Silicon oil leakage on PCB - how to remove?

[René]

Hi All,

We have a product containing a PCB filled with high-speed DSP's and
such (Video conference Codec). Above this PCB a mechanical damper is
placed to dampen the motion of a lid that can flip open.

We are now faced with malfunctioning dampers, causing Silicone Oil to
leak on the PCB.
Although the stuff seems non-conductive, the PCB's affected faithfully
stop working.

Questions:

1. How does Silicone Oil affect the electronics? (my meter can
measure up to 40 MOhm, but registers nothing - does the presence of
other materials / voltages etc. cause some reaction?)

2. Is there a way to get rid of the stuff and get the expensive
PCB working again? Wiping off does not seem to cut it - I fear some
irreversible chemical transformation took place.

I like to learn more about these phenomena - input welcome!

--
- René

 

[Spehro Pefhany]

On 27 Apr 2005 06:44:43 -0700, the renowned "Tim Shoppa"
<shoppa@trailing-edge.com> wrote:

How does Silicone Oil affect the electronics? (my meter can
measure up to 40 MOhm, but registers nothing - does the
presence of other materials / voltages etc.
cause some reaction?)

Any residue/dirt that the oil dissolves will increase its conductivity.
Unless you've got really high impedance stuff I'd be surprised if this
was a problem.

You seem worried about the board and conductivity of the oil, but I'd
be more worried about any sockets/plugs connecting to and from the
board and the oil making an insulating layer. Any socketed chips?

Tim.

We've put silicone oil on contacts to improve the reliability. ;-)
How do you know that it's silicone oil?

Any 32.768kHz etc. crystals?

I think the oil can "attract" a thin layer of moisture on the surface,
perhaps leading to subtle problems if the circuitry is really finicky.

It's probably really, really, hard to get rid of that stuff, BTW.


Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com

 

[Leon Heller]

"René" <rjz~nospam~@xs4all.nl> wrote in message
news:h84v61tomdfftfo2982vjnoa8ph24rb6aj@4ax.com...

Hi All,

We have a product containing a PCB filled with high-speed DSP's and
such (Video conference Codec). Above this PCB a mechanical damper is
placed to dampen the motion of a lid that can flip open.

We are now faced with malfunctioning dampers, causing Silicone Oil to
leak on the PCB.
Although the stuff seems non-conductive, the PCB's affected faithfully
stop working.

Questions:

1. How does Silicone Oil affect the electronics? (my meter can
measure up to 40 MOhm, but registers nothing - does the presence of
other materials / voltages etc. cause some reaction?)

2. Is there a way to get rid of the stuff and get the expensive
PCB working again? Wiping off does not seem to cut it - I fear some
irreversible chemical transformation took place.

I like to learn more about these phenomena - input welcome!

Silicone oil used to be used in Xerox copiers when I worked for Rank-Xerox
many years ago; we had a lot of trouble removing traces of it from panels
before repainting them (it stopped the paint adhering properly). We used a
special detergent from these people:

http://www.deconlabs.com

to remove it.

Don't get the stuff in your eyes, even a very small amount transferred from
the fingers can make them sore for hours. No permanent damage, though.

 

[René]

On 27 Apr 2005 06:44:43 -0700, "Tim Shoppa" <shoppa@trailing-edge.com>
wrote:

How does Silicone Oil affect the electronics? (my meter can
measure up to 40 MOhm, but registers nothing - does the
presence of other materials / voltages etc.
cause some reaction?)

Any residue/dirt that the oil dissolves will increase its conductivity.
Unless you've got really high impedance stuff I'd be surprised if this
was a problem.

You seem worried about the board and conductivity of the oil, but I'd
be more worried about any sockets/plugs connecting to and from the
board and the oil making an insulating layer. Any socketed chips?

Tim.

Hi Tim,
Thanks for fast response!
I would not be worried - if not the affected boards would fail
operation as soon as leakage occurs.

The PCB has several sub-functions, and the malfunction is clearly
tracable to the location of the "puddle".

If e.g. a drop of the stuff lands on the balanced mike amplifier - the
whole goes out of bias and the mike input is effectively dead. So
obviously it "does" something "conductive" with the circuit, though
directly measured (2 probes in a drop of the oil) does not reveal any
substantial conductivity. The circuits are not super-high impedant,
but on e.g. some amp circuits a ~1 MOhm phenomena would spoil the
function. Also some Audio / Video DA's were observed to stop.

Wiping the oil off sometimes clear the problem, but it may come back
later. A reliable repair method would probably involve more than just
wiping it off - but how?

