4
\$\begingroup\$

I'm using this brand and type paste flux, and it works very well... except for one small problem. After assembling 5 prototypes, most had bizarre issues, which were resolved when the flux was removed via 91% isopropyl alcohol bath, soak, scrub, and dry.

Now this flux claims to be "no-clean" and "non-conductive", however the behavior was such that it was at least partially conductive. Some of the signal paths are very high impedance (10M ohms) but ironically, most of the problems centered around SOIC-28 and 100+ pin LGA components at less than 5 volts.

What's your experience with no-clean paste flux? Do they become at least slightly conductive after heating?

\$\endgroup\$
  • 2
    \$\begingroup\$ Have I ever had a circuit where I was sure the flux residue was the cause of the problem? Yes. Did cleaning it fix the problem? Yes. Did I ever find a way to prove that was the problem? No. :) If you do some searching you'll see other anecdotal evidence of the same kind of issues. Would be nice if someone here knows more. \$\endgroup\$ – Some Hardware Guy Mar 12 '15 at 16:14
6
\$\begingroup\$

While flux may be pristine before it's used, its job is to clean the surfaces of the metals the solder is supposed to join, physically and electrically. So where does all the junk the flux dissolves go? Into the flux, of course, and if there's enough of it and it's the right kind of junk, it could well provide a conductive path through the cooled flux which will be removed when the board is cleaned.

\$\endgroup\$
  • \$\begingroup\$ The boards I'm using are ENIG gold-plated. I'd read others having corrosion issues after no-clean flux. Do you think it could be something in the gold plating process which creates extra impurities, which then causes the flux to become more conductive on this particular type of PCB? \$\endgroup\$ – rdtsc Mar 12 '15 at 18:00
  • \$\begingroup\$ @rdtsc: Sorry, I don't know. \$\endgroup\$ – EM Fields Mar 12 '15 at 18:21
6
\$\begingroup\$

Yes, many of us have had problems with conductivity of no-clean flux. It's not just anecdotal, a quick search will yield solid studies- which in part seem to blame the process conditions, but as a user, "whatever" it still can create invisible terrible problems, and if they're intermittent because of whatever the contract assembler is doing, that's actually worse.

To quote the article linked above:

In all three cases the assembly failed for voltage leakage or intermittent signal quality due to conductive ionic residue between fine pitch leads

It's even worse with precision analog circuitry (I have a design in which 5G leakage makes a difference because that causes significant error), and compounded by the difficult in cleaning off the "no clean" residue, which is an order of magnitude more difficult than cleaning ordinary RMA flux residue.

\$\endgroup\$
  • 1
    \$\begingroup\$ Thanks Spehro. It's been a long while... Last we talked was probably on the JAL list or PIC list a decade ago. :) But I will keep this in mind, especially with ENIG-plated boards. \$\endgroup\$ – rdtsc Mar 12 '15 at 18:05
  • 1
    \$\begingroup\$ @rdtsc Ah, now I recognize the name. Good to see you. First saw this problem back in the 1990s on a thermocouple temperature indicator PCB we were making in quantity. \$\endgroup\$ – Spehro Pefhany Mar 12 '15 at 18:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.