Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Chip failure and air corrossion products

82 views
Skip to first unread message

N_Cook

unread,
Mar 3, 2017, 3:36:43 AM3/3/17
to
Anyone any insight on this. Failure of 34 year old TTL chip, in
unpowered storage. Just 1 of 50 chips on a board, TTL,4000, analogue,
memory etc.
Only this one had silver tarnish looking appearance of black silver
sulphide on the pins. This make of TTL did for a while have silver
plating aded to the pin frame , for a period.
I cracked open the device and no obvious tarnish on internal frame
metal. Anyone know what the failure process was? was silver plating
found to be problematic?

N_Cook

unread,
Mar 3, 2017, 7:16:47 AM3/3/17
to
Silver tarnish appearance of that sort of granular/cellular/mottling
effect of silver tarnish rather than the very black soot-like appearance
of CuS.

pf...@aol.com

unread,
Mar 3, 2017, 8:34:44 AM3/3/17
to
Most silver plating is "fine" silver. And proper plating is quite complex starting with a copper layer, then the silver layer. Back in the day when coal was the main fuel of choice, and there was no differentiation between high and low sulphur coal, plate-silver used to tarnish (not corrode) quickly due to high levels of SO2 in the air. Hence the references in Victorian times to "polishing the silver" as a full time job. Also back in the day, the underlying materials were often brass or bronze, so quite stable in itself. Further, electroplating was not yet an established science so "plate" was made using a dipping process with sterling silver. Sterling silver is far more prone to tarnish than fine silver, and, why the 'very best' plate used fine-silver for surfaces exposed to food or drink. It does not tarnish under normal conditions.

Today, the underlying materials might be anything from steel to brass to copper to many other things. The copper layer is still necessary for the silver to adhere, and the silver coating will be the minimum amount absolutely necessary. The silver will be fine silver from this method and very resistant to tarnish.

So, here is what happens. As the plating ages, especially if there are heat/cool cycles, the silver coating will get tiny holes - right through the copper to the base material. Now, electrolysis can occur as O2 and moisture gets between the copper and the base material. This causes tiny flaking (granular) such as you describe. At this point, the silver actually makes things worse by using the copper as a sacrificial electrode.

At home, take a common nail and a bit of copper wire. Wrap the wire around the nail and drop both into a glass of tap water. Give it a week.

If you have a bit of silver wire (must be fine silver), do the same with both wrapped around the nail.

Peter Wieck
Melrose Park, PA

Sjouke Burry

unread,
Mar 3, 2017, 12:14:18 PM3/3/17
to
Silver oxide conducts.
So, it could have been enough to clean to body of the ttl chip.

pf...@aol.com

unread,
Mar 3, 2017, 12:41:17 PM3/3/17
to
a) Silver Oxide will not form in this case, as copper is sacrificial to silver.
b) Copper Oxide (Cu0) is a semi-conductor, and is also black as described, and under the condition described would have looked like soot. But, as a semi-conductor would have led to all kinds of erratic behavior.

But, that does not take away from your correct conclusion that it would have been enough (for now) just to clean. However, as long as there was now-exposed copper, the process would have continued until all the copper was consumed. By then, all the silver would have been dissipated as well.

N_Cook

unread,
Mar 3, 2017, 1:21:14 PM3/3/17
to
All this is one the external sections of the pins? how would these
effects progress into the active die or at least weld-spots ?
This piece of kit was not stored i na damp environment, so no chance of
capilliary action.

pf...@aol.com

unread,
Mar 3, 2017, 1:44:21 PM3/3/17
to
On Friday, March 3, 2017 at 1:21:14 PM UTC-5, N_Cook wrote:

> All this is one the external sections of the pins? how would these
> effects progress into the active die or at least weld-spots ?
> This piece of kit was not stored i na damp environment, so no chance of
> capilliary action.

This is all happening at the chemical level by electrolysis. So the corrosion would follow the path as the copper starts being exposed. Clearly the silver-over-copper extends into the body of the die - and the same oxides would form within. Once the process is established, no moisture is really necessary, just oxygen and time. You have to understand that the galvanic action is made much worse by the presence of silver making copper act as a sacrificial electrode to the silver reacting with anything attacking it and whatever the base metal might be.