--
- René

 

[John Woodgate]

I read in sci.electronics.design that Tim Shoppa
<shoppa@trailing-edge.com> wrote (in
<1114609483.567282.159800@f14g2000cwb.googlegroups.com&gt about 'Silicon
oil leakage on PCB - how to remove?', on Wed, 27 Apr 2005:

How does Silicone Oil affect the electronics? (my meter can
measure up to 40 MOhm, but registers nothing - does the
presence of other materials / voltages etc.
cause some reaction?)

Any residue/dirt that the oil dissolves will increase its conductivity.
Unless you've got really high impedance stuff I'd be surprised if this
was a problem.

You seem worried about the board and conductivity of the oil, but I'd
be more worried about any sockets/plugs connecting to and from the
board and the oil making an insulating layer. Any socketed chips?

Some of these oils have high penetrating power and a high dielectric

constant, so some capacitor values may have changed.

You can't wipe the stuff off because of its surface tension and
viscosity. You may be able to get a 'removing fluid' from the oil
suppliers. It may be costly.
--
Regards, John Woodgate, OOO - Own Opinions Only.
There are two sides to every question, except
'What is a Moebius strip?'
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

 

[John Woodgate]

I read in sci.electronics.design that René <rjz~nospam~@xs4all.nl> wrote
(in <nr6v611i2gjd8vosine7vqovjr6d7opmpg@4ax.com&gt about 'Silicon oil
leakage on PCB - how to remove?', on Wed, 27 Apr 2005:

A reliable repair method would probably involve more than just wiping
it off - but how?

Clean off with special cleaner and then conformally-coat the boards
affected.
--
Regards, John Woodgate, OOO - Own Opinions Only.
There are two sides to every question, except
'What is a Moebius strip?'
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

 

[René]

On 27 Apr 2005 07:19:34 -0700, "Tim Shoppa" <shoppa@trailing-edge.com>
wrote:

If e.g. a drop of the stuff lands on the balanced mike
amplifier - the whole goes out of bias and the
mike input is effectively dead.

OK, before you said video and codecs, which is rather low impedance
stuff, but a mike amp will be very sensitive to megohm level
impedances.

Also some Audio / Video DA's were observed to stop.

As Spehro asked, any CMOS crystal oscillators?

Tim.

The design (A Tandberg H323 Codec board) contains several dedicated
clock oscillators (complete Xtal+osc assemblies in a metal can) on
various frequencies. All work. (no oscillators on a "leftover" CMOS
gate)

Video parts also contain DA's / amps, that at some level run
impedances substantially higher than 75 Ohms.

The design has several separate operational parts that communicate
with each other - if one partial function does not chime in, the rest
may also not initialise. This makes cause / effect hunting a little
tricky. The balanced amp example is a nice "stand alone' function I
observed to fail when exposed to the oil.

Am I sure it's Silicone oil? Well - that is what the damper datasheet
tells me, but whatever special dopes are present I do not know.

Suppose "megaOhm" circuits were involved - would this be normal
behaviour for silicone oil?

I am now toying with Fluxclene - see what good that does...

Thanks for all suggestions mentioned so far by many. I will try the
Decon lead.

--
- René

 

[John Larkin]

On Wed, 27 Apr 2005 15:28:05 +0200, René <rjz~nospam~@xs4all.nl>
wrote:

Hi All,

We have a product containing a PCB filled with high-speed DSP's and
such (Video conference Codec). Above this PCB a mechanical damper is
placed to dampen the motion of a lid that can flip open.

We are now faced with malfunctioning dampers, causing Silicone Oil to
leak on the PCB.
Although the stuff seems non-conductive, the PCB's affected faithfully
stop working.

Questions:

1. How does Silicone Oil affect the electronics? (my meter can
measure up to 40 MOhm, but registers nothing - does the presence of
other materials / voltages etc. cause some reaction?)

2. Is there a way to get rid of the stuff and get the expensive
PCB working again? Wiping off does not seem to cut it - I fear some
irreversible chemical transformation took place.

I like to learn more about these phenomena - input welcome!

The stuff may be penetrating the epoxy of IC packages and causing
swelling, stressing chips and maybe breaking wire bonds. You might try
some side experiments to see if this is what's happening.

John

 

[RST Engineering (jw)]

We were using the stuff for compass fluid and had a similar problem. Put it
in the dishwasher on "hot" and use LIQUID dishwasher detergent, not the
granulated stuff that has an abrasive in it. After a couple of trips
through the dishwasher, rinse in 98% isopropyl alcohol and let it dry, dry,
dry. A warm location (NOT THE OVEN) of about 90°F helps it dry faster. A
small 5 watt bulb inside of a shoebox will get you into the general area.

