No-Clean Residue Shorts



No-Clean Residue Shorts
We have a PCBA with a QFN 44 using no-clean flux. We have many test failures during testing. Are the failures due to flux reside between terminations?
Board Talk
Board Talk is presented by Phil Zarrow and Jim Hall of ITM Consulting.
Process Troubleshooting, Failure Analysis, Process Audits, Process Set-up
CEM Selection/Qualification, SMT Training/Seminars, Legal Disputes
Phil Zarrow
Phil Zarrow
With over 35 years experience in PCB assembly, Phil is one of the leading experts in SMT process failure analysis. He has vast experience in SMT equipment, materials and processes.
Jim Hall
Jim Hall
A Lean Six-Sigma Master Blackbelt, Jim has a wealth of knowledge in soldering, thermal technology, equipment and process basics. He is a pioneer in the science of reflow.

Transcript


Phil
And welcome to Board Talk with Jim Hall and Phil Zarrow, the Assembly Brothers, who are also known as ITM Consulting. We are here to discuss your problems with processes related to electronics assembly and SMT, whether it is materials, components, procedures methodology, equipment.

Today we have an interesting question from TY. We have a PCBA with a QFN 44 using no-clean flux. We have encountered many test failures during testing. The test engineer claims that the failures are due to flux reside between terminations that affect the capacitance of the joints. Is this a valid claim?

Good question. And the answer is, yes it could be. Of course, it could be something else too but with regard to the flux, yeah.

Jim
First off, let's say that the QFN is a type of bottom-terminated component, BTCs, to keep our terminology straight with the IPC specs.

Yes, I agree that no-clean residues, under bottom-terminated components, QFNs, has been shown to have negative influences on certain assemblies.

Years ago when BTCs first appeared Terry Munson of Foresite, one of the true cleaning gurus of our industry, pulled some QFNs off the board and started looking at the residues underneath and found that no-clean residues were not completely deactivated.

Of course with a no-clean residue you are supposed to be able to leave the residue on the board because during the thermal cycle of the reflow the active flux materials are deactivated and the residues are benign and should not cause influence on the performance of the circuit, which would result in the test failures that are being seen here.

This theory, and a number of other people have advocated this, is that what is happening is that although the flux material under the BTCs reach the proper temperatures because they are confined under the chip complete evaporation does not occur.

Evaporation is one the mechanisms that no-clean materials use to deactivate the paste by evaporating the chemicals, chemically reacting them, and encapsulation.

The feeling and the observation when people pull BTC components off is that the no-clean residue are not as hard as they were before, indicating the presence of non-evaporated solvent type materials.

Also more recently some consortiums, particularly led by Mike Bixenman of Kyzen, are also evaluating and they are taking it a step farther to look at SIR values measured under the residues underneath.

They are also showing, from a quantitative standpoint, it has no-clean residue even when completely reflowed at the proper temperatures still fail SIR test after reflow.

The answer is to clean, even if it is a no-clean. It may be necessary for your parts to actually do a cleaning. That means coming up with a cleaning methodology, a chemistry and a process that will remove those no-cleaned residues from under those parts.

Phil
And I just want to add, I believe that I have seen papers along this situation even when you have properly reflowed residue that certain frequency ranges can interfere with the functionality of the circuit. You would have to research that and see if your application falls into these suspect frequencies.

I am sure one of our readers will comment on it. We are typically talking gigahertz range but I don't remember the specifics. Again, if that is the situation as Jim is saying, you may have to go to a water soluble cleaning process. It's a process, not just a material change, it is a process change too.

And if you have type of devices where certain parts can't go through a cleaning operation then we look at primary ops with a water soluble and thermal cleaning and then do a secondary op with a no-clean.

Jim
One more possibility. Another plug for current solder paste.

This is an issue that is being addressed by solder paste manufacturers and some of them have flux formulations that perform better in terms of deactivating under BTC components. Something else you might want to investigate to solve this problem.

Phil
So we hope that we shed some light on it. It puts you off in a few directions there.

You may also want to take a look at some of the solder paste companies, including your supplier and see if they have done some research on this or ask them about it. They know their formulation. They know their chemistry. Or they better or it is time to switch solder pastes.

You have been listening to Board Talk with Jim Hall and Phil Zarrow from ITM Consulting. We hope we have enlightened your mind in this particular fun subject.

