Conformal Coating Defects: When Things Go Horribly Wrong - Banner

Conformal Coating Defects: When Things Go Horribly Wrong

To consistently and repeatably conformal coat PCB's, all the variables need to be controlled like heat, humidity, time, contamination... and the list goes on and on. Failure to control your process can result is sudden defects and an assembly line that grinds to a halt. Learn from Pierce Pillon, Senior Field Engineer, and Kevin Pawlowski, Applications Specialist, on a discussion of case studies on coating processes that have gone off the rails, and how the problems can be diagnosed, solved, and (most importantly) prevented.
Watch the video or download a Comprehensive Conformal Coatings Defects Guide on PDF.

This webinar's not meant to be an all inclusive, in depth discussion of the conditions that may be deemed as a coating defect, but rather a broad overview, intended just to familiarize you with establishing and refining your coating program, by recognizing these conditions. First, you have to define what a defect is, and this can be determined in three broad manners. It can be determined by internal documents that your company generates through your engineering staff, your operation staff. It can also be customer driven, whether it's an internal customer, such as, for instance, let's say you have a company with many divisions and your area has been tasked to provide a part for another division of your company, that's an internal customer. Also, it can be customer driven by your outside customers. Contract manufacturers know exactly what this is. Those are determined by contract and the documents that are referenced and spelled out in those contracts.

They can also be determined by adoption, either in full or part of industry standards. And two of them that are widely used out there in the manufacturing field are IPC A-610 revision G, that's the acceptability of electronic assemblies, and also the IPC J standard 001. I believe it's revision F is what we're on now. And that's the requirement for soldered electrical in electronic assemblies. When you establish your criteria, there are three main categories of criteria. The first is your target properties of your coating. Essentially, this is a wishlist of what the perfect coating on your assembly is going to do, okay, what properties that has to show for acceptance, that is your wishlist. The next is actually your accepted criteria. They may not meet your target, but this is what is considered acceptable. And from there, you can further define the criteria for deeming it a defect and of course, in there there's inspection criteria, the methodologies you use, and so forth and so on, in your quality process.

So with those three main categories, your target properties, your acceptance criteria, and your defect criteria, you've got those established by whatever means you're going to use. We're going to use IPC A-610 as our prime set of examples throughout this presentation. And a defect is a condition and let's be quite plain in what a condition is, in this instance. It's a condition that does not allow, and I'm paraphrasing, the form, fit, and function of that assembly in its service environment or process indicator on the other hand, is a condition, it is not a defect. It's a condition that points out a characteristic, but it does not affect the form, fit, or function of that product. And it's important to understand that a process indicator neither fully meets acceptance criteria, yet it's not a defect. And I'll give you a really good example, and we'll see this later on.

Let's just say bubbles and voids are considered as conditions that may be a defect, okay. In one area of the board, then that might meet your criteria of a defect, but let's say you have a bubble or a void over on a totally empty non populated section of the board, is it a bubble or a void? Sure it is. But is it affecting the form, fit, and function of that product? If it doesn't, then it's just a process indicator. Is there something wrong? Yes. Your process is out of control in some manner and you should address it and you should resolve it. But based on this criteria, it should not be considered as a defect. That's the difference between a defect and a process indicator. As we said, we're using A-610 as our source document for this, and these are the criteria that are spelled out in A-610.

Your criteria may be different, but in A-610, this is the criteria that they use. If a coating is not cured, well, that's kind of common sense, if a coating is not cured, it's not functional. It's not providing the projection that you're putting in on there to provide. An example of this is silicone coatings. Mini silicone coatings are RTB coatings, polymers, a room temperature vulcanization, and as such, they require a minimum amount of moisture to initiate that cure mechanism. So let's say you're in Phoenix, in the middle of August, and that relative humidity in the facility is essentially nil, you may get incomplete or just partial curing on your coating, that can be resolved. But at that point, that condition, according to A-610, is considered as a defect. That's just one example of coatings not being cured.

