Beware of old AntiStatic foam

[intabits]

Unforunately, I wasn't aware of this previously, and maybe others aren't also...

Among my most prized possions are my first two microprocessors - a pair of Signetics 2650s with ceramic packages and gold plated pins, that I bought new back in 1976.
I dug them out a few months ago, just to make sure that they hadn't been lost, but I didn't take a closer look at them until today.
Not only had the antistatic foam they had been stored in for 40+ years decayed to dust, it had corroded the pins of the chips away in the process. The gold plating was no obstacle.





I found some discussion of this:-
https://vintage-radio.net/forum/showthread.php?t=131907

So to save them, (after DeOxiting and cleaning them) I'm thinking of soldering them to  machined pin IC sockets like this:-



It looks like the pins on these ceramic packages are not part of the leadframe, but separate pieces that are welded(?) to the frame. I hope they are welded, rather than soldered, so that they won't come adrift when I solder them to the socket.

Before I do this, does anyone here have any better suggestions for saving these chips?
A PCB adapter board with header pins maybe? (they don't need to be plugged into any existing socket, so the pinout can be anything)

[Zero999]

I've seen it where I work. Have you tried cleaning them with isopropyl alcohol? An ordinary detergent may help, but make sure it's properly rinsed and dried, before storage, or usage.

Modern polyethylene anti-static foam doesn't decompose like this and there are polypropylene anti-static packages which are pretty good. If you can't get that, then how about putting the ICs in ordinary sockets, to protect the pins and packaging them in silver or black ESD safe bags?

[Refrigerator]

Old foam deteriorates and gets either dry and crumbly of wet and sticky, looks like yours was the latter.
You see this often in old cars where the headliner foam deteriorates as well as the foam suspension in speakers.
I think it's humidity that causes this.
Also make sure you plug the socket into a breadboard to keep the pins straight when soldering.

[RoGeorge]

Have you tried using a pencil-eraser to clean the pins?

It cleans almost everything, and it can be cut or pierced to clean any facet of almost any shape.

[Refrigerator]

Modern polyethylene anti-static foam doesn't decompose like this and there are polypropylene anti-static packages which are pretty good. If you can't get that, then how about putting the ICs in ordinary sockets, to protect the pins and packaging them in silver or black ESD safe bags?

Cleaning with vinegar might be an option as well i've seen people clean battery leakage (and other corrosion) from PCB's this way.

[tszaboo]

IPA, or ultrasonic cleaner is your friend.

[zener]

An ultrasonic cleaner is NOT your friend.   The ultrasonic vibrations can break the bonding wires that go between
the chip carrier and the silicon wafer.

[tszaboo]

An ultrasonic cleaner is NOT your friend.   The ultrasonic vibrations can break the bonding wires that go between
the chip carrier and the silicon wafer.

Got any evidence to back this up?

[Bud]

I believe it was Adrian's Digital Basement youtuber who had very good results cleaning a badly corroded PCB by soaking the affected areas with WD40 for a few hours, then brushing it.

[Berni]

Wait are those pins just corroded or did the thin pin part disappear completely? Or are those pins just bent and such to look like it.

EDIT: Sorry typo

[intabits]

Thanks for the responses.

Good idea about stabilizing the pins while soldering.

So far, I have only applied some Deoxit, and the black munge then seems to scrape off fairly easily.

Next, I'll use IPA and a toothbrush (if I had a girlfriend, I'd use hers) to get the munge off.
I'll try the eraser idea on any stubborn spots.

"Wait are those pins just corroded or did the thin pin big disappear completely?"
Not sure what you mean there, but most of the pins are definitely very, very corroded.

Do people agree that soldering to a socket as shown is the best way to make these devices usable again?
(I'll first test a socket to make sure the pin tops take solder well - might have to remove some plating if not)

[DrG]

Thanks for the responses.

Good idea about stabilizing the pins while soldering.

So far, I have only applied some Deoxit, and the black munge then seems to scrape off fairly easily.

Next, I'll use IPA and a toothbrush (if I had a girlfriend, I'd use hers) to get the munge off.
I'll try the eraser idea on any stubborn spots.

"Wait are those pins just corroded or did the thin pin big disappear completely?"
Not sure what you mean there, but most of the pins are definitely very, very corroded.

Do people agree that soldering to a socket as shown is the best way to make these devices usable again?

