Ask the Experts
August 16, 2018
Gold Edge Contact Flux Contamination Failures
We received somereturned circuit board assemblies that failed after some time in the field. These assemblies were originally processed using no-clean flux. The assemblies had visible contamination on the gold edge contacts and analysis confirmed that there was flux residue on the goldedge contacts.
We sent the assemblies back to theEMS supplier to properly clean them. The EMS supplier claims that the residue on the gold edge contacts alone would not cause functionality problems. What do you say?
Expert Panel Responses
I'd say youneed to run some tests. If the flux residue is causing interruptions orintermittent contact, then the residue should be cleaned.
If you and your supplier are working IAW IPC-A-610E, you'll seein section 10.6.1 that the defect for all classes calls out "Note 3: Processesdesignated 'no-clean' need to comply with end-product cleanlinessrequirements." While that doesn't see like it's much help, it does give you,the user, the responsibility of defining how clean your "no clean" assemblyneeds to be. If that hasn't been done already, the supplier has a bit of wiggleroom on "How clean is clean?"
What seems moreapplicable to your situation is section 10.1 for Gold Surface Contact area. Thedefect condition states, "Solder, any metal other than gold, or any othercontamination in the critical contact area of the gold surface fingers, pins orother contact surfaces such as keyboard contacts."
Notice, "... orany other contamination in the critical contact area ..." Here's where your testswill come into play. As I stated above, If the flux residue is causing breaksin the contact, then the residue needs to be cleaned.
Unfortunatelythe marketing people coined the phrase "No Clean" as a sales gimmick for themore properly named, "Low Solids" or "Low Residue" fluxes. Just because it'scalled no clean doesn't mean the assembly doesn't have to be cleaned.
Check out the newIPC-CH-65B for a more in-depth discussion of cleaning "no clean" assemblies.
Manager of Assembly Technology
Kris Roberson has experience as a machine operator, machine and engineering technician and process engineer for companies including Motorola, and US Robotics. Kris is certified as an Master Instructor in IPC-7711 / 7721, IPC A-610 and IPC J-STD 001.
Boards that have beenmanufactured with "No-Clean" flux may still require cleaning prior todelivery to service. This depends on the environment to which theywill be placed. For example, if the "No-Clean" boards areplaced into a humid environment without a conformal coating, with temperaturevariations, and (optimally) a continual low-power current, the conditions areperfect for the dendritic growth to occur.
If the operatingenvironment has these characteristics, then you should be cleaning your"No Clean" boards. Others may tell you different,but you have the proof in your hands. The gold fingers can easilybe wiped with a good solvent blend and retested. If the problem stillexists, you probably have some shorted areas on the board. If youneed more info on remedies, feel free to contact me.
Rick Perkins is a chemical engineer with more than 33 years of Materials & Processes experience. He has worked with Honeywell Aerospace in high-reliability manufacturing, as well as with several oil-field manufacturing companies. He also has a good understanding of environmental, health, and safety regulations.
I cannot agree with the EMSsupplier. The gold edge contact is a critical interface surface formating with the connector. No clean flux while not necessarily having acorrosive nature can have an insulating effect which could cause intermittentfailure between the contact and the connector. From IPC-A-610E section 10.1contamination on contacts is a defect.
VP International Relations
David W. Bergman is Vice President of International Relations for IPC. He has worked at IPC for more than 30 years responsible for IPC standardization efforts, education and certification programs. Currently, David is responsible for globalization activities, including IPC's China and India offices and reps in Europe and Russia.
No-clean fluxes come in different classes, the ROL0 or ORL0 ispreferred for reliability. This would need to be verified.Functionality may or may not be affected since this is due tomany factors, however all no-clean fluxes need to see soldering heat tode-activate completely.
If flux gets into regions which do not see soldering heat, theseareas would normally need protection from flux ingress, if there is a potentialfor corrosion or electrical conductance.
If flux is suspected to be present in excess and has not seensoldering heat, it would be good to first confirm the presence of the flux'sweak organic acids. Determine the WOA using ionic testing such as ionicchromatography is a method often used. It should be very low and the region suspected of failure shouldnot exhibit electro-migration issues when examined under a microscope.
If it is determinedthat flux is the issue for the failure, it would require either the goldfingers to be covered to avoid flux contact or reduce the amount of fluxapplied to reduce the likelihood of excess flux permeating areas where it isnot desired. Also increasing heat contact by either increasing thepreheat temps and times or solder wave contact times will help de-activate theflux.