The next is coating is not applied to the required areas. Now those required areas generally are spelled out in your engineering drawings for that assembly. And it's pretty easy to compare that drawing with the actual assembly and look and see during your quality inspection, if all areas that require the coating are coated. Now, the flip side of that is you have coatings on areas that are required to be free of coating. These are your keep out areas. For whatever reason, those are deemed by the engineering drawings that they cannot have coating in that area, for whatever reason, by the designs. An example that was applied, and may have been whipped into a connector housing because it wasn't properly masked or booted, and now you have insulated coating around conductive pins or conductive female housing in the other half of that part. That's just an example.

If there's any bridging of adjacent pads, or the lands, or any exposed conductive surfaces that are caused by loss of adhesion, voids, bubbles, de wetting cracks, fish eyes, flaking, any entrapped material, thawed, that's entrapped in that coating that bridges the lands or the conductive surfaces adjacent conductive surfaces, if the circuitry is exposed, if it's less than the minimum clearance between the components, lands, or conductive surfaces, that's considered a defect.

Also, discoloration or loss of transparency. If a coating has been discolored by the environmental conditions, its service environment, for instance, urethanes over time, if they're exposed to sun, have a tendency to yellow. If it hazes in any way, it's considered at least a condition to consider deeming it a defect. And the reason this is important, is because the inspectors, the end users, if it's set up for rework, they have to be able to see through the coating for labels, markings, anything that they wish to inspect. The main causes for defects, I've decided to bring these conditions into three main categories, on where they may be looked at as part of the cause. And as we go through this and we hit all of these different conditions that may be considered as defects, you'll see these little symbols up on the top right hand side of the slide.

And that gives you areas to look at, to address that condition and find out a root cause to resolve it, okay. Cleaning or perhaps lack of cleaning, that's one big category where that can be ripe for creating these conditions that may be defects. And you can isolate that, you can take a new board or a new assembly, coat it, and see if it corrects that defect. And then of course you will want to validate that by additional testing. It could be an application error, an equipment setting, whether it's an improper setting or whether something has drifted, wet film thickness, if it's too little, too much, that can cause that condition to possibly be a defect, viscosity changes, variability, and by variability, I don't necessarily mean in equipment it's human variability, the human error that's introduced into your process by inconsistency of operator application, whether it's from one application to another using the same operator, or the same application using multiple operators over a shift, you're going to get some inconsistency.

And sometimes this inconsistency may result in a condition that could be a defect. And then of course you have the curing process, the time, temperature, in some cases, the humidity conditions in which you're curing that coating. If you're not doing it properly, if you're not following the the coating manufacturer's recommended guidelines, that could set you up for a coating failure or a defect. The coating must uniformly cover that board and all of the components. If it does not, it could be a condition that may be a defect. Now, I'm not sure how many of you know what shadowing is. Shadowing is caused by the geometry of the board, where you have a short component right next to a tall component. And your method of application may prevent that shorter component from being uniformly coated, just because it's being blocked by the taller, that's what's considered shadowing.

If you have any uneven coating application, that could be cause for concern. Brushing is an example of this. When you're using a brush to apply the coating, you're generally going from the beginning of your pass, to the ending of your pass on that brush stroke, you're going to apply more coating at the beginning of the pass and less coating at the end of the pass. There's also coating sag. And you'll see this in dipping. Now I do have to point out that this is specifically addressed in A-610, it is allowed, but as the coating settles from a dipping process, it accumulates at the lowest point. Could it be a defect? Possibly, but it is allowed, as long as it does not affect the form, fit, or function of that component or that part in its service environment, the end use environment, then it's not considered a defect, at least according to A-610. Your criteria may be different. If you have sharp surfaces, solder points, when coating is applied over these, they tend to be less consistent there than they are on a flat surface of the component or on the board laminate.

