13
\$\begingroup\$

I found this video where a technician with a lot of experience seems to clean his PCBs using water, and then he just puts it in the sun so it can dry. I was just amazed to see the circuit keeps working as intended, he says he keeps doing this all the time and several times, however I doubt this is normal.

Which leave me some questions:

  • What are the recommended methods to clean huge PCBs like the one in the video?
  • What happens to charged capacitors when water shorts them out?
  • Solder joints do not get any kind of corrosion?
  • Is the antisolder mask waterproof?

Video Thumbnail

\$\endgroup\$
  • \$\begingroup\$ are you sure it is normal water and not demineralized water? \$\endgroup\$ – T J Dec 3 '14 at 8:32
  • \$\begingroup\$ yeah its normal water he says so in the video \$\endgroup\$ – GoatZero Dec 3 '14 at 9:16
  • 2
    \$\begingroup\$ In 2000 I was working as a trainee in a small computer repair store when there was a big flood in my city. After that catastrophe, people sent us a lot of dirty computers to try save them. The only solution was to wash all of them with water just like the picture. And almost all of the computers were saved. But the hard drivers didn't have the same luck. \$\endgroup\$ – Daniel Grillo Dec 3 '14 at 10:55
  • 4
    \$\begingroup\$ I have seen dishwashers used to clean new boards that were built using solder with water-soluble flux, although de-ionized (DI) water is normally used. \$\endgroup\$ – Tut Dec 3 '14 at 13:21
  • 4
    \$\begingroup\$ I did some board assembly at a job about ten years ago. My boss had us clean the flux off with soap, water, and a toothbrush in the restroom. Once the board was clean, we blew the water off with compressed air. \$\endgroup\$ – Adam Haun Dec 3 '14 at 13:35
19
\$\begingroup\$

This is certainly possible, but a few things to keep in mind:

  • Rosin flux residue often isn't water soluble. For cleaning after assembly, water is not generally effective
  • Salts in the water will, especially if the pcb is left to dry in the sun and isn't mechanically dried, deposit on the PCB and ruin isolation distances and surface conductivity for sensitive circuits. To give you an example: IPA-rinsed FR-4 boards have in the order of \$10^8\$ Mohm * cm surface resistivity, a salt-contaminated board can go as low as \$10^3\$ Mohm * cm. Salt contamination is also more likely to attract local water condensation in high humidity environments.
  • Crevice corrosion will occur in places where the water is trapped and cannot effectively vaporize during drying

All these reasons are why isopropyl alcohol (IPA) is most commonly used for board cleaning; it is an effective solvent for most fluxes and other common contaminants (grease, thermal compound, etc.) on PCBs, it does not affect surface resistivity and it is extremely volatile, with even very high diffusion rates and low vapor pressure at freezing temperatures.

Water has a chance of causing problems, but if used with this in mind (and certainly if care is taken to use demineralized water) it can be an effective cleaning agent if nothing else is available. Certainly, it is better than a PCB covered in dust in many cases.

As for your specific question on charged capacitors: in general, when removing electronics from its indended application either the engineer who made it has to make sure the design will eventually discharge completely and/or to safe levels, or the user needs to recognize designs not intended to be serviced and discharge any capacitors prior to cleaning. If not, water is certainly slightly conductive but not very much so. Even a droplet right in between two terminals will only very slowly discharge a capacitor with any significant charge.

\$\endgroup\$
  • 3
    \$\begingroup\$ While Iso Propyl Alcohol (IPA) is useful, the fumes, handling precautions & general cost are nuisance factors. Depending on the quality of tap water in your area, you may find it easier to wash in H2O - and mechanically dry (blow dry with a hot air gun). \$\endgroup\$ – Alan Campbell Dec 3 '14 at 8:57
  • \$\begingroup\$ Does Mohm*cm make sense as a unit for surface resistivity? Since the surface is 2D I'd expect units of Mohms * cm per cm, or just Mohms. \$\endgroup\$ – user253751 Jan 16 '18 at 23:36
  • \$\begingroup\$ You can use water, it must be ultra pure, and only rinsed for a short time, some board houses use this process. I happen to have access to ultra pure water, and I use water to rinse my boards after the oven to clean off water soluble flux. \$\endgroup\$ – Voltage Spike Sep 24 at 17:36
8
\$\begingroup\$

What are the recommended methods to clean huge PCBs like the one in the video?

