|
Home > Archive > Electrical Engineering > April 2006 > Repairing corrosion damage to PCBs
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Repairing corrosion damage to PCBs
|
|
|
| There are many web pages dedicated to emergency procedures to take if you
spill a drink into a laptop. But I don't find any describing what to do about
remaining corrosion.
I was given a laptop that had coffee (no sugar, I'm told) spilled into the
keyboard. The owner turned off the power and removed the battery and drained
the computer as best he could, but did not take any steps to rinse out the
coffee. Some time later the computer quit.
There is evidence of corrosion on pc traces, SMD component and connector
leads:
http://www.hostmypic.info/uploads/27f71caf77.jpg
I plan to soak the board(s) in a weak solution of dishwashing detergent and
mineral-free h2o and dry for a few days. (Is there a more effective
solution?)
In preparation for examination and repair of any compromised conductors on
the pcbs, what's the best way to remove (ie, halt) the green & black residue
of corrosion?
Thanks,
--
DaveC
me@privacy.net
This is an invalid return address
Please reply in the news group
| |
| Smitty Two 2006-04-22, 4:21 pm |
| In article <0001HW.C06FC6DF03A7E32AF02845B0@news.readfreenews.net>,
DaveC <me@privacy.net> wrote:
> There are many web pages dedicated to emergency procedures to take if you
> spill a drink into a laptop. But I don't find any describing what to do about
> remaining corrosion.
>
> I was given a laptop that had coffee (no sugar, I'm told) spilled into the
> keyboard. The owner turned off the power and removed the battery and drained
> the computer as best he could, but did not take any steps to rinse out the
> coffee. Some time later the computer quit.
>
> There is evidence of corrosion on pc traces, SMD component and connector
> leads:
>
> http://www.hostmypic.info/uploads/27f71caf77.jpg
>
> I plan to soak the board(s) in a weak solution of dishwashing detergent and
> mineral-free h2o and dry for a few days. (Is there a more effective
> solution?)
>
> In preparation for examination and repair of any compromised conductors on
> the pcbs, what's the best way to remove (ie, halt) the green & black residue
> of corrosion?
>
> Thanks,
I'll take a shot at this, but YMMV. We clean our newly minted PWBAs in a
somewhat corrosive mixture of saponifier and distilled water. They're
then rinsed in fresh distilled water. Occasionally they're left in the
soap too long, or not rinsed well, and the nice shiny new solder joints
end up dull and dark.
I've cleaned gold contacts with liquid silver detarnish, but the best
thing for corroded solder is a drop of (can you believe it?) liquid
flux, and then reflow the joint.
If you've got actual crystals growing, or other severe surface
contamination, a stiff brushing (I use an acid brush with bristles cut
quite short to stiffen them) should preface the reflow work.
| |
| n cook 2006-04-22, 4:21 pm |
| DaveC <me@privacy.net> wrote in message
news:0001HW.C06FC6DF03A7E32AF02845B0@news.readfreenews.net...
> There are many web pages dedicated to emergency procedures to take if you
> spill a drink into a laptop. But I don't find any describing what to do
about
> remaining corrosion.
>
> I was given a laptop that had coffee (no sugar, I'm told) spilled into the
> keyboard. The owner turned off the power and removed the battery and
drained
> the computer as best he could, but did not take any steps to rinse out the
> coffee. Some time later the computer quit.
>
> There is evidence of corrosion on pc traces, SMD component and connector
> leads:
>
> http://www.hostmypic.info/uploads/27f71caf77.jpg
>
> I plan to soak the board(s) in a weak solution of dishwashing detergent
and
> mineral-free h2o and dry for a few days. (Is there a more effective
> solution?)
>
> In preparation for examination and repair of any compromised conductors on
> the pcbs, what's the best way to remove (ie, halt) the green & black
residue
> of corrosion?
>
> Thanks,
> --
> DaveC
> me@privacy.net
> This is an invalid return address
> Please reply in the news group
>
Wouldn't it be a multi-layer board with liquid conducted into the board by
capillary action ?