A solder point, essentially, is an upside down cone. When it's applied over the top, it tends to fall by gravity down to the base. And so you generally have a little bit less cutting coverage at the tip of that solder point. That may be concern, and it may be cause to consider as meeting the conditions of your defect criteria. You can have differences in the surface tension of the coating liquid, as opposed to the substrate energy, and that can cause non uniformity. In general, and this is in general, the liquid coating surface tension must be less than the substrate energy to allow it to flow for adequate wetting, and adhesion, and bonding energy to that substrate. De wetting is essentially where the coating pulls away from that surface because of differences between the surface energy. It could be a contaminant on there, as mentioned here, it could be little spots of silicone levels for mold releases, adhesive residues.

It could be oils from your hands and your fingers from handling the board. If you're not properly taking those precautions, again, surface tension and surface energy variations that coating allow or is not allowed to properly wet, and then bond to that surface. You can have some interaction with flux residues because of the binders, the Realogy modifiers, the ingredients, let's say that's leftover after the soldering process in these residues, sometimes do not allow proper wetting of the coating. And again, interaction with the coating on the surface, just because there are just differences between the surface tension of the liquid and the surface energy of the surface, whether it's a component casing, whether it's a board laminate and you see this in the bottom right hand picture, that obviously that coating is not leveled out and adhering, and wetting that entire surface.

Fish eyes, by and large, this is a point contaminant issue. And when I say a point contaminant, it could be a little spot of silicone, a little spot of oil, wax, possibly some unremoved or flight coating, if you've got a rework area going on. What that does, is as that coating goes over that little point contaminant, it raises up and it repels from that contaminant and it forms this it's almost like a bubble, it's actually a bulge, if you will. Another source, I mean, it could be bits of scoring dust. It could be junk coming in from your ventilation system, could be a piece of lint off of your smock. Anything that can settle on that surface, prior to coating, can be considered as a point contaminant or a potential point contaminant. And I do want to mention here one thing that I didn't put on the slide, but I've seen this quite a few times.

If you're using temporary solder mask in your process to protect from areas that you don't want soldered, let's say if it's a peelable mask, once the soldering process is completed and you peel that mask off, there could be a little spot of residue on there that you don't see with the eye. That can be a point contaminant. If you're using water washable temporary solder masks, those are generally composed of either clay based or cellulose based that are in an adhesive binder, a water-soluble adhesive binder. When it goes to the cleaning process, it's washed off, the binders dissolve, releases either the clay or the cellulose, and then it goes where it goes down the drain. But if you don't adequately clean it, now you've got little spots of clay, and these are very small, very small particulate size of the clay or the cellulose.

Those could be point contaminants. If you're cleaning an area, especially rework area, don't use the same thinner that you're using for your coating, if you have like a selected spray system or a spray gun, because sometimes that is not the proper solvent to clean those residues. And what that ends up doing is now you just smear that residue around that area. It just doesn't properly dissolve and remove a lot of these contaminants, so please be aware. Bubbles are essentially caused by, in the most part, by solvent that is flashed off. Now it is vaporized and is trying to escape that coating, and it can't. So it bulges that coating. What happens is, is that coating skins over faster than the solvent can flash, and is trapped. It will slowly diffuse out. But now that coating is cured just enough to maintain that bubble shape, and it can create a void.

Is it a defect? Depends on your criteria, according to 610-A, maybe. If it satisfies that that bridging criteria, then yes. But let's say it's way out on the edge of the board somewhere, it's in a non populated, just a blank area, it's uninhabited by components. You may have a bubble. Is it a defect? Well, it depends. If it does not fit that criteria, if it's not bridging lands, adjacent components, whatever criteria you have set up, then it's a process indicator. Something has gone wrong in your actual applications of process, and it should be addressed and it should be resolved, but it's not a defect. Now you'll see this phenomenon a lot at the base of leg components, where you have applied the coating over the top, around the adjacent areas. And some of that coating is now in that under fill area, in that standoff area, underneath the component.