I saw this once, with a couple of chips that were in a plastic drawer - but only a couple of chips were affected. I am sorry that I can't remember precisely, but I don't think they were on static foam at all - If they were, I do not remember it being all degraded.

It puzzled me as to why a few chips would get corroded when others stored in the same environment did not. To be clear, I think you are on solid ground suspecting the degraded foam, but I just don't know.

Mine were in worse shape and portions of some pins were gone:



For the heck of it, I kludged the IC onto a socket and it did still work https://www.eevblog.com/forum/vintage-computing/sn76489-reincarnation/msg2989066/#msg2989066

Personally, if I had some ICs that were extremely valuable to me, I think I would treat the pins with something like 99% isopropyl alcohol or ethanol, to remove any corrosive chemicals that might be on the pins, and store them in a plastic tube. But I don't know this for a fact. For example, in numismatics people used to use certain kind of soft PVC "flips" until learning that they could interact and damage the coins - that is well known now.

If some one has some hard facts or "museum curator"-type practices, I would like to know also.

Edited to add: The ICs had been in the plastic drawer for 30+ years.

[tooki]

Unforunately, I wasn't aware of this previously, and maybe others aren't also...

Among my most prized possions are my first two microprocessors - a pair of Signetics 2650s with ceramic packages and gold plated pins, that I bought new back in 1976.
I dug them out a few months ago, just to make sure that they hadn't been lost, but I didn't take a closer look at them until today.
Not only had the antistatic foam they had been stored in for 40+ years decayed to dust, it had corroded the pins of the chips away in the process. The gold plating was no obstacle.





I found some discussion of this:-
https://vintage-radio.net/forum/showthread.php?t=131907

So to save them, (after DeOxiting and cleaning them) I'm thinking of soldering them to  machined pin IC sockets like this:-



It looks like the pins on these ceramic packages are not part of the leadframe, but separate pieces that are welded(?) to the frame. I hope they are welded, rather than soldered, so that they won't come adrift when I solder them to the socket.

Before I do this, does anyone here have any better suggestions for saving these chips?
A PCB adapter board with header pins maybe? (they don't need to be plugged into any existing socket, so the pinout can be anything)

Oh dear, that’s rough!!

At work, the component drawers are full of foam that’s decaying, but there’s no apparent corrosion. What were these chips, and their foam, stored in, and under what conditions? I suspect that there are additional factors beyond age that come into play to cause this level of catastrophic corrosion.

[magic]

Beware that inserting the bottom pins of a female socket into another female socket will ruin the latter.

I would look for a similar frame with real male pins (if such things even exist) or create one by transplanting pins. Half an hour of boring work and you're in. Well, all of that assuming that the ICs are ever meant to be used.

[drussell]

Do people agree that soldering to a socket as shown is the best way to make these devices usable again?

Too bad you don't have access to whatever apparatus the Chinese recyclers use to weld on and polish up new pins on salvaged components when they sell all those cleaned and refurbished, sometimes remarked, sometimes even fake chips.   

[amyk]

It looks like the pins on these ceramic packages are not part of the leadframe, but separate pieces that are welded(?) to the frame. I hope they are welded, rather than soldered, so that they won't come adrift when I solder them to the socket.

They might be soldered with high-temperature solder. On the bright side, if they do come off, you're left with a DFN-40

[Berni]

Wow it does completely eat away the thin part of the pin.

Would have never thought that decomposing antistatic foam could be this corrosive. Even most RTC battery leaks are not as aggressive as this.

[intabits]

What were these chips, and their foam, stored in, and under what conditions?

They were stored in a parts draw cabinet. Admittedly, in a room exposed more to the outside air, so with higher humidity.
There's no other black foam-mounted parts in the drawers to compare with, but other parts in the drawers show no similar corrosion, and other chips stored in aluminium foil over styrofoam in the same environment are all OK. So it seems the foam (possibly plus humidity) caused it.

Beware that inserting the bottom pins of a female socket into another female socket will ruin the latter.

I wasn't aware of that. Also, the pins on those sockets are easily broken off when straightening a bent one. The leftmost one in the photo is already bent.

I don't need to insert them into any standard socket, I just want to connect them up to some sort of demo/display system.

So I'm starting to lean towards the idea of a PCB with header pins, like an Arduino shield.
I'd just cut off the thin parts of all the pins, and just "surface mount" the chips onto traces going to the header strips.
If the headers become damaged, they can be replaced without ever touching the chips. And removing the chips would be a little easier if that ever became necessary. And while I'm at it, put basic items like bypass caps & a crystal on there as well...