--
Diverse Devices, Southampton, England
electronic hints and repair briefs , schematics/manuals list on
http://home.graffiti.net/diverse:graffiti.net/
| |
|
| Thus spake n cook:
> Wouldn't it be a multi-layer board with liquid conducted into the board by
> capillary action ?
So, the board's toast? No reason to attempt repair?
Opinions welcome.
--
DaveC
me@privacy.net
This is an invalid return address
Please reply in the news group
| |
| Roy L. Fuchs 2006-04-22, 6:21 pm |
| On Sat, 22 Apr 2006 11:41:19 -0700, DaveC <me@privacy.net> Gave us:
>
>I plan to soak the board(s) in a weak solution of dishwashing detergent and
>mineral-free h2o and dry for a few days. (Is there a more effective
>solution?)
You may have to bake it at 60 C for an hour or so since there are
things on the board that were not on it when it was soldered and water
washed. A good vacuum cycle is good for evaporating those last little
bits of remaining water as well.
| |
| Roy L. Fuchs 2006-04-22, 6:21 pm |
| On Sat, 22 Apr 2006 11:41:19 -0700, DaveC <me@privacy.net> Gave us:
>In preparation for examination and repair of any compromised conductors on
>the pcbs, what's the best way to remove (ie, halt) the green & black residue
>of corrosion?
Use a fine child's tooth brush on it while you are doing the water
wash.
BTW, if there are any transformers on the board, all bet are off.
You would HAVE TO bake it or vacuum it out (the water).
| |
| Roy L. Fuchs 2006-04-22, 6:21 pm |
| On Sat, 22 Apr 2006 20:12:42 +0100, "n cook" <diverse8@gazeta.pl> Gave
us:
>DaveC <me@privacy.net> wrote in message
>news:0001HW.C06FC6DF03A7E32AF02845B0@news.readfreenews.net...
>about
>drained
>and
>residue
>
>Wouldn't it be a multi-layer board with liquid conducted into the board by
>capillary action ?
The material PCBs are made from, regardless of the layer count, are
hygroscopic and will absorb water.
If he cleans it with water, he will need to bake it out at the very
least.
| |
| Palindr☻me 2006-04-22, 6:21 pm |
| DaveC wrote:
> Thus spake n cook:
>
>
>
>
> So, the board's toast? No reason to attempt repair?
>
> Opinions welcome.
Toast. Not because of the above but because it stopped working whilst
powered up with the contamination in place. The odds are that
irreversible damage has been done.
--
Sue
| |
| bloggybob 2006-04-23, 6:21 am |
| No...... clean it and use a conductive ink pen .. i have fixed many laptop
keyboards and mainboards using one of these .. you basically draw a new line
on top of the old one with the pen which is silver conductive ink ... last i
checked MCM electronics carries them .. i still have one from years ago and
it is still half full ... hope this helps you out
cheers ..
"n cook" <diverse8@gazeta.pl> wrote in message
news:e2dv8p$fl6$1@inews.gazeta.pl...
> DaveC <me@privacy.net> wrote in message
> news:0001HW.C06FC6DF03A7E32AF02845B0@news.readfreenews.net...
you[color=darkred]
> about
the[color=darkred]
> drained
the[color=darkred]
> and
on[color=darkred]
> residue
>
> Wouldn't it be a multi-layer board with liquid conducted into the board by
> capillary action ?
>
> --
> Diverse Devices, Southampton, England
> electronic hints and repair briefs , schematics/manuals list on
> http://home.graffiti.net/diverse:graffiti.net/
>
>
>
| |
| Tzortzakakis Dimitrios 2006-04-23, 5:21 pm |
|
"Palindr?me" <me9@privacy.net>
news:124l560qd8o3j8f@corp.supernews.com...