So you're going to apply thermal heat or a thermal cure process, while obviously being exposed to the heat around the edges there, that coating's going to cure at a quicker rate than the coating in that standoff area, below the connector, or not the connector, but the component, excuse me. So now you're starting to cure and skin over that coating, at the base, and around the edges of that component. And now the solvents underneath are starting to flash. They're picking up that heat, well they've got to go somewhere and they want to go out from underneath that component. And they hit that partially cured or initially cured coating, and because of the vapor pressure, it causes that coating to bulge and create a bubble. Now, if you're brushing, when you dip your brush in the coating, it can trap bubbles between those bristles.

And then when you make your pass with your brush, it releases that coating with the entrained air bubbles in there. And sometimes those bubbles will dissipate as the coating levels out, and sometimes they become trapped. So that's a condition to warrant further inspection. If you're using a spray gun, improper setup, or your settings have drifted, they can trap air bubbles. If your travel pass across that board or across its assembly is too slow, it can entrain air as it hits the surface of the board, possibly air that's underneath that component that we just talked about. If the distance between your nozzle and your substrate is too close, now you've introduced some turbulence in that coating, as that higher pressure coating hits the surface, and that can trap bubbles. If your air pressure, and by the air pressure, I'm talking about your push pressure, either in your selective system, which you generally don't see that in a selective system, but it can happen.

It's more obvious in a spray gun. If that's too low, it can entrain bubbles. If you have excessive coating thickness, if your wet film is way too thick, and especially if it's a high solid content coating, it doesn't cure evenly. It doesn't allow that that solvent to evaporate evenly, and especially at the surfaces. And then you get strains and stresses in that coating, and it cracks. If you're force drying, if you're curing at too high of a heat, it can create cracks. The bottom left hand picture, there's what we call mud cracks. It's also evident in the bottom right hand picture around the component labeled as 105. You can see the crack across just outside the top of the component, on the top edge in the picture. That's just caused by uneven curing, and stresses and strains within that coating, as it tries to cure. And it just cracks.

Delamination is not common, but it does occur. And the bad part is, is you're generally not going to see this in your manufacturing facility, by your inspection team. One of the causes is, again, excessive cutting thickness. That interface of the coating at the substrate is improperly cured, or it does not cure at the same rate as the rest of the volume of that coating. And it does not for good adhesion with the substrate surface, whether it's a component or the laminate on the board, whatever it may be. Various surface contaminants can prevent good bonding. You have little silicone on there. Maybe you've got a part that's a peel off, that stuck on there. Any of those can prevent good bonding. If the substrate energy is too low, number one, it's going to feel very slick just on a bare board, it's going to be extremely slick.

If it's lower than the surface tension of that coating, it can cause delamination. Now what it's going to look like, is it's going to look good, it's going to look like it's bonded to that surface, but six months down the line, it's going to pop up. That adhesion is going to fail. Sometimes they just flake. You can see up at the top, in the top picture, you can see areas where that coating has just kind of flaked off. In the bottom, they're flaking, but it looks like almost the entire coating has pulled up from the substrate. Now, if you're doing multiple coats, use thin coats, not heavy wet coats. Multiple thin coats is a much better adhesion practice than heavier coats. They just don't tend to meld together, okay. And we always recommend on the previous coat, allow it to become tacky. Don't overcoat it while it's still essentially wet. You're running the risk of those coatings not adhering to each other, actually as color cohesion. Orange peel is a very rough, very uneven texture.

You can see that actually, you can see the texture and the picture of the orange peel, but you can see it in that bottom picture on that component. Sometimes excessive film thickness and using heavy wet overcoats will cause this. Sometimes if you have a wet coat and you're applying another coating on top of it, that pressure from an aerosol, from a spray gun, can kind of push it. And as that bottom picture, you can see, the coating on the left hand side looks thinner and it looks like it's been waved up over to the far right hand side. And you can see the peaks, and the valleys, and that uneven texture there. If your substrate is too warm relative to the temperature of the coating, it can cause this. Let's say you just cleaned your board. It's gone through the dry cycle in your wash system.