The thickness of the coating - at least several molecules - is plenty wide enough, if you will excuse the expression.

Benderthe.evilrobot

unread,
Mar 3, 2017, 3:01:59 PM3/3/17
to

"N_Cook" <div...@tcp.co.uk> wrote in message
news:o9b9r3$71c$1...@dont-email.me...
There have been problems in the past where gold plated IC pins have
developed a microscopic insulating layer - but I don't know about silver
plated.

Silver plating develops black silver sulphide after long exposure to air,
with NOS parts you'd just clean it off and solder as normal. Silver is a
component of several types of solder alloys. So I don't think it should
cause any problems.

pf...@aol.com

unread,
Mar 3, 2017, 3:28:17 PM3/3/17
to
Mpfffff..... Some clarification required.

FINE silver plating (most of it) *will not* tarnish from exposure to air, nor will it form sulphur compounds - without other contributing factors. Sterling silver (.925 silver) is an entirely different matter.

What is going on here electro-galvanic (galvanic action) *corrosion* of the copper plating under the silver - and most things electro-silver-plated other than brass are first copper-plated to guarantee proper adhesion.

The black material noted is CuO, not tarnished silver.

John Robertson

unread,
Mar 3, 2017, 3:51:17 PM3/3/17
to
Thanks, we've had many an Eprom with corroded legs that are from the
70s, in some cases the legs break off inside the IC sockets when the
EPROM is removed. Also this is common on some other ICs used in the
early 80s, such as Galaga which has Namco custom ICs, where again the
legs are corroding right off them.

Is there any way to arrest the process? Be nice to have something to dip
the legs into to cancel the corrosion, similar to using white vinegar
and water (50/50) to neutralize alkaline battery leakage corrosion.

John :-#)#

--
(Please post followups or tech inquiries to the USENET newsgroup)
John's Jukes Ltd. 2343 Main St., Vancouver, BC, Canada V5T 3C9
(604)872-5757 or Fax 872-2010 (Pinballs, Jukes, Video Games)
www.flippers.com
"Old pinballers never die, they just flip out."

pf...@aol.com

unread,
Mar 3, 2017, 4:05:25 PM3/3/17
to
On Friday, March 3, 2017 at 3:51:17 PM UTC-5, John Robertson wrote:

> Thanks, we've had many an Eprom with corroded legs that are from the
> 70s, in some cases the legs break off inside the IC sockets when the
> EPROM is removed. Also this is common on some other ICs used in the
> early 80s, such as Galaga which has Namco custom ICs, where again the
> legs are corroding right off them.
>
> Is there any way to arrest the process? Be nice to have something to dip
> the legs into to cancel the corrosion, similar to using white vinegar
> and water (50/50) to neutralize alkaline battery leakage corrosion.
>
> John :-#)#


The best way I have found - if at all possible - is a coat of flexible lacquer over the entirety. I admit to having a bottle of flexible fingernail polish hidden away amongst my other solvents for this sort of thing. Stop the air, stop the corrosion.

Benderthe.evilrobot

unread,
Mar 3, 2017, 4:07:38 PM3/3/17
to

"John Robertson" <sp...@flippers.com> wrote in message
news:J_OdnT_d4oCiSiTF...@giganews.com...
> On 2017/03/03 12:28 PM, pf...@aol.com wrote:
>> Mpfffff..... Some clarification required.
>>
>> FINE silver plating (most of it) *will not* tarnish from exposure to air,
>> nor will it form sulphur compounds - without other contributing factors.
>> Sterling silver (.925 silver) is an entirely different matter.
>>
>> What is going on here electro-galvanic (galvanic action) *corrosion* of
>> the copper plating under the silver - and most things
>> electro-silver-plated other than brass are first copper-plated to
>> guarantee proper adhesion.
>>
>> The black material noted is CuO, not tarnished silver.
>>
>> Peter Wieck
>> Melrose Park, PA
>>
>
> Thanks, we've had many an Eprom with corroded legs that are from the 70s,
> in some cases the legs break off inside the IC sockets when the EPROM is
> removed. Also this is common on some other ICs used in the early 80s, such
> as Galaga which has Namco custom ICs, where again the legs are corroding
> right off them.
>
> Is there any way to arrest the process? Be nice to have something to dip
> the legs into to cancel the corrosion, similar to using white vinegar and
> water (50/50) to neutralize alkaline battery leakage corrosion.

