Can conformal coating affect performance?

[fx991ex]

So I have this very sensitive fully analog board - an EMG amplifier.

After I covered the board with urethane conformal coating the sensitivity of the amplifier plummeted. The amplifier still works but the signals that it outputs are much weaker.

I used the following conformal coating: https://uk.farnell.com/mg-chemicals/4223-55ml/chemical-coating-55ml/dp/2889756

The coating is fully died. It has been more than 24 hours since I coated the board.

[EEEnthusiast]

If some of your pins of the ICs are too fine pitch (0.4 mm ) then the conformal coating can add additional capacitance between them. They will also add additional resistance, but once it dries up, the impact of resistance would be negligible.
What kind of circuits is it? If it is very low frequency, then there should be minimal impact to the design.

[fx991ex]

Yeah, there are SOT23-8, MSOP-8 and 0402 components on the board. And the board is pretty densely populated. And the frequency is fairly high.

[EEEnthusiast]

For high frequency circuits, conformal coatings are a bit tricky. The extra capacitance created by the coating layer dielectric, will cause frequency response changes. Circuits could oscillate and sometimes may loose bandwidth. There are specific chemicals made for RF boards, but I would still not use them without caution.

[fx991ex]

Can you recommend me any product?

I intend to coat the board with conformal coating and then encapsulate it in polyurethane resin.

[coppice]

Conformal coating inevitably affects capacitance, but it will also affect resistive leakage if you don't apply it carefully. You need to have a really clean board, thoroughly dried, before applying the coating. If you don't you will trap moisture under the coating, which has a nasty tendency to be a lot more leaky than the surface moisture was before coating.

[daqq]

There can actually be some leakage induced by conformal coating, compared to, say, perfect air, as well as additional capacitance. That could certainly knock a very sensitive circuit off balance.

Also, it's still permeable a bit to water (see: https://www.analog.com/en/analog-dialogue/articles/does-my-voltage-reference-design-hold-water.html?ADICID=EMAL_WW_P1274_SUB-NL-PN_547&9039 ).

[EEEnthusiast]

Trapped moisture is more deadly as it has no way to evaporate inside the coating. Without the coating, as the board warms up, the moisture evaporates and the circuit starts to perform better. Recommendation is to bake the board before applying the coating. This will ensure moisture is already gone. I cannot recommend a coating for high frequency as I never use them on RF or high frequency boards. To protect RF boards, I use a metal shield over the sensitive parts. Then seal any holes or edges with epoxy. After this is done, a fine coating of general PU spray could be applied over the entire board. This ensures that the RF or high frequency part of the board remains unaffected by the coating.

[Tomorokoshi]

It appears that urethanes have dielectric constant between around 8 and 11:
https://gallaghercorp.com/white-papers/electrical-properties-urethane/

While acrylics have a dielectric constant around 2.6:
https://chasecorp.com/HumiSeal/products/acrylics/1b73/

It may be that an acrylic coating like HumiSeal 1B73 could reduce some of the capacitive effect. Not sure about moisture permeability or other factors. You might need to make several test boards to measure the properties. As mentioned above a cover of some type with a gasket might be a better option.

[fx991ex]

Thank you all for the answers and the resources. That article from Analog Devices was especially interesting.

I willl contact some different manufacturers of conformal coatings and see what they say.

I went and bought PLASTIK 70 acrylic conformal coating from Kontakt Chemie and coated another board with it (same board, another instance).

The acrylic conformal coating seems to not affect the board's performance however I have some concerns regarding its effectiveness. For example If I touch an operational amplifier on the board, the output signal will spike even though the board is conformaly coated. Maybe the coating that I sprayed on is too thin?

As I mention before conformal coating is just a step prior to potting the board in polyurethane resin.

[Tomorokoshi]

In some applications using a spray-on conformal coating requires two coats. If the board is user-accessable you may need 2 or 3.

[fx991ex]

I will certainly try that. Although I really should wait 24 hours for the acrylics to dry before I make a judgement.

[fx991ex]

Yeah, acrylic conformal coating won't do anything if I pot the board afterwards in urethane.

I found this potting resin: https://electrolube.com/product/er1448-specialist-radio-frequency-encapsulant/

Any thoughts? It seems to be what I am looking for.