You take it out and it's still warm, but you're in a hurry, and your operators need to cut it. Well, now they're applying this coating. And what happens, is that as soon as it hits that warm substrate, there's such a Delta there, that that solvent starts flashing, and it's flashing unevenly. And this is going on as you're depositing additional coating on top of that. And so the solvent is starting to flash in a random manner. And they form these bulges, and they form that really uneven rough peak and valley type texture. If the solvent evaporation rate is improperly staged, it won't level. And by that, what I mean is, when a coating manufacturer is developing the coatings, they pick solvents, and they pick the amounts of solvents, where it allows for a staged evaporation rate. And most formulators will include what they call a tailing solvent.

It is generally the smallest amount of solvent, yet it is the slowest evaporating solvent. And that allows that coating to remain wet in the final stages of curing, to allow it to level out to that nice, smooth, even finish that you've come to expect. But if it's not staged properly, if it just flashes out too quickly, you're going to end up with a lumpy mess and or orange peel. If you're spraying, especially spray guns, again, pressure too low, it can cause this. Again, if you're spraying, your passes across that board must be parallel to each other, and at the same distance, with a slight overlap, and they have to be perfectly perpendicular to the substrate. If you're spraying at an extreme angle, you're putting pressure behind that coating. And again, it can push that coating at the point of contact. You can push it away from that point of contact on your previous coat.

And it can result in orange peel. It can also result in other conditions for potential defects, wrinkles and waves. There's a couple of causes of this. One of them is that if you're curing in an oven, if you're curing in an IR conveyor system, that thermal cure energy is too high, and that tailing solvent flashes too quickly. In other words, all of your lighter insolvents are flashing, but your tailing solvent is flashing at the same time. It's just too much heat for it to handle. It needs to be done in a staged manner. And you're going to end up with a wrinkly lumpy mass. If your previous coat is still wet and you're coating over that, instead of letting it come to a tacky condition first, it can create that wave effect, like we discussed before. It's just pushing that coating away from that point of contact.

And it looks like waves lapping over one another. Okay, here's a different take on that same chart that we had earlier. And now what we've done, we've kept the main three categories of generic causes here. But what we've done is, we have included these conditions that we've gone over and we've categorized them according to a cleaning issue, an application issue, or a curing issue. Now they're not mutually exclusive. If you can see delamination can be caused by a cleaning problem, it can also be caused by a curing problem, but hopefully this has given you at least some areas to investigate, so you can resolve that problem. Cleaning your PCB before coating, well, here's the deal, coating loves clean. The cleaner it is, the better chance you're going to have of good bonding energy, good adhesion, good coverage, and less potential defects.

And non cleaning or improper cleaning or lack of cleaning, these are common causes like fish eyes, de wetting, delamination, flux residues. If they're not clean, they can absorb moisture or cause corrosion, even under the coating. So we recommend that your assemblies be clean before your coating process. And this is including no clean fluxes. I'm a little worried about no clean fluxes, no clean doesn't mean you shouldn't clean. And the reason that no clean fluxes are called no clean, is because the residues do not, according to the manufacturers, contain any ionic ingredients, or any ionic portions of that residue that can promote indritic growth. However, no cleans as well as a lot of other fluxes can cause reactions with the coating, cause adhesion issues. There have been some instances where these are routed back to the use of no clean fluxes. It is not every no-clean flux. It is not every coating. It's just certain combinations.

People coat over no clean every day, but there that potential exists of having a coating and a no clean flux residue that are incompatible with each other. So we always recommend cleaning before you coat. And you can avoid these coating failures due to surface contamination, such as flux residues, release agents, silicone contamination, adhesive residues, scouring dust, junk from your ventilation system, whatever it may be, anything that settles on there, or is on there due to the assembly process, the soldering process, anything like that, those are all potentials to cause problems in your coating process. So please avoid those, clean, and you'll have a whole lot less agony in your life.

Please reach out to us with questions, problems. Let's say you're qualifying a new product, a new process, you're running into problems, okay. Our job is to make your job easier, okay. So please reach out to us.

Thank you everyone.

Ask A Technical Question


Stay up-to-date on Chemtronics news, products, videos & more.

Related Categories
Previous Article Next Article
You did not finish submitting your information to request a sample