appear that NaCl testing doesn't work in this case. I offered to examine other tests, to see if there is one that would work for our case (SIR, other surface contaminents, etc..)but, if I find one we would need to buy the equipment and do periodic testing ourselves. Of course, that was a big NO. So, finally, we said screw it-just have them use a water soluble flux and run it thru the "dishwasher".

To add insult to injury-they want about $1.80/bd more for the water wash. Our board is only 9 sq inches!

Yes Sir, I think I'll buy those specs first thing Monday morning!



reply »



Re: Wow! Those are low limits. | 11 November, 1999

1. Most ionic testing is measured in NaCl, that is, the ionic residues are quantified in terms of NaCl equivalence. The test does not determine what the ionic residue is, just its relative strengh. 2. I use IPC cleanliness specs on OA fluxes but not on no clean fluxes. I've been away from no cleans for a while, but when I worked with them, the vendor freely admitted ionic testing would fail. In the case of that specific flux, it would easily dissolve in alcohol and yield a very high reading. But how often do your electronics function in an alcohol condensing environment? By the way, this situation is what spawned the SIR testing. SIR (surface insulation resistance)is just another method to quanify residues in terms of impact on circuits. 3. SIR testing is usually 'jobbed out' to a test facility and used to qualify the flux materials and the assembly processes.

Using OA flux and washing is a good solution!

Hope it helps, Boca

reply »

Used SMT machines

reflow oven profiler