Senior Market Development Engineer
Mr. Biocca was a chemist with many years experience in soldering technologies. He presented around the world in matters relating to process optimization and assembly. He was the author of many technical papers delivered globally. Mr. Biocca was a respected mentor in the electronics industry. He passed away in November, 2014.
Sincethe edge connector contacts wipe across the surface they open up new materialso they might not cause initial functionality problems but the residue on thesurface might very well cause corrosion/long term reliability problems down theroad. A lot depends on the reliability requirements and expectations for thepart. You need to understand the long term implications, not just the shortterm effects.
Process Solutions Consulting Inc.
Lee Levine has been a Process Engineer and Metallurgist in the semiconductor industry for 30 years. He now operates his own company Process Solutions Consulting Inc where he consults on process issues and provides SEM/EDS and metallography services.
It really depends onwhat the gold pads will be used for when the product is in use. Mosttypically, these gold edge pads (or "gold fingers") are used as apress fit connector into another assembly, and it is critical for the gold tomake good electrical contact into whatever type of substrate or mother board itis being plugged into. For this reason, most assemblers try to preventanything at all from contaminating the gold fingers. The prevention ofdebris onto these gold surfaces will insure reliable contact when they are"plugged in" to another board.
If cleaning is anoption at this point, I would suggest pursuing that. Contaminated goldfingers can impact functionality in the field.
General Manager - Electronic Assembly Americas
Mr. Smith has been supporting customers in the electronics assembly industry since 1994. His expertise is focused on solder paste printing and reducing soldering defects. He holds a BS in Chemical Engineering and an MBA in Marketing. He has authored several papers in trade magazines and at industry conferences. He is an SMTA Certified Process Engineer.
Find a new EMSsupplier. Flux residues on thelaminate will cause premature failures and also promote ionic migration,pending signal frequency and premature failure. A response like this,tells you a great deal about the competency and integrity of your assemblyhouse.
Capital Equipment Operations Manager
Specialty Coating Systems
Rodney is currently Operations manager at SCS coatings, Global Leader in Parylene and Liquid Coating equipment. Rodney applies his BS in Computer Integrated Manufacturing from Purdue University, along with 20+ years of Electronic manufacturing and Equipment Assembly, to direct the Equipment business at SCS Coatings. "We provide unique, value added coating equipment solutions for our customers". Including conformal, spin and Parylene coating expertise.
A number of questionscame up as I read about your problem. I will attempt to touch on all of them aswell as answer your question.Many, if not all"gold plated edge connectors" consist of fingers or pads on thePWB which slide into a mating connector, which is itself mounted onto abackplane that holds all of the circuit card assemblies (CCAs) in the device orproduct.
The mating connector is usually either press-fit or soldered to thebackplane. I assume you are speaking of this type of design? Or could it be anedge-mounted pin connector, rather than gold-plated PWB pads?
This may be irrelevant,because nearly all connectors have a very small contact area, or pinch point,if you will. Very few, if any, connector designs make contact over the fulllength of the pads or pins, with the exception of mating pins/socket types.
Or it can be a combination of any or all of the aboveconditions occurring. Hopefully this and others contributions will provide alist of root causes to investigate and eliminate. But if contamination can beseen and detected, I would start by eliminating that first. Go for "pristineappearance" at final inspection.
- Itis always possible that the target mating connector (the one mounted to thebackplane) may be at fault due to flux from rework, or flux fromrework/assembly of adjacent components. This could be the possible source ofthe no-clean flux found on the edge-connector on "your" CCA.
- Nomatter the source or type, anyflux or any other FODthat is visible or verified present on the edge-connector or in the matingconnector is sufficient to cause a malfunction. This is because the pinch pointarea is very small. A single fiber or very light residue of no-clean flux canvery easily cause an open connection, or a high-resistance connection, leadingto electrical malfunction. At room temperature, the no-clean fluxresidue may be of a "greasy" appearance, but at hot or coldoperating temperatures it may be a very hard residue. Even if only slightlysticky, any particle of FOD can stick to it and roll under thepinch-point, causing an electrical malfunction.