Jim

2. Is there a way to get rid of the stuff and get the expensive
PCB working again? Wiping off does not seem to cut it - I fear some
irreversible chemical transformation took place.

I like to learn more about these phenomena - input welcome!

 

[John Woodgate]

I read in sci.electronics.design that René <rjz~nospam~@xs4all.nl> wrote
(in <u39v61pmg4ldj7qrthsr46c9iv17iudl4r@4ax.com&gt about 'Silicon oil
leakage on PCB - how to remove?', on Wed, 27 Apr 2005:

Suppose "megaOhm" circuits were involved - would this be normal
behaviour for silicone oil?

You won't have such high-impedance circuits in a mic amp. It's really
very surprising that that fails, unless some component is damaged by oil
penetration.
--
Regards, John Woodgate, OOO - Own Opinions Only.
There are two sides to every question, except
'What is a Moebius strip?'
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

 

[Glen Walpert]

On Wed, 27 Apr 2005 15:28:05 +0200, René <rjz~nospam~@xs4all.nl>
wrote:

Hi All,

We have a product containing a PCB filled with high-speed DSP's and
such (Video conference Codec). Above this PCB a mechanical damper is
placed to dampen the motion of a lid that can flip open.

We are now faced with malfunctioning dampers, causing Silicone Oil to
leak on the PCB.
Although the stuff seems non-conductive, the PCB's affected faithfully
stop working.

Questions:

1. How does Silicone Oil affect the electronics? (my meter can
measure up to 40 MOhm, but registers nothing - does the presence of
other materials / voltages etc. cause some reaction?)

Perhaps your PCB or some device packages are absorbing the oil,
swelling, and the resulting mechanical distortion is causing BGA ball
failure or some other mechanical failure. I would expect board
swelling to be visible as warpage, but package swelling could be hard
to see. If it is occurring, oil absorption should be measurable as
weight gain in a test which should be done at least at the peak
temperatures seen in service or accelerated by slightly higher
temperatures.

2. Is there a way to get rid of the stuff and get the expensive
PCB working again? Wiping off does not seem to cut it - I fear some
irreversible chemical transformation took place.

If the damage is from swelling it is irreversible.

And Sphero is right, the stuff is just about impossible to completely
remove even if only adsorbed (surface coating) vs absorbed (into
volume). I once knew someone who tried hundreds of methods to remove
silicone mold release from polycarbonate aircraft windshields prior to
CVD coating, without complete success. They had to make the process
tolerate some residual silicone. Acetone and some other solvents will
reduce it to a thin layer, however.

I like to learn more about these phenomena - input welcome!

It is also possible that the damper oil is not pure silicone oil but
has some additive(s) which attack some polymers - such as the seal on
the damper, and your board or packages.

 

[Paul Probert]

René wrote:

....

1. How does Silicone Oil affect the electronics? (my meter can
measure up to 40 MOhm, but registers nothing - does the presence of
other materials / voltages etc. cause some reaction?)

....

One of my projects has circuit boards immersed in high voltage oil. The oil induced
failure I get is when I have a radial electrolytic capacitor mounted standing up on the
circuit board. The rubber plug where the leads come out soaks up oil, swells, and pulls
the leads right out of the circuit board. It is pretty much an irresistible force.
Anything that has rubber in it is going to have trouble.

Also on these boards we have fiber optic cables. At first we tried the common orange kind,
but that orange stuff, whatever it is, soaked up the oil and became very stiff. It also
expanded enough to yank the fibers out of the connectors.

Paul Probert
University of Wisconsin

 

[Daniel Haude]

On Wed, 27 Apr 2005 10:28:55 -0700,
Chris Carlen <crcarleRemoveThis@BOGUSsandia.gov> wrote
in Msg. <d4oi4n02lb8@news2.newsguy.com>

2. potassium hydroxide in methanol - really effective at stripping
grease off glass, and leaving a surface completely free of foreign
material.

This is interesting. What concentration? Does the KOH just dissolve in
methanol, or does one add some water? Does NaOH work as well?

I work in the UHV field, and somethimes we have to get small traces of
stuff off of surfaces.

3. Boiling perchloric acid - oh, this stuff was just way too much fun.
Used to clean trace residues of metals and metal oxides/hydroxides off
glassware used for trace analysis. Good for cleaning out the frits of
sintered glass crucibles. Don't treat anything organic or boil it to
anhydrous, or else KA-BOOM!