No idea whether its related - but when I serviced VGA monitors, I often
encountered brown glue that was probably applied to prevent components from
fatiguing their leads.

This stuff eventually became brittle and crumbly - and highly corrosive. It
frequently etched away component leads.

Once brittle it was fairly easy to chip away, but in the early stages it was
elastic like contact adhesive, and lighter brown like Evo-stik.

Very often a component like a resistor would come out encased in this stuff
and the leads completely eaten away.

pf...@aol.com

unread,
Mar 3, 2017, 4:40:47 PM3/3/17
to
Yep. "Elephant Snot" turned out to be a very bad idea over time. Much like foam speaker surrounds of recent memory.

I use an acid-free neutral-formula flexible hot-melt for the purpose. I source from AdTech.

N_Cook

unread,
Mar 3, 2017, 5:07:01 PM3/3/17
to
I've quite often found problems with conformal coating, over decades, it
seems to trap moisture underneath as though hygroscopic/delequescent or
whatever its called

N_Cook

unread,
Mar 3, 2017, 5:07:55 PM3/3/17
to
On 03/03/2017 21:08, Benderthe.evilrobot wrote:
>
sorry wrong followup.
You failed to mention it going conductive

N_Cook

unread,
Mar 3, 2017, 5:11:03 PM3/3/17
to
I've quite often found problems with conformal coating, over decades, it
seems to trap moisture underneath as though hygroscopic/delequescent or
whatever its called , an oily/ jelly type composition beneath, complete
with green copper carbonate corrossion product

Phil Allison

unread,
Mar 4, 2017, 12:19:22 AM3/4/17
to
Sjouke Burry wrote:
>
> >
> Silver oxide conducts.
>

** You got proof of that ?

In any case, the black tarnish on silver exposed to air is Silver Sulphate - which is an insulator.


.... Phil

Phil Allison

unread,
Mar 4, 2017, 12:25:50 AM3/4/17
to
Phil Allison wrote:

>
>
> In any case, the black tarnish on silver exposed to air is Silver Sulphate - which is an insulator.
>
>

** That should be "Silver Sulphide" = Ag2S


.... Phil

pf...@aol.com

unread,
Mar 4, 2017, 7:40:35 AM3/4/17
to
On Saturday, March 4, 2017 at 12:25:50 AM UTC-5, Phil Allison wrote:

>
> ** That should be "Silver Sulphide" = Ag2S
>
>
> .... Phil

True for sterling (.925) silver.
Fine silver - what is deposited by electroplating - does not form compounds with sulphur at anything near normal room temperatures.

Phil Hobbs

unread,
Mar 4, 2017, 11:10:45 AM3/4/17
to
Parylene is about the most impervious common coating, but any plastic
has orders of magnitude higher gas diffusion rates than any metal. In a
humid environment, water will penetrate until the net flux is zero, so
if you have nasty hygroscopic corrosive crud underneath, your board is
still doomed.

Cheers

Phil Hobbs


--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

Benderthe.evilrobot

unread,
Mar 4, 2017, 4:07:12 PM3/4/17
to

<pf...@aol.com> wrote in message
news:342bcce6-1434-4f63...@googlegroups.com...
> Yep. "Elephant Snot" turned out to be a very bad idea over time. Much like
> foam speaker surrounds of recent memory.
>
> I use an acid-free neutral-formula flexible hot-melt for the purpose. I
> source from AdTech.

Some manufacturers use hot melt - but some power components get hot enough
to melt it.

