Production PCB assembly and moisture issues question

[alexg]

Hello, I did some homework on this topic already but for clarity I wanted to ask few specific questions, perhaps someone with greater experience can help me out to understand who is doing what wrong and how to remedy that.
We have batch of boards FR170, no visible issues, hand assembles two as a test hot air reflown and no issues, however when we received batch of same boards from US based assembly house, some boards have issues with moisture under ICs, to the point where it alters Voltage regulator's performance as well as main IC performance, at first I did not know what is going on until I accidentally traced issue to moisture under IC when I was reflowing it and it popped, if you apply generous amount of flux you can even see at you heat it up some bubbles seeping out from under IC for awhile. I also took one of the problem boards, soaked few ICs with denaturated alcohol then dried it slowly with FAN and it seem to perform better after that. Also assembly house did tell me that they use closed loop water washing system so apparently it is a water under there.
So my few questions are:
1. Once water accumulated under IC, how long can it stay there, will it ever evaporate on its own?
2. Will drying board with hot air for some time get water out of there, and once it out, is it necessary to seal it with conformal coating or once out it will never get under there from the air?
3. Are there any specific chemicals that can displace water from under all components, kind of similar to what WD-40 does in metalworking industry but for PCBs?
4. What is the best option to avoid washing the board at all, my assemble provider uses lead free solder paste, I am not sure how necessary it is to wash the board after that, if there are no chemicals left after baking the board that will corrode it overtime, i'd rather just let it in there without washing.

Any advice on consumer level electronics production solder paste is highly appreciated, and thank you in advance for any help in this.

[free_electron]

water is not your problem. surface cleanliness is ...

you have flux remnants under the device . if they use a water soluble flux that flux is hygroscopic and turns into soap if it absorbs moisture.

Things to do to solve the problem

- NO silkscreen underneath IC packages. The thickness of the paint is sufficient to become a trap an have an impact on how good you can clean devices.
- if you are using DFN packages with a heat slug : provide venting channels in the soldermask so the flux has a chance to bake out during the soldering process, and cleaning fluid can get in.

[alexg]

water is not your problem. surface cleanliness is ...

you have flux remnants under the device . if they use a water soluble flux that flux is hygroscopic and turns into soap if it absorbs moisture.

Things to do to solve the problem

- NO silkscreen underneath IC packages. The thickness of the paint is sufficient to become a trap an have an impact on how good you can clean devices.
- if you are using DFN packages with a heat slug : provide venting channels in the soldermask so the flux has a chance to bake out during the soldering process, and cleaning fluid can get in.

Thank you, a follow up question: can I rewash the board in home lab setup? I'm thinking to buy distilled water and perhaps just rinse it for 15 minutes by agitating water manually and then drain with heat gun? Will that work?

[Benta]

water is not your problem. surface cleanliness is ...

you have flux remnants under the device . if they use a water soluble flux that flux is hygroscopic and turns into soap if it absorbs moisture.

Things to do to solve the problem

- NO silkscreen underneath IC packages. The thickness of the paint is sufficient to become a trap an have an impact on how good you can clean devices.
- if you are using DFN packages with a heat slug : provide venting channels in the soldermask so the flux has a chance to bake out during the soldering process, and cleaning fluid can get in.

Thank you, a follow up question: can I rewash the board in home lab setup? I'm thinking to buy distilled water and perhaps just rinse it for 15 minutes by agitating water manually and then drain with heat gun? Will that work?

I don't think distilled water will help. I'd propose IPA, but you need to analyze if there are sensitive components on the PCB (typically sensors or relays), ICs, resistors and caps are no problem.

[DaJMasta]

If you're looking for home lab cleaning, maybe an ultrasonic cleaner would get you better results?  They cost something, but unless your board is pretty large, you can likely get a competent frequency sweep cleaner for less than $100.

[beeryt]

Things to do to solve the problem
- if you are using DFN packages with a heat slug : provide venting channels in the soldermask so the flux has a chance to bake out during the soldering process, and cleaning fluid can get in.

Can someone describe in more detail what venting channels may look like?