Tin whiskers in transistors other than the AF11x series. - Page 2 - UK Vintage Radio Repair and Restoration Discussion Forum

Philips210 · 29th Apr 2016, 8:32 pm

Hi.

I have experienced tin whiskers affecting an AC188 transistor from a Roberts R707. As has been pointed out, if the whiskers are to the case from either collector, base or emitter then the fault won't show up unless the case of the transistor is connected to anything. The AC188 in the R707 has an heat-sink fitted to it and this is bolted to the chassis.

Out of interest, has anyone come across tin whiskers in an npn device such as AC127 or AC187?

Regards
Symon

G6Tanuki · 29th Apr 2016, 8:44 pm

Quote:

Originally Posted by greg_simons

It's one of the strangest phenomenons I know of, nasa has published quite a lot of stuff about this, it even occurs in new unused semiconductors for satellites and such like.

Yes, it's a physical/chemical crystal-growth thing and it occurs irrespective of whether there is any applied electrical 'stimulus' to the transistor(s). 40-year-old new-old-stock unused AF11x transistors come ready-whiskered to trap the unenlightened when they use them as replacements!

Arguing whether it happens more on P-N-P or N-P-N transistors is akin to arguing whether positive- or negative-earth cars rusted faster (the answer is no).

Philips210 · 29th Apr 2016, 9:00 pm

Hi.

That's very true about old unused AF11x transistors being afflicted by the tin whisker problem. I have a few in my collection of NOS transistors.

Why is it that some really grotty looking and corroded used AF117s in my collection, taken from scrapped radios, appear to be absolutely fine whereas the clean looking NOS ones in plastic bags are affected?

Regards
Symon.

paulsherwin · 29th Apr 2016, 9:24 pm

What they look like on the outside makes no difference.

ukcol · 29th Apr 2016, 9:56 pm

Quote:

Originally Posted by ukcol

……… when I repeated the diode test on the AC128 I had removed the short circuit had gone!

Quote:

Originally Posted by Philips210

the fault won't show up unless the case of the transistor is connected to anything.

When I tested the removed AC128 I failed to check between the case and each electrode

.

I fished it out of the bin and tested it again. It measure less than 20 ohms between the base connection and the case

.

So it looks likely that this particular AC128 has a tin whisker problem.

Lucien Nunes · 29th Apr 2016, 10:09 pm

Sounds like it needs to be cut open...

ukcol · 29th Apr 2016, 10:13 pm

Any advice on that Lucien? Are there any nasty chemicals inside for example?

Philips210 · 29th Apr 2016, 10:24 pm

Hi.

We've been talking about cases of tin whiskers in European devices, notably Philips/Mullard but have there been any recorded cases of tin whiskers in the Japanese TO-1 style germanium transistors?

Symon.

Lucien Nunes · 29th Apr 2016, 11:38 pm

I'd turn it very slowly in a lathe, to avoid any metal particles getting in that might be mistaken for whiskers, while minimising centrifugal force that might collapse them. I don't think there's anything harmful inside.

paulsherwin · 30th Apr 2016, 1:02 am

I agree that there's unlikely to be anything toxic inside. For an amateur investigation you should be able to hold the transistor in a small vice and cut the top off with a Dremmel cutting disc or equivalent. This will contaminate the transistor, but it should be easy enough to distinguish between swarf/dust and tin whiskers using a microscope or even a strong hand lens.

CambridgeWorks · 30th Apr 2016, 4:27 am

Do not assume only old parts affected. I am aware of problems on tin plated screening, like often used on rf screening on pcb modules. I once saw one metal screen with whiskers up to abou 6 mm long on something that was manufactured about 15 years previous. All whiskers that I have seen usually grow at an angle away from the surface.
Some intermittent faults have been caused by these whiskers. Low current areas around cmos are easily affected as well. With the miniaturised equipment today, expect to see more problems.

turretslug · 30th Apr 2016, 8:12 am

Not to mention the use of unleaded solder- alloying with lead suppresses tin's tendency to grow whiskers. Critical applications were exempted but consumer items went unleaded. A cynic might venture that modern stuff has a short life-cycle anyway, so whiskering would assist in culling any survivors and make even the stubborn buy new things.

Philips210 · 30th Apr 2016, 9:10 am

Quote:

Originally Posted by G6Tanuki

Arguing whether it happens more on P-N-P or N-P-N transistors is akin to arguing whether positive- or negative-earth cars rusted faster (the answer is no).

Hi.

