0
\$\begingroup\$

I am new to circuit design and debugging them. I have recently helped in building a PCB with certain kind of circuits on it which includes opamp, boost converter, microcontroller and accelerometer. The board is battery operated. It basically goes on a device which is supposed to be water proofed. But by any reason the water can go inside the device.

The circuit is works fine it draws minimal sleep current in micro amps when in sleep mode and set current when its measuring. But when the liquid(approximately a drop of liquid and it would have already dried up "the liquid is mostly sugary") gets into the board "the board still works it even measures" but it starts drawing more current. Hence the battery gets drained faster.

My question is Can I debug the boards and bring it back to its normal working condition? If so how to debug it? I have 4-5 boards which are in the same condition will they all have same problem if I debug 1 board?

It would be really helpful if I could get some help from this forum.

Thanks

\$\endgroup\$
4
  • 2
    \$\begingroup\$ Distilled water is cheap in most places, and very easy to make where it is not. Wash the board with it and move on. \$\endgroup\$ Jul 31 '15 at 18:27
  • \$\begingroup\$ I agree with ignacio, wash the boards and then conformal coat them. \$\endgroup\$
    – Kvegaoro
    Jul 31 '15 at 18:29
  • \$\begingroup\$ Thanks.I washed the board with Isopropyl alcohol. But its still drawing more current. Is it so that I need to use only distilled water? \$\endgroup\$
    – User999
    Jul 31 '15 at 18:36
  • 1
    \$\begingroup\$ Depending on the stuff you might want to use both liquids to clean the board, some things will solve in water and get washed away and others will solve in isopropyl alcohol. Don't use too strong solvents as they might damage your components/pcb. After cleaning make sure it has time to dry off. \$\endgroup\$
    – Arsenal
    Jul 31 '15 at 19:22
1
\$\begingroup\$

First remove the battery if it's attached to this board. Bake them for about 30 minutes at around 40C-50C (maybe check data sheets for temperature ratings but this should be okay). This should remove any moisture. Let them cool down and then perform your current consumption and functional test again.

In the future, conformal coating should help if they aren't coated. Also, maybe seal up the enclosure better with some epoxy or silicone. You could also if there's a place/room add or zip-tie a moisture absorbing packet (silica gel) to the board. Water/moisture usually finds a way even if you don't think it's possible.

\$\endgroup\$
4
  • \$\begingroup\$ If water has penetrated the packages themselves then you will need to bake for longer. See the manufacturer recommendations regarding pre-reflow baking. \$\endgroup\$ Jul 31 '15 at 19:03
  • \$\begingroup\$ Yeah, you may need to bake for longer. Use your best judgement and discretion based on the parts. \$\endgroup\$ Jul 31 '15 at 19:07
  • \$\begingroup\$ Thank you everybody. I will tryout this technique. Apparently we checked the boards and they have short. So when we finally digged in the short was found on the microcontroller which is BGA package. So seems that the liquid seeped under the BGA? And it didnt dry out. But hopefully if I bake it I can get rid of this problem too am I correct? \$\endgroup\$
    – User999
    Jul 31 '15 at 22:31
  • \$\begingroup\$ If you operate electronics in high humidity environments you can also develop conductive paths on the surface of the board (electrochemical migration), and even inside the board, that are difficult to get rid of. So depending on how long they were operated with moisture, you may have developed some irreversible conductive paths. \$\endgroup\$
    – dbrwn
    Aug 8 '15 at 19:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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