Board Cleaning

[glee]

I have inherited a circuit board that has become "soiled" somehow - there are green and white powdery corrosion products on the connection pins of several ICs, brown stains (from some liquid) on the surface of the board itself, fingerprint shaped residue marks everywhere, and in places it just looks dull and tarnished. It may be that a battery leaked in the case at some point. Board appears to be working ok.

I'm guessing that alcohol of some kind will clean up most of it. The corrosion worries me. Is alcohol appropriate? Will it halt the process or should I be looking for something else?

Thanks.

[Rerouter]

the green powder is from a battery over discharge, the white is likely water soluble residue, and the brown is likely also from a battery,

for the powders isopropl is more than enough, the brown hoever if it is battery residue i would reccomend acetone,

as for the corrosion, is it on the traces, or the pads or the joints? if its on the traces, and it has no solder mask, grab a piece of sandpaper in the 1000's grit, go over it lightly till you see shiny copper and then spray the board with a laquer, if its on the pads and joints, desolder and resolder,

hope this helps

[glee]

Thanks for the tips. That's helpful. I'll try cleaning with isopropyl first and get a better idea of what remains to do. Maybe I'll post some photos if it's unclear. Thanks!

[ptricks]

For really corroded or boards that have been damaged by leaked caps or dirt, a solution of baking soda (the kind you put in the fridge for odors), and water with a tooth brush for scrubbing works well. Baking soda is alkaline and will neutralize anything like acids and the slight abrasive will remove other stuff without harming anything. Rinse in water and then alcohol being careful of plastics that might get damaged.   I actually use a garden hose with some water pressure to clean dirty electronics then I use an oven or hair dryer to completely dry .

[glee]

Nice. :-)

While I was trawling around for answers I discovered a video of a guy washing a board with detergent in his sink and scrubbing with a toothbrush. He did take care to rinse it thoroughly. Then let it sit in a "airing cupboard" (a rack) for "3 days". I couldn't figure out why he didn't go with active drying of some kind. But anyway it appeared to work (in that case, at least).

[ablacon64]

I don't recommend letting it "sit for 3 days", drying must be faster than that, use a reflow oven, a DIY stove (aluminum coated paper cardboard box with a 150W light for example) or a hair dryer, like suggested by ptricks.

It's ok to wash using water as long as you dry it properly and fast (at least faster than the sun temperature will).

[T4P]

Or use a BGA preheater or a SMD rework station at 100C if you have one and you can also use a hair dryer
Automated heating ... If you have a thermal chamber like dave hacked one for, that can be useful

[glee]

Here are some photos. I didn't really have the right lens for this ...









I cleaned it up best I could using isopropyl alcohol, water and a toothbrush today. A hairdryer was fine for evaporation. It came up ok but not anything like new, of course.

Seems that a battery had definitely leaked in it's compartment. The images here show small isolated parts of the board. Damage was widespread but in localised regions. I can't really see how it could have happened that way. It's not like there are signs of continuous acid flow across the surface. More like it was transfered through the electrical conductors and emerged in different parts of the board. Or maybe it was vapour? Who knows.

Now I'm curious. Can anyone explain the differentiated colours? Why some regions have surface oxidation and in others metal appears to be stained brown?

[ablacon64]

It's not that bad. Apply some flux, hot air and then IPA with a small brush.

[ptricks]

Now I'm curious. Can anyone explain the differentiated colours? Why some regions have surface oxidation and in others metal appears to be stained brown?

My guess is that what did the damage is rain water. Rain water tends to be more acidic than tap water , combine that with the various metals present on the board, copper, tin, possibly lead if not ROHS , the metals mix, electrons move  and it corrodes parts.  The clue for me that it was something acidic is the copper turning green, copper will not do that in a standard 7.0 ph tap water.  If it were salt water you would see way more corrosion.  You can do the same thing with a copper wire and some zinc, place both in a slightly acidic, 6.0 ph,  solution and watch the process turn the copper green and the copper plate itself onto the zinc.

[glee]

Ok, that's interesting.

Mysterious that it's so localised in different regions of the board, and remote from both the battery compartment and the input jack sockets.

I wonder if acid rainwater can explain the dark brown staining to the lower left corner of the board in the last image? This discolouration seems to be deep in the fibreglass itself. Certainly seems as if at least that corner has been exposed to some bulk liquid. The case seals seem original and intact, so it's a little hard to understand.

I guess it's feasible that rainwater came in through the connection jacks and mingled with the battery directly below - but then most of that part of the board seems relatively ok, tarnished and dull but not actually corroded or stained. 

Anyway, understanding "exactly what happened" is not the goal here. I'm just trying to develop a better sense of interpreting the outcome for future reference.

By the way, can anyone identify this (inside the rear case shell of a Fluke 87-V)? You can see that one of the springs that are supposed to provide electrical contact is missing, and some staining in the metal under the paper at lower right. Thanks.

[ptricks]

By the way, can anyone identify this (inside the rear case shell of a Fluke 87-V)? You can see that one of the springs that are supposed to provide electrical contact is missing, and some staining in the metal under the paper at lower right. Thanks.

That is the piezo for the sounds that the meter makes like the beeps or tones.  Just about any other piezo should work if that one is too corroded.

http://en.wikipedia.org/wiki/Buzzer

[SeanB]

Just remember that you need one with a feedback connection, which has 3 wires. You might have to attach 3 short thin wires to the new one and stick it to the board if you cannot get one with the same pattern as the fluke one.

[glee]

Ah, of course. Makes sense. Thanks for the help.