5
\$\begingroup\$

I have to design a electronics device that is specified for environmental operating conditions of 0°C to 60°C with 5% to 95% relative humidity (non condensing).

I think the temperature part should be covered with commercial component selection (0°-70° operating temp.), but what about the humidity requirement?

  • Is this humidity specification hard to achieve?
  • Do I need to protect the PCBA with conformal coating?
  • What are the general recommendations for different humidity specifications?

Of course I will need to perform some tests in the climate chamber. But I think there should exist some general guidelines for the design regarding humidity?

\$\endgroup\$
13
  • 1
    \$\begingroup\$ What do you mean by non condensing? I don't think conformal coating would be adequate. \$\endgroup\$
    – K H
    Commented Feb 8, 2021 at 7:50
  • 3
    \$\begingroup\$ I have often seen this "non condensing" addition to relative humidity specifications. I think it means that when you perform tests in the climate chamber, you need to carefully control the climate in order not to reach the dew point. \$\endgroup\$
    – Stefan Wyss
    Commented Feb 8, 2021 at 7:54
  • 1
    \$\begingroup\$ If you have high impedance analogue and sensitive circuits, at around 0 degC, water will collect on PCB surfaces if there's humidity. So, I'd be tempted to ensure it works down to -10 degC and see what happens when you gradually warm it up through 0 degC. \$\endgroup\$
    – Andy aka
    Commented Feb 8, 2021 at 8:53
  • 1
    \$\begingroup\$ @Andyaka Thanks for your comment. I have no high impedance analogue and sensitive circuits. It is mainly a digital device. Why do you think that 0°C will be special in term of humidity? When I look at a psychrometric chart, I see no differences to other other temp. ranges. \$\endgroup\$
    – Stefan Wyss
    Commented Feb 8, 2021 at 9:13
  • 1
    \$\begingroup\$ @StefanWyss - ice that forms on PCB surfaces at below 0 degC starts to melt and becomes more conductive as temperature passes positively through 0 degC and I've had problems with that before that have been cured by varnish/lacquer. \$\endgroup\$
    – Andy aka
    Commented Feb 8, 2021 at 9:48

2 Answers 2

Reset to default
2
\$\begingroup\$

For an electronic circuit, that specification is almost completely "bog standard". You likely don't have to do anything special to meet it.

Examples of things that are "non-standard" that you might have to take into account:

  1. something that measures a field, capacitive proximity sensing, for instance, since humidity changes the permittivity of the air

  2. something that has a "diffusion interface"--lots of biochemical sensors have a layer that the active species needs to cross and that can be sensitive to humidity

  3. something that has a weird, non-standard package that doesn't hermetically seal the chip

These are unusual unless you're in a specific domain, and you generally know if you are.

Normally where people get burned is that their boards get put in a "condensing" environment even though everything is specified as "non-condensing". But that's not a design problem so much as a specifications issue.

tl;dr Don't sweat it and just do your design.

Good luck.

\$\endgroup\$
0
\$\begingroup\$

Spray application of an acrylic conformal coating would be sufficient for your needs however if you truly wanted to ensure the parts are protected and completely unaffected by its working environment, use Parylene C.

\$\endgroup\$
0

Not the answer you're looking for? Browse other questions tagged or ask your own question.