Ask the Experts
August 3, 2018
Glued SMT Components Falling Off
We are having issues with glued SMD capacitors falling off after going through the oven curing process. We are using well known adhesive from a quality supplier.
The issue is erratic and we have no problem with SMD resistors. Why does this problem happen only with capacitors and not resistors. The components are very similar in size?
Expert Panel Responses
This could be afunction of your glue placement....
Cap's 'pop' off theboards generally because of voids in glue. The glue is semi cured, andvolatiles trapped mechanically pop. You mentioned this is happening inthe curing oven, not the wave... which is another similar soldering issue.
The erratic patternyou've described should be tracked. Tantalum Capacitors, and"similar" package sizes of Ceramic Capacitors can have differentcomposition. When soldering, we found, that testing could not be doneuntil 12hrs had passed on some packages, due to heat retention altering thevalue of capacitance. This consideration, might be the difference in your"similar" packages, and could be coupled with glue voids, which arepushing your parts from the assembly.
To minimize this, youmay look at alternative method of glue... minimizing the profile to leastpossible compression of glue dot. Less is more, when it comes to SMDglue. For this reason, we used a stencil printer to improve performanceof our SMD / Type II assemblies. The profile allowed us to reduceprofiles, create different shapes of glue patterns and eliminate the 'popping'in the wave soldering process. I'm sure this would apply in yourglue/curing process as well.
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.
First, some key assumptions:
If the answers to the above questions are both "Yes" then thekey to solving this is determining whether the failure was adhesive (failed atan interface) or cohesive (failed within the adhesive) and if adhesive, atwhich interface the failure is occurring.
If the failure is adhesive in nature and the adhesive remains onthe board, look at the following potential root causes:
- There was in fact sufficient adhesive present.
- You've done a thermal profile of the curing process and verifiedthat complete cure is occurring.
If the failure is adhesive in nature and the adhesive remains onthe component, look at the surface condition of the soldermask as a root cause:
- Compare the bond strength on the capacitors to that of theresistors. Test to see if they are well-adhered using a push test. It may wellbe that the resistors also have low bond strength but enough to hold them on(they tend to be lower-profile and have lower mass, and they may have arougher, more porous bottom surface).
* If bond strength is low on both components, look at the delaybetween adhesive dispense (or print) and placement, and look at the adhesivelot-to-lot variation storage conditions and shelf life.
* If the bond strength of the resistors is normal, while the bondstrength of the capacitors is very low, then look at the surface of thecapacitors for contamination that may be inhibiting adhesion.
If the failure is cohesive in nature, look at the following:
- Contamination on board? Test by cleaning PWB prior to assemblyand comparing bond strength.
- Soldermask material issue? Some soldermasks may containmaterials that cause low bond strength if present on the surface. Thesematerials include fillers that may be used to modify surface appearance or toreduce cost.
By carefully examiningthe failure mode and testing/eliminating possible causes in a structured,quantitative way, you should be able to rapidly diagnose the source of theproblem.
- Curing problems, including excessive bubbles within adhesivewhich are indicative of improper curing profile, or softness of the "cured"material when probed, indicating inadequate cure.
- Mechanical impact. The fact that the caps are physically tallerthan the resistors can lead to them being more susceptible to being knockedoff. If this is the case, the bond strength of remaining components will benormal, and there may be evidence of the impact in the area of the missingcomponents.
Fritz's career in electronics manufacturing has included diverse engineering roles including PWB fabrication, thick film print & fire, SMT and wave/selective solder process engineering, and electronics materials development and marketing. Fritz's educational background is in mechanical engineering with an emphasis on materials science. Design of Experiments (DoE) techniques have been an area of independent study. Fritz has published over a dozen papers at various industry conferences.
Suggest you check thethickness of the stencil and the stand-off distance of the Capacitors vs. theresistors. Maybe the Caps have bigger stand-off.
Vice President Technology
For over 18 years, Dr. Coleman has been the vice president of technology for Photo Stencil, working closely with customers to understand their printing requirements. His efforts have resulted in several new stencil products.
It would be nice to know if the glue is stilladhering to the resist on the board or if it is only sticking to the componentbody. If it sticks to the resist then could it be down to a Mold Releasecontamination of the Component. If stuck to the component then could be due tosurface contamination or more likely under-cured solder resist leaching out andnot allowing the adhesive to bind properly. This would be shown worse inheavily grounded areas of the PCB.
Technical Sales Manager
BLT Circuit Services Ltd
Greg York has over thirty two years of service in Electronics industry. York has installed over 600 Lead Free Lines in Europe with Solder and flux systems as well as Technical Support on SMT lines and trouble shooting.
I would be suspicious of the capacitors' surfaceproperties, i.e. contamination. Has your vendor changed ANYTHING in hismanufacturing process? Has his supplier of raw materials used in thecap manufacturing changed anything? The surface of the caps has changedwhich is why they do not adhere to the adhesive, if, in fact, there have beenno changes of any kind in the adhesive.
Jim Willimas is a PhD Chemist in Polymers and Materials Science. He specialize in printing, cleaning, inks, and coatings used in electronics manufacturng operations. Williams has more than 30 years experience.
Thereare a number of reasons why components might fall off, but if this is happeningafter cure but not with any outside force applied such as wave soldering thenthe reason is most likely one of the following.
The adhesive is either not biting into the surface of theseparticular capacitors due to a smoother finish for some reason and this issometimes seen on glass diodes or the supplier has used a mould release agentin the manufacture of the component and this is causing the adhesive notto adhere properly and this causes the component to fail.
Ifthe components are falling off in a wave solder process then there are a hugenumber of other reasons that might be the cause.
Global Product Champion
Richard Boyle is a Global Product Champion at Henkel Electronics. He has over 25 years experience in the electronics assembly industry and is responsible for the global technical service of all of Henkel's solder materials.
Firstcheck the cleaning process. There could be mold release left on thecapacitors.
Wayne Wagner has over 25 years in the conformal coating industry and is the president of Krayden Inc., a leading distributor of engineered materials.
The issue is all of the above, plus a few more factors, including a combination of the material properties (CTE) of the capacitors, the CTE of the PWB, and the CTE of the chipbonder you are using. Cleanliness also plays a role.
Most chipbonder adhesives are very hard materials when cured, with a Shore D hardness in the range of 80 or higher. They have little modulus of elasticity. If there is a major difference between the CTE of the board and the component, in this case the leadless caps, then the stress is placed on the chipbonder. The typical chipbonder such as Loctite 3609 or 3621 does not adhere well, it will work just fine to hold the parts long enough for reflow to take place, but if the ramp rate is too fast, the chipbonder will snap off due to the major difference in expansion rates between the FR-4 and the ceramic chip caps.
Try slowing down the ramp rate slightly and see if the issue is reduced. In cases like this, it is usually not any one issue, but a perfect storm of the adhesive, the cleanliness of the boards and parts, the reflow ramp rate, and perhaps other factors as well.
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.