> DaveC wrote:
by[color=darkred]
>
> Toast. Not because of the above but because it stopped working whilst
> powered up with the contamination in place. The odds are that
> irreversible damage has been done.
>
>
>
> --
> Sue
>
>
>
>
I left my 10 year old VCR an afternoon in the rain after it failed to
playback, considering it as garbage.I discovered after that some kid had
tampered with the switches and set them to NTSC playback.I plugged it still
wet, no on leds.After drying it out with a fan heater, it worked as
new.Probably has 10 more years of life left.(Real story).
--
Tzortzakakis Dimitrios
major in electrical engineering,freelance electrician
542nd mechanized infantry batallion
dimtzort AT otenet DOT gr
| |
| ampdoc 2006-04-23, 6:21 pm |
| Another factor to contend with on multilayer PCB's are the VIA's (some call
them feed-throughs) that rout traces through the board. If the metal has
been etched out of the VIA by corrosive action, then you have to take a very
small drill bit and clean out the thru-hole in the board an pass a small
wire through and solder it to the traces, meaning a lot of work if many of
them are bad. Also if the PCB is more than 2 layers it may be impossible to
repair.
Also if you have heavy soil on a board or it is contaminated with Glycol
(CRT Coolant for PJTV's) you can wash it with an ammonia based cleaner then
rinse with distilled water. Windex works wonders :-)
Jammy
| |
| Palindr☻me 2006-04-23, 8:21 pm |
| Tzortzakakis Dimitrios wrote:
> Ï "Palindr?me" <me9@privacy.net> Ýãñáøå óôï ìÞ_õìá
> news:124l560qd8o3j8f@corp.supernews.com...
>
>
> by
>
>
> I left my 10 year old VCR an afternoon in the rain after it failed to
> playback, considering it as garbage.I discovered after that some kid had
> tampered with the switches and set them to NTSC playback.I plugged it still
> wet, no on leds.After drying it out with a fan heater, it worked as
> new.Probably has 10 more years of life left.(Real story).
>
There is a difference between something that won't power up after
getting contaminated, say because the psu shuts down, and something
which powers up, works and then dies.
In the former case, almost all of the electronics has had no voltage
applied at all. And power supply circuitry tends to be simpler, works
with higher amplitude signals and lower impedances and with bigger track
spacing than what is being powered.
In the latter, I believe that dendritic growth can occur very quickly
between tracks with a voltage difference - once a ion-bearing liquid
bridges the gap between them. The unit will continue to function
"normally" whilst this growth is going on - until the gap is bridged and
a much higher current flows. It is this current which does the
destruction - hence the unit will work for some time before failing
catastrophically.
The problem with attempting to clean a board is that contamination may
exist in the gap under components,between components and board - which
may also contain many tracks. If contamination has already started to
crystalise out, it may not be readily soluble. Even an ultrasonic
cleaning bath may not remove these. If high voltage components or
sections of board have been contaminated, the currents that can flow
when growth has allowed tracking to take place can be "exciting".
--
Sue
| |
| Roy L. Fuchs 2006-04-23, 11:21 pm |
| On Sun, 23 Apr 2006 15:49:06 -0500, "ampdoc" <ampdoc@hotmail.com> Gave
us:
> Another factor to contend with on multilayer PCB's are the VIA's (some call
>them feed-throughs) that rout traces through the board. If the metal has
>been etched out of the VIA by corrosive action, then you have to take a very
>small drill bit and clean out the thru-hole in the board an pass a small
>wire through and solder it to the traces,
NEVER, I repeat NEVER "take a very small drill bit" to ANY of your
vias. Adding and soldering the wire is fine. Use FLUX for that
operation, NOT a friggin drill bit. That (drilling) would almost
ensure that there will be detached layers.
> meaning a lot of work if many of
>them are bad. Also if the PCB is more than 2 layers it may be impossible to
>repair.
Drilling the fucking thing will surely produce that result.