- There should never be any evidence of flux residues, conformal coating, FOD, oranything else on the connector contact lands or pins. Small solder splattersare allowed as long as they are not in the contact area (connector mating areawhere the pinch point is located), per IPC standards. Careful masking of edgeconnectors to prevent contamination is required, and if tape is used itshould be a low-residue, ESD-safe tape made just for this purpose, or some typeof slip-on gasketing material, perhaps as part of the assembly fixturing. Thefixturing in turn must also be cleaned. For pin connectors, some type of shroudshould be used to prevent solder splatters or other contamination. These areavailable in ESD-safe plastics for most connector types.
- Sometimes,but very rarely, press-fit mating connectors (in the backplane) candevelop an open condition.
- Sometimes connector contacts in the pinch-point area can develop opens due tovibration in the device while in operation (especially true of any automotive,machinery, or avionics application) which causes the pinch point to rubback and forth across the gold pad, exposing the basis metal underneath thegold plating (almost always nickel). This is called fretting. The exposednickel oxidizes very quickly, creating an open or high-resistance contact. Inthis case, any other FOD or flux can wrongly be blamed. Fretting can sometimesbe very difficult to detect, even at 30x magnification. Remember, it only hasto be present right at the really small pinch-point location to causeproblems. Keep in mind on some types of connectors, a gel is purposelyapplied to the connectors to lubricate them, to prevent fretting. While it mayserve that purpose, it can also attract and hold dirt particles (FOD).
Richard D. Stadem
Richard D. Stadem is an advanced engineer/scientist for General Dynamics and is also a consulting engineer for other companies. He has 38 years of engineering experience having worked for Honeywell, ADC, Pemstar (now Benchmark), Analog Technologies, and General Dynamics.
LS this failure scenario is likely a result of several factors.A no clean flux when processed correctly will meet many acceptable standards,Bellcore, IPC, etc."No Clean" is actually "Low Residue". However now you still havea contaminant residing on a circuit. Is it benign and non-conductive? More thanlikely yes it is considering the hours of development by the flux and solderpaste suppliers, so according to your EMS he is also correct, yes it isperfectly acceptable to leave this low residue on the pcb.
In a perfect world with a clean environment and atmosphere youwill likely get many years of successful operation. However, place this circuit out in the field the tables are nowturned against you. Salt air along coastal regions, smog and exposure to avariety of other industrial pollutant and fine air particles in our environmentcan result in a visible contaminant now seen on your pcb's that will result ina failure. This flux residues can act like a sponge and absorb atmosphericcontaminants.For many of my customers No Clean flux does not mean no cleaningat all. What are you building? Is the extra cost of cleaning up front justifiedor do you risk a product recall?The added cost to clean insures in your favor that no fieldreturns will occur. In this specific case there is hope. I have seen field returnsin the past cleaned of the offending flux residues and recovered at >90%success rate. If you wish to discuss this further please do not hesitate to contactme.
Technical Expert Sales Support
Charlie Pitarys has over thirty years of industry experience and has been with KYZEN for twenty-one years. Charlie is a former Marine and a retired Sargent First Class in the Army Reserves. His previous employers include Hollis and Electrovert. Charlie continues to use his expertise on cleaning processes and machine mechanics to help KYZEN customers and partners improve their cleaning operations.
Theonly way to find out whether flux would cause functionality problems or notlater on is to assess the assembly cleanliness level or the electricalperformance before it is sent out to the field.
Climatic chamber test (i.e. Surface insulation resistance test) could be conducted first to investigate the detrimental effect of any potentialcontamination ( normally done at 85C and 85 % relative humidity).Contamination reduces the insulation resistance. Here, you could test theresistance of the unclean (no clean) board versus a cleaned board and confirmhow cleaning would improve your surface insulation resistance. At the same timeclimatic chamber test would also predict whether the no-clean board could failin the field later on if it is not cleaned.
If the assemblies fail the climatic chamber test, as a next step, the root causes need to be investigated. If suspected contaminant is ionic, then the ROSE test or an ion chromatography test should be conducted. Ion chromatography test is a more detailed oriented test method which would provide the information on the amount of cations, anions and weak organic acids potentially present on the board surface one by one.
For testing non-ionic (organic, rosin etc.) species, the most common method is Fourier-Transform Infrared (FT-IR) spectroscopy.
Application Technology Manager
Mr. Tosun has published numerous technical articles. As an active member of the SMTA and IPC organizations, Mr. Tosun has presented a variety of papers and studies on topics such as "Lead-Free Cleaning" and "Climatic Reliability".