A range of IBM VGA monitors had clusters of tall electrolytics in the PSU
with blobs of RTV joining the tops together.

I'm not entirely convinced it served the intended purpose and it was a PITA
cutting it away when the caps had to be replaced and/or tested.

Benderthe.evilrobot

unread,
Mar 4, 2017, 4:08:30 PM3/4/17
to

"N_Cook" <div...@tcp.co.uk> wrote in message
news:o9cpc2$er4$2...@dont-email.me...
That was pretty much implied by it becomes corrosive.

Benderthe.evilrobot

unread,
Mar 4, 2017, 4:13:45 PM3/4/17
to

"Phil Allison" <palli...@gmail.com> wrote in message
news:1879ee73-2e28-400f...@googlegroups.com...
It would be sulphate if exposed to sulphuric acid.

The sulphide is mainly produced by a component of car exhaust - or your
farts if you've been eating eggs.

Catalytic converters stop your silverware going black (or make it go black
even quicker - I can't remember which) - and kill thousands of asthma
sufferers.

Phil Allison

unread,
Mar 4, 2017, 9:05:45 PM3/4/17
to
pf...@aol.com wrote:
>
> > ** That should be "Silver Sulphide" = Ag2S
> >
> >
>
> True for sterling (.925) silver.
> Fine silver - what is deposited by electroplating - does not form
> compounds with sulphur at anything near normal room temperatures.
>
>

** As you seem to be the only person in the world who thinks that - you need to post a link to back it up.

Do you know what "black leg" corrosion is ?

Commonly seen on the leads of small signal transistors made in Japan in the 70s. It often results in the device going noisy or failing completely. Marantz stereo amps were notorious for it.



.... Phil





John Robertson

unread,
Mar 4, 2017, 9:29:18 PM3/4/17
to
> ..... Phil
>

Hey Phil,

Can you provide a link to more in depth discussion on your black leg
corrosion? I did a quick search and could only find one reference, and
it merely said watch out for 'black leg corrosion'?

I have run into a lot of Namco branded ICs from Japan, made in the late
70s that have this problem...

Thanks!

Phil Allison

unread,
Mar 4, 2017, 10:17:05 PM3/4/17
to

John Robertson wrote:
>>
> >>> ** That should be "Silver Sulphide" = Ag2S
>
> >
> > Do you know what "black leg" corrosion is ?
> >
> > Commonly seen on the leads of small signal transistors made in Japan in the 70s. It often results in the device going noisy or failing completely. Marantz stereo amps were notorious for it.
> >
>
> Can you provide a link to more in depth discussion on your black leg
> corrosion? I did a quick search and could only find one reference, and
> it merely said watch out for 'black leg corrosion'?
>

** Not much to be found on the net, but something most service techs have seen.

Bit like the "yellow glue" problem that has plagued countless items and still is.


> I have run into a lot of Namco branded ICs from Japan, made in the late
> 70s that have this problem...
>

** It's got be due to silver plating the leads.



.... Phil

John Robertson

unread,
Mar 5, 2017, 1:43:05 AM3/5/17
to
I'm sure they thought that silver plating was a good idea, instead of
planned obsolescence.

>
>
> ..... Phil
>
Thanks,

pf...@aol.com

unread,
Mar 5, 2017, 7:57:36 AM3/5/17
to
No, Phil, it is not *DUE* to the silver plating on the leads, nor is the silver plating on the leads even contributory to the primary failure, though it does contribute to the visual results.

Let me start over as you clearly need the little-steps-for-little-feet approach.

a) Silver electroplating does not adhere well to anything but copper, low-tin bronze and very low zinc brass.
b) So, most materials to be silver-plated are first flash-plated with pure copper. Otherwise the failure of the silver-plate due to simple peeling would be quick and fast.
c) Over time, the different coefficients of expansion, simple age, heat/cool cycles and other environmental factors cause the copper plating to crack, microscopically, initially.
d) At which point, electrolysis takes place between the copper and base-metal of the lead/pin/whatever.
e) The first product of failure is CuO - because typically the substrate (base metal) would be sacrificial to the copper, in this case, the copper becomes sacrificial to the silver.
f) CuO is a black semi-conductor, still (very rarely) used in very heavy duty rectifiers.
g) Silver reactions with sulphur do not produce a grainy material, Even extensive reactions.
h) Copper oxides, with the fine silver still attached, are grainy. The fine silver will be entirely covered by the copper oxide, but will not actually react itself.

