Any further use for corroded RAM?

I was getting some persistent errors in my Pavilion A1130N. I ran
memtest for a few days and kept getting occasional errors on test
#7.So I took a look at everything and saw that there was some kind of
corrosion-looking substance on my 512Mb DDR 400, Micron
Mt8VDDT6464AG-40BDB, that came with the PC when I first bought it. I
can't tell exactly what the substance is, but it looks like a whitish
corrosion.

This "substance" is in the suspicious pattern of what looks like a
thumb-print on one side and several fingers on the other. If they are
fingerprints, then it was not I, the sole owner, that did it, since I
have never even touched the stock modules before.

Could this "stuff" just be the oils from someone's fingers that has
chemically changed over the years to become something that can cause a
short?

I would just dismiss the apparent "finger pattern" as a coincidence,
except for the fact that there is a "large" mass of the "stuff" on one
part of the board that is far away from any corrodible metal.

If this stuff is just some careless idiots finger crud, then would you
think that I could clean the module in some way and possibly get it
working reliably again? If so, then what kind of cleaning method would
you recommend?

I also noticed that there is a similar, yet not as noticeable,
discoloration pattern on the other stock 512 module. I ran memtest
with it for several days but never got any errors. Do you think I
should just go ahead and toss it out? Or do you think it might be
salvageable as well?

I would replace the corroded memory. Apparently in the "factory" where the
computer was assembled, there are not policies and procedures in place to
minimize mishandling of the electronic parts... Ben Myers

On Tue, 12 Feb 2008 19:22:21 -0800 (PST), ShadowTek <>
wrote:

>I was getting some persistent errors in my Pavilion A1130N. I ran
>memtest for a few days and kept getting occasional errors on test
>#7.So I took a look at everything and saw that there was some kind of
>corrosion-looking substance on my 512Mb DDR 400, Micron
>Mt8VDDT6464AG-40BDB, that came with the PC when I first bought it. I
>can't tell exactly what the substance is, but it looks like a whitish
>corrosion.
>
>This "substance" is in the suspicious pattern of what looks like a
>thumb-print on one side and several fingers on the other. If they are
>fingerprints, then it was not I, the sole owner, that did it, since I
>have never even touched the stock modules before.
>
>Could this "stuff" just be the oils from someone's fingers that has
>chemically changed over the years to become something that can cause a
>short?
>
>I would just dismiss the apparent "finger pattern" as a coincidence,
>except for the fact that there is a "large" mass of the "stuff" on one
>part of the board that is far away from any corrodible metal.
>
>If this stuff is just some careless idiots finger crud, then would you
>think that I could clean the module in some way and possibly get it
>working reliably again? If so, then what kind of cleaning method would
>you recommend?
>
>I also noticed that there is a similar, yet not as noticeable,
>discoloration pattern on the other stock 512 module. I ran memtest
>with it for several days but never got any errors. Do you think I
>should just go ahead and toss it out? Or do you think it might be
>salvageable as well?

"ShadowTek" <> wrote in message
news:a3aa26ce-e5c4-4c96-a0e9-...

> If this stuff is just some careless idiots finger crud, then would you
> think that I could clean the module in some way and possibly get it
> working reliably again? If so, then what kind of cleaning method would
> you recommend?

I usually clean the contacts with a plain piece of paper (like laser printer or
inkjet paper)
It's abrasive enough to remove tarnish yet is gentle on the actual metal fingers.
Some
people use the Pink Pearl method. That's rubbing the contacts with an eraser.
If you rub the contacts briskly with the paper, look at the paper afterwards. If
it has a grey streak on it, you're getting off tarnish and corrosion. After
cleaning, you
should treat the contacts with a contact protector and lube (usually available at
Radio Shack
or similar stores. If your sockets are gold plated while your RAM board fingers are
tin
plated (or vice-versa) you will eventually have problems with corrosion. You should
match
gold to gold and tin to tin.