Whatever you do, however you're soldering those, whatever capacitance problem you may have, please don't solder them like my brother.  

Jim
Don't solder like my brother.

Comments

A few questions, have these failing boards been X rayed or has CSAM acoustic microscopy done? What kind of failure analysis is being used? Are you absolutely sure there are no residual solder balls intermingled here? Like others have mentioned is this an RF part and circuit? We definitely see issues with to much flux. Seems like the proper paste may fix your problem. Possibly you are also printing to heavy? Can you make printing adjustments or stencil mods? One lesson I am having to push very hard to our students is board cleanliness, flux removal and final inspection.
Fred Haring, North Dakota State University
You guys are old enough to remember the origin of “no clean” fluxes They were originally designed for surface mount components in wave soldering. The thick gooey rosin fluxes couldn’t flow around these pin head sized components - so the fluxes were thinned to allow better penetration. Then, because the residue was no longer visible, it, by marketing terms became “ no clean” Several generations later no clean is taken at face value with resultant consequences. I might also add that lead spacing back then was like picket fence posts compared to today. Just because you can’t see. It doesn’t mean it isn’t there!
Ray Chartrand, Chartrain Consulting
This is a perfect example of the adage "No Clean? No Way!" ... You see, if you have a high reliability component, or if your assembly is built up and going through test, why would you NOT want to remove the residues residing between runs or leads? Seriously. Why would you even risk it? Cleaning processes are NOT expensive. In this case of a bottom terminated QFN type of part, you need to use a super low surface tension fluid to move out the contamination. Personally, I would use a vapor degreaser, or vapor defluxer or even a vacuum vapor defluxer. There are various sizes of equipment, but for small parts they start, with automation, in the $24K price range.

Pair that with a solvent based on 80% trans dichloroethylene, and Bam; clean parts with no water, no moisture bakeout, no long process. Its not that expensive, cleans under the parts, automation ... How can you afford to NOT clean your boards, if you want reliable parts. That is, unless you are building electronic assemblies that you want to fail after a limited service life ... then I understand why you would leave the crud on the board. Repeat after me: "No Clean? No Way!" LOL
Rick Perkins, Chem Logic
This sound like a Win-Win for Business, Environment, pricing and availability of Lithium and so on. Could Lithium (and other elements) be recovered from seawater as well?
Calvin Mosher, Nautel Ltd
I suspect that this is a leakage current issue. With wet flux under the QFN that's a low standoff component, the solvents in the flux allow for ions that cause dendritic growth & a path for leakage current. This is a big issue in higher power applications that use QFNs & MOSFETs. Flux chemistries with enhanced SIR & ECM reliability will certainly help.
Karthik Vijay, Indium Corporation
It would be interesting to know which kind of activation has the flux used, if it is an L0 or L1, or others, and if the solder paste is compliant to IPC J-STD-004. Then it would be interesting to know which kind of tests are done on the boards, if they are doing humidity tests or not, and if the PCBA is "as-is" or closed inside a plastic or metallic box. This information can help the discussion. Typically a no-clean flux compliant to J-STD-004 has a 100MOhm minimum impedance, so quite difficult that it makes shorts, unless it hasn't received enough heat. Maybe a temperature profile on this particular device (a probe on the top of the component, a probe under the component) should help to better understand if the soldering profile is OK or not.
Pierpaolo Galli, Meta System SpA
I've generally wondered, being an RF guy, why anyone would use a no-clean flux, since by definition it leaves "crud" behind. It HAS to have some resistive and dielectric properties, and looks like crap, whereas water-soluble flux yields beautiful looking, clean boards. No solvents needed anymore, and water is cheap as a cleaning agent.
Allan Knox II, Knox Associates Design
Solder paste flux residue may indeed change the capacitance of a solder joint. SIR is because of the in-phase, resistive impedance between closely spaced conductors. Capacitance is the out-of-phase impedance. No-clean solder paste flux residue is a dieletric material; it contains a lot of rosin or resin, which is put there to improve SIR. Unfortunately at >GHz frequencies, this additional dielectric material can affect the circuit cpacitance and change signal responses.
Karen Tellefsen, Alpha Assembly Solutions

Submit A Comment


Comments are reviewed prior to posting. You must include your full name to have your comments posted. We will not post your email address.

Your Name


Your Company
Your E-mail


Your Country
Your Comments