Most soldering processes involve flux that must be cleaned from the board after soldering. Some fluxes are water soluble, so yes, using filtered water is fine. Most electronic components that can be machine mounted are washable. Buzzers often have stickers covering the hole to prevent water infiltration, and other components may have warnings or special requirements for washing, but generally most components are not damaged by brief exposure to water when not in use.

Other fluxes require different solvents for cleaning.

Typically the solvent, whether water or something else, is blown off the PCB and components after washing. Air drying may leave residue on the components, and exposes the various metals to water for longer. Heating may reduce exposure, but may still leave residue.

What happens to charged capacitors when water shorts them out?

Washing is done immediately after production, and prior to testing. Typically boards are not powered up until after the board is dry and clean.

If you need to wash a board after its been powered up, for instance after rework, then you'll need to discharge any capacitors and remove, cover, or otherwise meet the washing requirements of special components.

Solder joints do not get any kind of corrosion?

Solder joints quickly form a microscopically thin layer of oxide soon after soldering due to air moisture. This layer protects the joint from further oxidation and degradation. Water, over the short term exposure needed for washing, will not cause greater oxidation.

This layer of oxide is one of the main reasons to use solder with flux.

Is the antisolder mask waterprof?

Yes, typically it is.

\$\endgroup\$
4
\$\begingroup\$

This cleaning process is practical and does work. Old guy I worked with used it routinely and placed boards in oven at 50degC to dry if there was no sun.

Many modern boards are aqueous cleaned so up to the task. Some are not well suited to it especially older boards with open frame potentiometers or other open components.

I should just mention a couple of things to keep in mind.

Rinsing with IPA or DI water is a good idea to get rid of any lingering dissolved salts by diluting them away.

Using oven, air or a centrifuge to speed up drying is a good thing. All water must be removed before restoring power. Connectors and any components with moving parts are at risk of trapped water.

If there are keep alive batteries then you should work fast or remove the battery. Water with salts will corrode under battery power pretty fast an keep working under ICs if water is trapped there. I do not think caps will usually be charged by the time you have a board out of a instrument but they will work like a battery if the water is conductive but for a shorter time. If DI water then the caps could remain charged and catch the unwary.

If dirt on the board is more than water soluble dust or mud then other detergents or solvents will be required. They are usually much more expensive. Clean Green or similar is a powerful surfactant that is required in small quantities. Certain board cleaning solvents use freon compounds that dry fast but seem to dissolve oily dirt but not conformal coatings. Acetone, thinners, methanol, ethanol are all options but carry the risk of dissolving some or other plastic or ink or crazing some plastics so should be used with caution and experience.

\$\endgroup\$
1
\$\begingroup\$

Back in the days of vacuum tube radios, etc., the Navy used to use an automotive Gunk Tank (carb cleaner) and the whole assembly would be put in the tank after tubes were removed. Then air compressor blow dry and place in a pizza oven at low temperature. Variable capacitors with ball bearings had to be re-lubed and sometimes rotor brushes burnished with business card stock. The rotary ceramic trimmer capers had to disassembled with a soldering pencil and the gunk removed, then reassembled and soldered together. I saw a few units where dial decals came off! Also one shop had to buy a second gunk tank for the teletype machines due to lots of grease and oil contaminated the solution used for electronic assembly washing.

I have seen small ultrasonic jewelry cleaners used for small PCBs.

\$\endgroup\$
  • \$\begingroup\$ Be careful with ultrasonic cleaners. They will destroy any MEMS devices on the board -- accelerometers, gyros, silicon resonators, some barometers and microphones… \$\endgroup\$ – duskwuff Sep 24 at 18:42
  • \$\begingroup\$ @duskwuff I've heard that this can be somewhat alleviated with more expensive frequency sweeping cleaners. Plus I think the damage can extend to more than just MEMs device, but MEMs devices would be particularly vulnerable. \$\endgroup\$ – DKNguyen Sep 24 at 19:14
0
\$\begingroup\$

This is fine if:

  • You use ultra pure water (deionized and demineralized, the impedance should be in the MegΩ, with a water impedance tester)
  • No components on the board are sensitive to humidity or water
  • Using water soluble flux.

I know some board houses that use ultra pure water to clean off water soluble flux. I have also adopted the process for prototyping. There are a few MEM's and optical sensors that cannot be used with water. Water will damage the components.

It is also not ok to reflow after getting water on a board, this could lead to humidity building up in some parts and exploding and damaging the part (or in the least displacement of packages may occur.

\$\endgroup\$

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.