4. Hydrofluoric acid - also good at restoring wetability to glass. But
even more scary than HCl04.

Are you talking about smoking, concentarted, boiling HCl? Your post
doesn't sound like you believed in diluting anything ;-)

--Daniel

 

[Daniel Haude]

On Wed, 27 Apr 2005 16:08:15 GMT,
Glen Walpert <gwalpert@notaxs.com> wrote
in Msg. <bvbv615arcnq2n6mtlur1bjcqn03qv38rp@4ax.com>

If the damage is from swelling it is irreversible.

This can probably be excluded if the damage occurs immediately after
exposure to oil because I guess the swelling/soaking would take some time.

--D.

 

[John Woodgate]

I read in sci.electronics.design that Daniel Haude
<haude@kir.physnet.uni-hamburg.de> wrote (in
<slrnd713t8.oru.haude@kir.physnet.uni-hamburg.de&gt about 'Silicon oil
leakage on PCB - how to remove?', on Thu, 28 Apr 2005:

Are you talking about smoking, concentarted, boiling HCl?

No, perchloric acid HClO4 is quite different from hydrochloric acid HCl.

Your post doesn't sound like you believed in diluting anything ;-)

Naturally, he's a chemist. (;-)

BTW, KOH does dissolve in methanol. Whether you could use NaOH is
another matter.
--
Regards, John Woodgate, OOO - Own Opinions Only.
There are two sides to every question, except
'What is a Moebius strip?'
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

 

[Aubrey McIntosh, Ph.D.]

Chris Carlen wrote:

René wrote:



My favorite:

3. Boiling perchloric acid - oh, this stuff was just way too much fun.
Used to clean trace residues of metals and metal oxides/hydroxides off
glassware used for trace analysis. Good for cleaning out the frits of
sintered glass crucibles. Don't treat anything organic or boil it to
anhydrous, or else KA-BOOM!

4. Hydrofluoric acid - also good at restoring wetability to glass. But
even more scary than HCl04.

no. no. no. no.

 

[Aubrey McIntosh, Ph.D.]

John Woodgate wrote:

You can't wipe the stuff off because of its surface tension and
viscosity. You may be able to get a 'removing fluid' from the oil
suppliers. It may be costly.

I am assuming that the problem oil is something like a
polydimethylsiloxane
( http://www.chemexper.com/chemicals/supplier/cas/63148-62-9.html )

I would worry that the KOH / Methanol will "clean" the polymers off of
the pcb silkscreen, chips, and even the PCB substrate. I have seen 250
mL beakers that I used as a KOH/Methanol bath display significant
etching. It dissolves skin. Use gloves. It dissolves corneas faster.
Use goggles. Spatter is invisible on clothing, and transfers to
unprotected skin. Change clothes and shower after handling.
NaOH/Methanol forms insoluable crud, so KOH is preferred.

A couple of web sites mention benzene. You can't (or shouldn't) get
that now, but toluene or xylene may substitute. I'm a little surprised
that they work, and don't remember trying them directly. I suspect that
you can get mixed xylenes at an art supply house as some or other paint
thinner or brush cleaner. They burn easily -- act as if you are
handling gasoline.

Someone else mentioned diethyl ether. The flammability of (this) ether
makes gasoline look safe. More problematic is that you will be tempted
to keep the left over material. On exposure to oxygen, even in a
"closed" container, it forms explosive material with LEGENDARY sensitivity.

Another web site did not offer any information specifically relevant to
this newsgroup at all: http://www.indecentblogging.com/blog.php?id=5906

Despite the general acceptance of opinion over lab work in most
newsgroups, I do mention the entry from the paper mill. Use your
browser to search for "Table 1" about a third of the way down the
article. My gut sense was ethyl acetate would be a good solvent, and
their finding of methyl ethyl ketone fits in nicely with that. Their
finding of hexane surprises me some, but hey, they went to the lab and I
didn't.

http://tappsa.co.za/archive2/APPW_2004/Title2004/Determination_of_silicone_defo/determination_of_silicone_defo.html

 

[John Woodgate]

I read in sci.electronics.design that "Aubrey McIntosh, Ph.D."
<newsposter@spam.vima.austin.tx.us> wrote (in
<1171nrtg6nl5c3e@corp.supernews.com&gt about 'Silicon oil leakage on PCB
- how to remove?', on Thu, 28 Apr 2005:

A couple of web sites mention benzene. You can't (or shouldn't) get
that now, but toluene or xylene may substitute. I'm a little surprised
that they work, and don't remember trying them directly. I suspect
that you can get mixed xylenes at an art supply house as some or other
paint thinner or brush cleaner. They burn easily -- act as if you are
handling gasoline.

