Soldermask requirements for high-CTI boards

[dramu]

Hello,

I've got a question about manufacturing boards with a high-enough Comparative Tracking Index (CTI) and solder mask applied. I think I've got an understanding on the subject, but since I basically lack both the experience and material/chemical knowledge, I would like to ask people here. Thank you for your help.

Let's assume I have a PCB with solder mask (in order to get an understanding of how manufacturing for CTI works, please don't ask if the solder mask is required). Let's also say that the PCB's FR4 laminate is made of high-CTI material. The question: If the solder mask is only normal/low CTI, will it reduce the CTI-rating of the finished board?

My understanding is that CTI (although it also depends on the material thickness) is primarily a surface material property. That is, a low-CTI coating applied to a high-CTI base will not have the same high-CTI performance as the base material itself. Also, given a high-CTI coating, due to the dependence on material thickness, to get the high-CTI rating for the whole product, the coating either 1) needs to have a high enough thickness, or 2) the base material will also need to be high-CTI. That is, it is possible to have a high-CTI product if only the solder mask is high-CTI (depending on the mask's properties), but it is never enough to make only the base laminate high-CTI but not the solder mask (assuming that there is a solder mask at all). Is my understanding correct on these?

Thanks for helping me getting a grip on the subject.

[T3sl4co1l]

It's my understanding, there is no such thing as a "high CTI FR-4".  It is <= 250 CTI, i.e. the worst possible class (e.g. IEC 60950-1 class IIIb I think?), which comes down to having an organic base that quickly carbonizes.

I'm not aware of any soldermask that is able to change that.  How would that work?  Even if it were, say, a sputtered ceramic coating, the base material would still eventually heat up and carbonize (meanwhile, cracking and flaking the coating away); it might be delayed (which IS good enough for testing purposes), but I don't see it being significantly better.

Unfortunately I don't know what materials are available with an actually high CTI.  I would search around for alternate materials.  Call up some manufacturers and see if they have any ideas.

I'm sure a ceramic PCB can be made with high CTI, but that may be a more extreme case and something with a better cost-CTI compromise may exist.  Then again, if I've not seen it, it may not even be as high volume, so that there is no lower-cost middle case... no idea.

Good luck,

Tim

[wraper]

If the solder mask is only normal/low CTI, will it reduce the CTI-rating of the finished board?

You could simply remove solder resist or even make a slot in places where high creepage distance is required (and is based on CTI). This is how it's often done in power supplies.

[dramu]

It's my understanding, there is no such thing as a "high CTI FR-4".  It is <= 250 CTI, i.e. the worst possible class (e.g. IEC 60950-1 class IIIb I think?), which comes down to having an organic base that quickly carbonizes.

Yes, I know that "high-CTI" is not specific enough. There are PLC numbers and so called material classes, representing various voltage ranges. For the question at hand it doesn't matter if we are talking about 300V or 600V CTI, the point is merely being higher than "normal" (which is about 150-200V), paired with a normal solder mask. If there is no high-CTI FR4, I'm sorry, but I think whether it is FR4 or some other type of material is really secondary. The question is, does a low-CTI coating reduce the board's total CTI performance if the "core" (whatever material it is) is high-CTI?

I'm not aware of any soldermask that is able to change that.  How would that work?  Even if it were, say, a sputtered ceramic coating, the base material would still eventually heat up and carbonize (meanwhile, cracking and flaking the coating away); it might be delayed (which IS good enough for testing purposes), but I don't see it being significantly better.

Unfortunately I don't know what materials are available with an actually high CTI.  I would search around for alternate materials.  Call up some manufacturers and see if they have any ideas.

Here is a specific example that is advertised as high CTI solder mask, but it is easy to find a lot more using Google: https://www.holderstechnology.com/wp-content/uploads/2014/11/PROBIMER_65-7203-2_Green.pdf

As for the surface coating not necessarily changing the carbonization properties of the base material, that is true, but I'm unsure if that matters as far as CTI is concerned. CTI is used for ensuring that creepage does not occur, which happens on the surface. So in my logic, if you use a high-CTI base material with a low-CTI mask, creepage could happen (almost) just as easily as if your base wasn't high-CTI at all. I feel my logic is further supported by the fact that there are high-CTI solder masks at all, or else why would they exist in the first place? But again, I'm not sure, and I'd be glad if someone who really knows and is not just trying to make sense of it like me could tell me if i'm wrong.