So, that you state something *MUST* be due to something else does not make it so. Jewelers and metalsmiths have been exploiting this phenomenon for many years and it is very well understood. You need to learn that there is an entire world outside your cave.

Phil Allison

unread,
Mar 5, 2017, 8:04:22 AM3/5/17
to
pf...@aol.com wrote:


> >
> > ** It's got be due to silver plating the leads.
> >
>
> No, Phil, it is not *DUE* to the silver plating on the leads,
> nor is the silver plating on the leads even contributory to the
> primary failure, though it does contribute to the visual results.
>


** Shame you have NO links whatever to back up this arrant nonsense.


> Let me start over as you clearly need the little-steps-for-little-feet
> approach.


** Go fuck yourself - asshole.

Wieck is a psychotic, raving lunatic.

I snipped his manic ravings.

To spare gentle the readers here.




.... Phil

pf...@aol.com

unread,
Mar 5, 2017, 8:53:05 AM3/5/17
to
file:///C:/Users/Peter/Downloads/corrosion_of_silver_plated_copper_conductor.pdf

Phil:

It's a big world out there - first link I found, and pretty much describes the situation nicely.

But, I am sure you will spin it in your own inimitable manner.

Michael A. Terrell

unread,
Mar 5, 2017, 12:17:50 PM3/5/17
to
pf...@aol.com wrote:
> file:///C:/Users/Peter/Downloads/corrosion_of_silver_plated_copper_conductor.pdf


Come on, Peter. Another link to your C: drive?


--
Never piss off an Engineer!

They don't get mad.

They don't get even.

They go for over unity! ;-)

pf...@aol.com

unread,
Mar 5, 2017, 3:15:53 PM3/5/17
to
https://escies.org/download/webDocumentFile?id=62170

See if this works - or use the title:

Corrosion of Silver Plate Copper Conductor

If one hits the link, it automatically downloads the .pdf, making a C: Drive link - which I did not notice.

Benderthe.evilrobot

unread,
Mar 5, 2017, 4:52:37 PM3/5/17
to

"John Robertson" <sp...@flippers.com> wrote in message
news:LbadnR13oZ3vLibF...@giganews.com...
> On 2017/03/04 7:17 PM, Phil Allison wrote:
>>
>> John Robertson wrote:
>>>>
>>>>>> ** That should be "Silver Sulphide" = Ag2S
>>>
>>>>
>>>> Do you know what "black leg" corrosion is ?
>>>>
>>>> Commonly seen on the leads of small signal transistors made in Japan in
>>>> the 70s. It often results in the device going noisy or failing
>>>> completely. Marantz stereo amps were notorious for it.
>>>>
>>>
>>> Can you provide a link to more in depth discussion on your black leg
>>> corrosion? I did a quick search and could only find one reference, and
>>> it merely said watch out for 'black leg corrosion'?
>>>
>>
>> ** Not much to be found on the net, but something most service techs have
>> seen.
>>
>> Bit like the "yellow glue" problem that has plagued countless items and
>> still is.
>>
>>
>>> I have run into a lot of Namco branded ICs from Japan, made in the late
>>> 70s that have this problem...
>>>
>>
>> ** It's got be due to silver plating the leads.
>>
>
> I'm sure they thought that silver plating was a good idea, instead of
> planned obsolescence.

My first job was fault finding huge desk calculators that only did the most
basic functions and contained about 200 DTL chips.

Most of the chips had silver plated pins and I never had a fault because of
that - a few chips had gold plated pins, there may have been less than
half-dozen faults because the alloy interface became an insulator (there was
no inherently dodgy RoHS soldering back then).