You can get toluene as solvent for impact adhesive.

Someone else mentioned diethyl ether. The flammability of (this) ether
makes gasoline look safe. More problematic is that you will be tempted
to keep the left over material. On exposure to oxygen, even in a
"closed" container, it forms explosive material with LEGENDARY
sensitivity.

True, but IIRC it takes years to produce sufficient concentration to go
pop.
--
Regards, John Woodgate, OOO - Own Opinions Only.
There are two sides to every question, except
'What is a Moebius strip?'
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

 

[Chris Carlen]

Aubrey McIntosh, Ph.D. wrote:

John Woodgate wrote:


You can't wipe the stuff off because of its surface tension and
viscosity. You may be able to get a 'removing fluid' from the oil
suppliers. It may be costly.

I am assuming that the problem oil is something like a polydimethylsiloxane
( http://www.chemexper.com/chemicals/supplier/cas/63148-62-9.html )

I would worry that the KOH / Methanol will "clean" the polymers off of
the pcb silkscreen, chips, and even the PCB substrate. I have seen 250
mL beakers that I used as a KOH/Methanol bath display significant
etching. It dissolves skin. Use gloves. It dissolves corneas faster.
Use goggles. Spatter is invisible on clothing, and transfers to
unprotected skin. Change clothes and shower after handling.
NaOH/Methanol forms insoluable crud, so KOH is preferred.

The KOH/MeOH, NaOH, HClO4 and related mentions were more for fun than
considered to be applicable to the problem at hand, as they are
certainly NOT. That was spelled out in my post. The diethyl ether was
the serious recommendation.

A couple of web sites mention benzene. You can't (or shouldn't) get
that now, but toluene or xylene may substitute. I'm a little surprised
that they work, and don't remember trying them directly. I suspect that
you can get mixed xylenes at an art supply house as some or other paint
thinner or brush cleaner. They burn easily -- act as if you are
handling gasoline.

Someone else mentioned diethyl ether. The flammability of (this) ether
makes gasoline look safe. More problematic is that you will be tempted
to keep the left over material. On exposure to oxygen, even in a
"closed" container, it forms explosive material with LEGENDARY sensitivity.

It's not a big deal if dealt with using normal precautions for volatile
flammables. An explosion proof fume hood typical of chem labs is the
best idea, otherwise a garage or outside patio with a check for
potential ignition sources nearby should be appropriate levels of
precaution.

Most diethyl ether is stabilized with some addatives, so the peroxide
problem isn't that big of a deal. I think we generally disposed of
ether older than 6 months or something like that. Not too difficult to
do a little research to determine prudent practices for this time. Bill
Sloman might have something to say. Or go on over to sci.chem and ask
about how to keep oneself safe with ether.

Another web site did not offer any information specifically relevant to
this newsgroup at all: http://www.indecentblogging.com/blog.php?id=5906

Despite the general acceptance of opinion over lab work in most
newsgroups, I do mention the entry from the paper mill. Use your
browser to search for "Table 1" about a third of the way down the
article. My gut sense was ethyl acetate would be a good solvent, and
their finding of methyl ethyl ketone fits in nicely with that. Their
finding of hexane surprises me some, but hey, they went to the lab and I
didn't.

http://tappsa.co.za/archive2/APPW_2004/Title2004/Determination_of_silicone_defo/determination_of_silicone_defo.html

Ether really was the only thing that worked for silicone vacuum grease.
Oils might be a little easier to solvate.


good day!



--
_______________________________________________________________________
Christopher R. Carlen
Principal Laser/Optical Technologist
Sandia National Laboratories CA USA
crcarleRemoveThis@BOGUSsandia.gov
NOTE, delete texts: "RemoveThis" and "BOGUS" from email address to reply.

 

[Zak]

Chris Carlen wrote:

Most diethyl ether is stabilized with some addatives, so the peroxide
problem isn't that big of a deal.

ISTR metals catalyze the peroxidation... thus tins would colect a nice
explosive crust where vapour would exit from the container and mix with
air. Don't screw that cap off...

article. My gut sense was ethyl acetate would be a good solvent, and
their finding of methyl ethyl ketone fits in nicely with that.

I have a bottle of pinkish MEK that was sold as a nail varnish remover
"without acetone".


Thomas