[dramu]

You could simply remove solder resist or even make a slot in places where high creepage distance is required (and is based on CTI). This is how it's often done in power supplies.

While that may be a practical solution in many cases (assuming there is no other good reason for a solder mask), this will not help me understand how various coatings influence total CTI. The goal of my original question is to research exactly that.

[wraper]

As for the surface coating not necessarily changing the carbonization properties of the base material, that is true, but I'm unsure if that matters as far as CTI is concerned. CTI is used for ensuring that creepage does not occur, which happens on the surface. So in my logic, if you use a high-CTI base material with a low-CTI mask, creepage could happen (almost) just as easily as if your base wasn't high-CTI at all. I feel my logic is further supported by the fact that there are high-CTI solder masks at all, or else why would they exist in the first place? But again, I'm not sure, and I'd be glad if someone who really knows and is not just trying to make sense of it like me could tell me if i'm wrong.

Does it even matter if the current path happens to be carbonized solder resist itself.

Here is a specific example that is advertised as high CTI solder mask, but it is easy to find a lot more using Google: https://www.holderstechnology.com/wp-content/uploads/2014/11/PROBIMER_65-7203-2_Green.pdf

Comparative Tracking Index (CTI) IEC 60112 600 – 0.0 V 1)
1) on CTI 400 laminate or with double coating

In this case it actually improves CTI

[floobydust]

It's just that CTI is also used as a figure of merit for an insulator.
How an insulator endures voltage stress through a PCB laminate, due to traces on either side or on inner layers, is just as important to safety and is also why high CTI is specified.
OP, by your reasoning you could use low CTI PCB laminate with low CTI soldermask- but apply a conformal-coating with high CTI rating and have no issues.

If you really want to understand how to bypass the safety standards, look at the parent IEC 60664-1 (bold added by me)

4.8   Insulating material, 4.8.1 Comparative tracking index (CTI) 
4.8.1.1
Behaviour of insulating material in the presence of scintillations
With regard to tracking, an insulating material can be roughly characterized according to the damage it suffers from the concentrated release of energy during scintillations when a surface leakage current is interrupted due to the drying-out of the contaminated surface. The following behaviour of an insulating material in the presence of scintillations can occur:
–    no decomposition of the insulating material;
–    the  wearing  away  of  insulating  material  by  the  action  of  electrical  discharges  (electrical erosion);
–    the  progressive  formation  of  conductive  paths  which  are  produced  on  the  surface  of  insulating  material  due  to  the  combined  effects  of  electric  stress and electrolytically conductive contamination on the surface (tracking).
NOTE   Tracking or erosion will occur when
–    a liquid film carrying the surface leakage current breaks, and
–    the applied voltage is sufficient to break down the small gap formed when the film breaks, and
–    the  current  is  above  a  limiting  value  which  is  necessary  to  provide  sufficient  energy  locally  to  thermally decompose the insulating material beneath the film. Deterioration increases with the time for which the current flows.

4.8.1.2
CTI values to categorize insulating materials 
A  method of classification for insulating materials according to 4.8.1.1 does not exist. The behaviour of the insulating material under various contaminant and voltages is extremely complex. Under these conditions, many materials may exhibit two or even all three of the characteristics stated. A direct correlation with the material groups of 4.8.1.3 is not practical.
However, it has been found by experience and tests that insulating materials having a higher relative performance also have approximately the same relative ranking according to the comparative tracking index (CTI). Therefore, this standard uses the CTI values to categorize insulating materials. "

[dramu]

Does it even matter if the current path happens to be carbonized solder resist itself.