You could always tell the silver plated ones because the pins were black.
The tinned ones never got past dull grey.

Benderthe.evilrobot

unread,
Mar 5, 2017, 4:55:04 PM3/5/17
to

"Michael A. Terrell" <mike.t...@earthlink.net> wrote in message
news:3_OdnerXncqr1SHF...@earthlink.com...
> pf...@aol.com wrote:
>> file:///C:/Users/Peter/Downloads/corrosion_of_silver_plated_copper_conductor.pdf
>
>
> Come on, Peter. Another link to your C: drive?

Couldn't open that - maybe someone formatted it.......................

Phil Allison

unread,
Mar 5, 2017, 8:28:01 PM3/5/17
to
pf...@aol.com wrote:
>
> https://escies.org/download/webDocumentFile?id=62170
>
> See if this works - or use the title:
>
> Corrosion of Silver Plate Copper Conductor
>

** The link is about another topic entirely.

It neither supports or contradicts what has been said here.

The Wieck troll had no basis for his mad asssetions.

It's just his OWN theory.

Fuck off you bullshitting idiot.


.... Phil

pf...@aol.com

unread,
Mar 5, 2017, 8:50:56 PM3/5/17
to
What you missed twit, is that the silver is not contributory to the corrosion. That copper is sacrificial to silver, and that the cause is air/water getting to the copper due to cracks/failure in the plating.

Again, this process is well understood and often exploited by jewelers and metalsmiths. It's a big world out there, jackass - try learning something.

Phil Allison

unread,
Mar 5, 2017, 10:31:54 PM3/5/17
to
pf...@aol.com wrote:

> What you missed twit,



** Fraid you missed the entire point.

Cos you are an absolute, fucking nut case.

Piss OFF you damn TROLL.



..... Phil


pf...@aol.com

unread,
Mar 6, 2017, 7:06:40 AM3/6/17
to
Forgot to add:

a) NOT due to the silver.
b) Sulphur and sulphur compounds are not involved at all.

N_Cook

unread,
Mar 6, 2017, 11:20:06 AM3/6/17
to
On 03/03/2017 18:44, pf...@aol.com wrote:
> On Friday, March 3, 2017 at 1:21:14 PM UTC-5, N_Cook wrote:
>
>> All this is one the external sections of the pins? how would these
>> effects progress into the active die or at least weld-spots ?
>> This piece of kit was not stored i na damp environment, so no chance of
>> capilliary action.
>
> This is all happening at the chemical level by electrolysis. So the corrosion would follow the path as the copper starts being exposed. Clearly the silver-over-copper extends into the body of the die - and the same oxides would form within. Once the process is established, no moisture is really necessary, just oxygen and time. You have to understand that the galvanic action is made much worse by the presence of silver making copper act as a sacrificial electrode to the silver reacting with anything attacking it and whatever the base metal might be.
>
> The thickness of the coating - at least several molecules - is plenty wide enough, if you will excuse the expression.
>
> Peter Wieck
> Melrose Park, PA
>

Any connection with metalisation creep ? I associate with silver plating
of the segments of ceramic resonators, the thin segment gets metal
migrating over the edge and going ohmic and failure of RF/IF stages.

pf...@aol.com

unread,
Mar 6, 2017, 11:37:31 AM3/6/17
to
On Monday, 6 March 2017 11:20:06 UTC-5, N_Cook wrote:
>
> Any connection with metalisation creep ? I associate with silver plating
> of the segments of ceramic resonators, the thin segment gets metal
> migrating over the edge and going ohmic and failure of RF/IF stages.

Absolutely a connection. Silver-Mica disease in vintage radios is one obvious manifestation of such creep.

A very simple explanation here:

http://www.gdsiswitches.com/silver-migration.html

As silver is an excellent conductor, it takes very little for a current path to develop.

Although this is not absolutely specific to your question - the phenomenon and mechanisms are the same.

It is interesting to note that what is common to all these various discussions around silver and how it affects most electronics does not involve Sulphur, or actual corrosion of the silver itself.
0 new messages