> Also if you have heavy soil on a board or it is contaminated with Glycol
>(CRT Coolant for PJTV's) you can wash it with an ammonia based cleaner then
>rinse with distilled water. Windex works wonders :-)
It's a laptop, not a CRT.
| |
| Roy L. Fuchs 2006-04-23, 11:21 pm |
| On Sun, 23 Apr 2006 23:31:28 +0100, Palindr?me <me9@privacy.net> Gave
us:
> hence the unit will work for some time before failing
>catastrophically.
Or simply failing to continue working, but not catastrophically.
I have repaired more than a few calculators that had soda or coffee
spilled on them, and they only required removal of the foreign media
and drying.
| |
| Charlie+ 2006-04-24, 6:21 am |
| On Sat, 22 Apr 2006 11:41:19 -0700, DaveC <me@privacy.net> wrote as
underneath my scribble :
Some actual experience:
I did a resurrect of a Sony MZ-R30 Minidisk Recorder/Player which had
been dropped in seawater and this had stopped working immediately!
This unit had been unwashed and returned around the world so had had
about a month or more to corrode, seawater is unforgiving so I wasnt
hopeful.
I removed the 1/32" FG PTH 4 layer boards completely, then washed
them in hot water, then blow dried with compressed air, you have to be
sensible with this as physical damage is possible with compressed air,
scanned them up to X10 for inspection then warmed boards (hot to
touch) then coated with WD40 for about 30 minutes, compressed air
again to remove most of the WD40 then baked in a warm oven for about 4
hours to flash off the remaining oil, repired some obviously corroded
SM joints by flux/solder and iron reflow (under magnification!) and
then reassembled. The point of the WD40 soak was to arrest any
ongoing corrosive action, it is thin enough to wick anywhere the
original liquid has gone.
Worked like a champ ever since, about 3 years now.
Your board has failed under power perhaps well after the event though
and this points towards electrolytic corrosion and this may not be
recoverable whatever you do, in a multilayered board.
Whatever you do I would advise you against soaking a board in anything
for a few days, imo that would be asking for trouble!
Charlie+
>There are many web pages dedicated to emergency procedures to take if you
>spill a drink into a laptop. But I don't find any describing what to do about
>remaining corrosion.
>
>I was given a laptop that had coffee (no sugar, I'm told) spilled into the
>keyboard. The owner turned off the power and removed the battery and drained
>the computer as best he could, but did not take any steps to rinse out the
>coffee. Some time later the computer quit.
>
>There is evidence of corrosion on pc traces, SMD component and connector
>leads:
>
>http://www.hostmypic.info/uploads/27f71caf77.jpg
>
>I plan to soak the board(s) in a weak solution of dishwashing detergent and
>mineral-free h2o and dry for a few days. (Is there a more effective
>solution?)
>
>In preparation for examination and repair of any compromised conductors on
>the pcbs, what's the best way to remove (ie, halt) the green & black residue
>of corrosion?
>
>Thanks,
| |
| Jon Elson 2006-04-27, 8:21 pm |
|
DaveC wrote:
>Thus spake n cook:
>
>
>
>
>So, the board's toast? No reason to attempt repair?
>
>
>
Unless the board has delaminated, the interior layers are darn near
hermetically
sealed in the laminating process. It is very unlikely that anything
like coffee
could penetrate a properly-built board. But, it sure can play hob with the
outer layers, especially when the power is on. Connectors are also easy for
liquids to wick into. I'd disconnect all the connectors, and some may
be the
flex-cable type that just presses a flex-print cable against the board,
and clean
anything there that needs cleaning. If glop got into the hard drive
(usually under
the keyboard!) that will likely be toast. The fans (if any) may have
been glopped
up, causing the delayed failure. Just tear it apart (it isn't working
any more,
anyway) and look for obvious problems. Gentle cleaning with alcohol and
an old toothbrush, followed by a day of open-air drying should fix most
fixable problems. Conductive glop getting between fine connections can
pop chips, so it may not be salvageable. But, worth a try.
Jon
|
|
|
|
|