Tom Lake

ShadowTek <> wrote:
> I was getting some persistent errors in my Pavilion A1130N. I ran
> memtest for a few days and kept getting occasional errors on test
> #7.So I took a look at everything and saw that there was some kind of
> corrosion-looking substance on my 512Mb DDR 400, Micron
> Mt8VDDT6464AG-40BDB, that came with the PC when I first bought it. I
> can't tell exactly what the substance is, but it looks like a whitish
> corrosion.
>
> This "substance" is in the suspicious pattern of what looks like a
> thumb-print on one side and several fingers on the other. If they are
> fingerprints, then it was not I, the sole owner, that did it, since I
> have never even touched the stock modules before.

remove the module and clean it with a new toothbrush and hot distilled
water. Let it soak for a few minutes. you can do this on a plate.

Really, the distilled water part is key. Don't use tap water, it just
makes electronics dirtier.

Dry things off with compressed air, or a hair dryer. Reinstall the memory
and try again.

There wasn't any corrosion on the socket connectors. The stuff was
just on the upper two-thirds of the module.

Is there any definitive way to tell if this stuff is, in fact,
corrosion that has compromised the integrity of the metal? If the
components of the module are still good, and all it needs is a "bath",
then I would like to give that a try. But I don't want to waste my
time on something that might result in my data files getting corrupted
when I try and start using the module in my system again.

I guess I could stick the it under my microscope after I clean it, and
look over the surface for any signs of pitting or wear in the circuit
pathways.

Aside from that, what RAM manufacturer is know for the best
reliability and chemical durability?

This is pretty ridiculous that BOTH the Micron modules have been
chemically degraded.

I have owned several Kingston modules over the years, and have one 1Gb
module in THIS system at the moment, and I have never had any problems
with any of them. They still look as shiny as the day they were
purchased. Once you blast em with a can of duster, that is.

ShadowTek <> wrote:
> There wasn't any corrosion on the socket connectors. The stuff was
> just on the upper two-thirds of the module.
>
> Is there any definitive way to tell if this stuff is, in fact,
> corrosion that has compromised the integrity of the metal? If the
> components of the module are still good, and all it needs is a "bath",
> then I would like to give that a try. But I don't want to waste my
> time on something that might result in my data files getting corrupted
> when I try and start using the module in my system again.
>
> I guess I could stick the it under my microscope after I clean it, and
> look over the surface for any signs of pitting or wear in the circuit
> pathways.
>
> Aside from that, what RAM manufacturer is know for the best
> reliability and chemical durability?
>
> This is pretty ridiculous that BOTH the Micron modules have been
> chemically degraded.

how did you conclude "chemically degraded"?

> I have owned several Kingston modules over the years, and have one 1Gb
> module in THIS system at the moment, and I have never had any problems
> with any of them. They still look as shiny as the day they were
> purchased. Once you blast em with a can of duster, that is.

memory is not made to submerged in ****. PC magazine doesn't run tests on
which memory works best under chemical x.

Is it possible the stuff you're seeing is heatsink grease, that somebody
maybe got all over their hands?

distilled water is cheap and a powerful cleaner. Give it as shot.

> how did you conclude "chemically degraded"?

Well, that's my way of saying "corroded", since corrosion is the
result of a chemical process. This is, assuming that this stuff is
actually corrosion.

> Is it possible the stuff you're seeing is heatsink grease, that somebody
> maybe got all over their hands?
>
> distilled water is cheap and a powerful cleaner. Give it as shot.

That could very well be what it is. But don't they usually use
nonconductive grease like silicone? I thought that the only people who
used conductive thermal interface material where overclockers who buy
that Arctic Silver stuff.

I mean, wouldn't the substance have to be conductive, causing a short,
to cause a problem? For what other reasons could a foreign substance
cause a RAM error, if not for causing a short?

Also, I have a can of some electrical contact cleaner lying around.
Would this be safe to use on RAM?
http://www.crcindustries.com/auto/co....aspx?PN=05102

I stuck the module under my microscope and, to my suprise, managed to
get some decent pictures with my digital camera. I found a few places
that could be causing shorts.