It actually does, as the breakdown voltage is usually a lot higher than the tracking index. So there is actually a wide voltage span where a surface can be susceptible to tracking, but otherwise it wouldn't be possible for it to break down. Meaning inside this voltage span, it is impossible for the solder resist to carbonize due to electrical stress, except if caused by tracking.

Comparative Tracking Index (CTI) IEC 60112 600 – 0.0 V 1)
1) on CTI 400 laminate or with double coating

In this case it actually improves CTI

Which says the same thing that I did in the first post: If the solder resist is applied in high thickness (here, double coating), the base underneath it need *not* be high-CTI. For single layer coating however, the manufacturer of the mask requires a high-CTI base in order to guarantee the CTI performance of the mask.

[dramu]

How an insulator endures voltage stress through a PCB laminate, due to traces on either side or on inner layers, is just as important to safety and is also why high CTI is specified.
OP, by your reasoning you could use low CTI PCB laminate with low CTI soldermask- but apply a conformal-coating with high CTI rating and have no issues.

Yes, that is kind of my reasoning actually, but of course only as long as *only* CTI is concerned. Note I am not looking to replace strong insulators with only a high-CTI coating. My question relates solely and only to preventing creepage, while at the same time knowing well that for full and proper insulation, other aspects beside creepage must also be addressed.

Also, please do not misunderstand me: I am not the slightest trying to "bypass" (as you put it) safety standards, actually quite the contrary. I have a colleague who argues that a solder mask coated PCB does not need a high-CTI solder mask if the base laminate is already rated for high CTI. Contrary to that, I am arguing that as long the solder resist is present, it must also be high-CTI. But I am not 100% sure so I decided to ask others too.

In your post, it was new info for me that for insulators applied to a PCB, the CTI is also used for roughly characterizing its breakdown, not just its tracking performance. While I didn't know that, doesn't that actually support my original suspicion? It would mean if a low-CTI coating is applied to a high-CTI board, then the outer layers of the finished product (the cured coating) can still easily break down and/or cause tracking, which negates the benefit of the base being high-CTI and a good insulator.

[T3sl4co1l]

Here is a specific example that is advertised as high CTI solder mask, but it is easy to find a lot more using Google: https://www.holderstechnology.com/wp-content/uploads/2014/11/PROBIMER_65-7203-2_Green.pdf

Ah, cool.  So such materials definitely exist!

I wonder what the CTI=400 substrate was...

As for the surface coating not necessarily changing the carbonization properties of the base material, that is true, but I'm unsure if that matters as far as CTI is concerned. CTI is used for ensuring that creepage does not occur, which happens on the surface. So in my logic, if you use a high-CTI base material with a low-CTI mask, creepage could happen (almost) just as easily as if your base wasn't high-CTI at all. I feel my logic is further supported by the fact that there are high-CTI solder masks at all, or else why would they exist in the first place? But again, I'm not sure, and I'd be glad if someone who really knows and is not just trying to make sense of it like me could tell me if i'm wrong.

Yeah, it should be a surface phenomenon, at least at first, and you'll win a higher CTI the longer it holds off destruction of itself and the substrate.  So I can see a good coating improving a substrate (if not aggressively so!).

Same here, I haven't looked into this myself (can you guess? ) and am interested to see what's out there.

Tim

[ahbushnell]

Rodgers makes hi CTI material.  I have used it in high voltage applications.  Note CTI is important in a dirty wet world.  And don't use FR4 for high votlage.

https://www.rogerscorp.com/index.aspx

[floobydust]

What I find obtaining product certifications from UL/CSA is that they look at the PCB laminate specifications.
The solder mask (and silkscreen) is considered a paint/coating and not the main insulator. It might be something to do with the surface grain, but probably because more important is the PCB environment, as far as dirt and moisture.

It's very difficult to find soldermask/silkscreen with high-voltage ratings like CTI or PLC (performance level category) because they are tested on what surface? You can find a dielectric rating but nothing for tracking.

High-CTI (>275-400) PCB laminate can be found, in low cost it is common in appliance "dishwasher" applications.
Exotic, expensive (microwave) laminates can be found with high CTI as well.