[magic]

They simply are thicker than male pins that normally go into those sockets.
Also, male pins have two ends and the wrong end is too thick as well.
Surprising complexity for such a seemingly simple connector

[Ian.M]

This paper on gold/tin solder bond strength may be of interest: https://www.osti.gov/servlets/purl/1142490

Look at Aries 40-600-10 flat pin headers which are readily available.  If you can find stock, Aries 40-600-11 would be even nicer as that's the gold plated version.  If the old pins were cropped, sanded and lightly tinned and the inside face of the fork terminals on the header was similarly tinned, after they were formed to fit the slight package over-width due to the old pin remnants, it should be possible to extract the individual pins, put them in a jig (e.g. a turned pin socket that's been oversized by reaming with a 0.55mm HSS drill bit in a pin vise) for alignment, and sweat them onto the well fluxed pin stubs.

[intabits]

Thanks for the reference and detailed possible solution. It does seem a bit fiddly though, and like the machined socket idea, still involves direct connection of chip pins to socket pins, which in both cases are fairly delicate. Any pins getting damaged would mean re-subjecting the chip to soldering stresses.

As mentioned, there is no need to make these able to be plugged into any standard DIP40 receptacle, so that's why I'm pretty much settled on the PCB adapter idea, which should firmly and safely mount the chip once and for all, while separating it from any possible damaged pin issues (and the header pins are stronger anyway).

A manufactured PCB with white (or black?) soldermask, and silkscreen printing showing pin designations  would make a nice centrepeice in a 2650 demo/display project. Some bypass caps (but no crystal) can also be fitted to the board.

Thanks to all who responded.

[rsjsouza]

Thanks for the heads up, and I am very sorry this happened to you.

I have a few ancient EPROMs and some TO-18 ICs that suffered corrosion over the years. I will try the Deoxit and WD-40 to see what good can come up from it.

Having moved to US I now have good access to high quality antistatic foam, but historically we did not have access to this. My dad's solution was to carefully discharge himself and then wrap the devices in aluminium foil on all his CD4000 devices and some have been stored in perfect condition for more that twenty years. I have several CMOS (and some NMOS) processors and memory stored this way as well. And yes, all them still work.

This method is annoying and does not hold if you use the packaging regularly (the foil tears easily), thus I will probably keep putting most of the devices on foam but keeping a close eye as time goes on. One aspect is that I created lots silica-gel bags using coffee filters and use them copiously on the boxes where I store these parts. Hopefully this reduces the time to decay.

[drussell]

One aspect is that I created lots silica-gel bags using coffee filters and use them copiously on the boxes where I store these parts. Hopefully this reduces the time to decay.

Don't you mean increases the time to decay?   

[Ian.M]

I don't like the direct to board mounting idea - the coefficients of thermal expansion of FR4 and ceramic are very different, and there's no flex in the arrangement so you risk ripping the metalization off the package either by temperature extremes or the board flexing, at which point its game over.  Originally the pins flexing could accommodate the thermal expansion difference.

Also, that metalization is pretty thin - you really don't want to get non-gold alloy solder  anywhere except the faces of the pin remnants as its probably severely vulnerable to leeching. (Similar to the vintage Tektronics silver on ceramic terminal strip problem.) 

Therefore I'd suggest  using a header even if you reject my loose pins idea.  You'll probably have to splay its terminals slightly or crop and file the top pins and solder with a fillet to the remaining flats.

[DrG]

/---/
They were stored in a parts draw cabinet. Admittedly, in a room exposed more to the outside air, so with higher humidity.
There's no other black foam-mounted parts in the drawers to compare with, but other parts in the drawers show no similar corrosion, and other chips stored in aluminium foil over styrofoam in the same environment are all OK. So it seems the foam (possibly plus humidity) caused it.

That situation (other chips stored right next to the chips that corroded) was also what I saw. It begs the questions as to whether the foam was the culprit. As I said before, I would also point to the foam since it is breaking down. Were there any of the same chips (2650s) nearby but not in the foam and not corroded?

Other possibilities are contaminants on those chips only and also the possibility that it is a manufacturing issue. True, you are more interested in what to do now rather than how it happened.

I wonder if these old cases would work for your display edit [your pcb would look better] but, at least for storage?

https://chipscapes.com/collections/just-the-chips/products/intel-8008-the-first-8-bit-microprocessor-c8008