40X magnification
http://img145.imageshack.us/my.php?i...m000218ub6.jpg
http://img164.imageshack.us/my.php?i...m000220ks5.jpg
http://img204.imageshack.us/my.php?i...m000224cs2.jpg

This picture, taken at 100X magnification, shows a little "hair-like"
strand that reaches across from one metal "pad" to another. Could this
be that "metal growth" that can cause shorts inside electronics? Or
does this just look like a regular fiber or hair?
http://img404.imageshack.us/my.php?i...m000222yo9.jpg

On Wed, 13 Feb 2008 06:43:19 -0500, Ben Myers <> wrote:

>I would replace the corroded memory. Apparently in the "factory" where the
>computer was assembled, there are not policies and procedures in place to
>minimize mishandling of the electronic parts... Ben Myers

http://www.caliburn.nl/topposting

>On Tue, 12 Feb 2008 19:22:21 -0800 (PST), ShadowTek <>
>wrote:
>
>>I was getting some persistent errors in my Pavilion A1130N. I ran
>>memtest for a few days and kept getting occasional errors on test
>>#7.So I took a look at everything and saw that there was some kind of
>>corrosion-looking substance on my 512Mb DDR 400, Micron
>>Mt8VDDT6464AG-40BDB, that came with the PC when I first bought it. I
>>can't tell exactly what the substance is, but it looks like a whitish
>>corrosion.
>>
>>This "substance" is in the suspicious pattern of what looks like a
>>thumb-print on one side and several fingers on the other. If they are
>>fingerprints, then it was not I, the sole owner, that did it, since I
>>have never even touched the stock modules before.
>>
>>Could this "stuff" just be the oils from someone's fingers that has
>>chemically changed over the years to become something that can cause a
>>short?
>>
>>I would just dismiss the apparent "finger pattern" as a coincidence,
>>except for the fact that there is a "large" mass of the "stuff" on one
>>part of the board that is far away from any corrodible metal.
>>
>>If this stuff is just some careless idiots finger crud, then would you
>>think that I could clean the module in some way and possibly get it
>>working reliably again? If so, then what kind of cleaning method would
>>you recommend?
>>
>>I also noticed that there is a similar, yet not as noticeable,
>>discoloration pattern on the other stock 512 module. I ran memtest
>>with it for several days but never got any errors. Do you think I
>>should just go ahead and toss it out? Or do you think it might be
>>salvageable as well?

Barry
=====
Home page
http://members.iinet.net.au/~barry.og

ShadowTek <> wrote:
>> how did you conclude "chemically degraded"?
>
> Well, that's my way of saying "corroded", since corrosion is the
> result of a chemical process. This is, assuming that this stuff is
> actually corrosion.
>
>> Is it possible the stuff you're seeing is heatsink grease, that somebody
>> maybe got all over their hands?
>>
>> distilled water is cheap and a powerful cleaner. Give it as shot.
>
> That could very well be what it is. But don't they usually use
> nonconductive grease like silicone? I thought that the only people who

not anymore. The standard white stuff, which is zinc oxide in a silicone
doesn't conduct. It can separate or dry out though. It's usually not very
running unless it hasn't beem mixed well. The stuff in your photos almost
looks more like a grease than heatsink compound.

> used conductive thermal interface material where overclockers who buy
> that Arctic Silver stuff.

The grey stuff is silver bearing, and while I've not tested it with a
meter, probably has some conductive properties. They tend to use this for
CPUs that use lots of power (basically anything these days).

One use thermal pads are the rage now. They print them onto the heatsinks.

> I mean, wouldn't the substance have to be conductive, causing a short,
> to cause a problem? For what other reasons could a foreign substance
> cause a RAM error, if not for causing a short?
>
> Also, I have a can of some electrical contact cleaner lying around.
> Would this be safe to use on RAM?
> http://www.crcindustries.com/auto/co....aspx?PN=05102


that stuff should remove grease or heatsink compound. Worst case, you
already have a dead DIMM with a value under $50 anways.


If I had any left, I'd use "blue shower" spray on that DIMM.