That's a fair point but it would still be interesting to find out if any npn devices have been found to be affected. From all the cases found so far they appear to have been pnp but then there's significantly less npn germanium devices than pnp types in existence anyway. Perhaps that's the reason why.

Regards
Symon.

mhennessy · 30th Apr 2016, 10:50 am

Quote:

Originally Posted by Philips210

but then there's significantly less npn germanium devices than pnp types in existence anyway. Perhaps that's the reason why.

I think you're dead right, Symon. Just statistics

But yes, I have seen the occasional NPN one...

dseymo1 · 30th Apr 2016, 11:24 am

It's difficult to see why there would be a difference between PNP and NPN devices. If I understand correctly, the actual semiconductor structure isn't invoved in the whisker growth.

ukcol · 30th Apr 2016, 12:01 pm

OK, I've opened up my suspect AC128 and pictures are shown below.

I am at the limits of my photographic skills and the technical spec of the camera but I think I can see whiskers.

paulsherwin · 30th Apr 2016, 12:17 pm

You may well be right. It can be difficult to spot them if the grease is opaque. It may help if you use a single LED torch to shine a bright light into it.

ukcol · 30th Apr 2016, 12:35 pm

Hi Paul,

I edited my post in which there is a better picture.

My reason for editing should read "I meant to show TWO pictures". But the 30 minute editing time has expired so I can't correct it.

mhennessy · 30th Apr 2016, 1:37 pm

That's an excellent couple of pictures, Colin - definitely whiskers there

I've had a few of these open themselves up, thanks to hairline cracks in the can - a gentle tug is all you need to pop the can off. Sadly, they didn't have whiskers, but I kept one to photograph (must get around to that!). I think the goo inside looks like silicone grease, but I can't really tell for sure. Amazing to think that it is the only medium to conduct the heat away from the junction...

ms660 · 30th Apr 2016, 2:23 pm

It might not be the end of the problems, tin is allotropic, it starts below 13 centigrade, white tin turns to pest tin, the resistivity of pest tin is many times higher than white tin so far as I'm aware.

I used to mine this stuff, the most abundant ore is Cassiterite (SnO2) When crushed it can be reduced with charcoal in the Rayburn (or owt else) to produce the metal, if you've got any lead kicking around it should be possible to make yer own solder

Lawrence.

29th Apr 2016, 8:32 pm	#21
Philips210 Nonode Join Date: May 2007 Location: Redruth, Cornwall, UK. Posts: 2,573	Re: Tin whiskers in transistors other than the AF11x series. Hi. I have experienced tin whiskers affecting an AC188 transistor from a Roberts R707. As has been pointed out, if the whiskers are to the case from either collector, base or emitter then the fault won't show up unless the case of the transistor is connected to anything. The AC188 in the R707 has an heat-sink fitted to it and this is bolted to the chassis. Out of interest, has anyone come across tin whiskers in an npn device such as AC127 or AC187? Regards Symon

29th Apr 2016, 9:00 pm	#23
Philips210 Nonode Join Date: May 2007 Location: Redruth, Cornwall, UK. Posts: 2,573	Re: Tin whiskers in transistors other than the AF11x series. Hi. That's very true about old unused AF11x transistors being afflicted by the tin whisker problem. I have a few in my collection of NOS transistors. Why is it that some really grotty looking and corroded used AF117s in my collection, taken from scrapped radios, appear to be absolutely fine whereas the clean looking NOS ones in plastic bags are affected? Regards Symon. Last edited by Philips210; 29th Apr 2016 at 9:07 pm. Reason: missing word

29th Apr 2016, 9:24 pm	#24
paulsherwin Moderator Join Date: Jun 2003 Location: Oxford, UK Posts: 27,943	Re: Tin whiskers in transistors other than the AF11x series. What they look like on the outside makes no difference.

29th Apr 2016, 10:09 pm	#26
Lucien Nunes Rest in Peace Join Date: Apr 2008 Location: London, UK. Posts: 2,508	Re: Tin whiskers in transistors other than the AF11x series. Sounds like it needs to be cut open...

29th Apr 2016, 10:13 pm	#27
ukcol Rest in Peace Join Date: May 2003 Location: Harlaxton, Lincolnshire, UK. Posts: 3,944	Re: Tin whiskers in transistors other than the AF11x series. Any advice on that Lucien? Are there any nasty chemicals inside for example?

29th Apr 2016, 10:24 pm	#28
Philips210 Nonode Join Date: May 2007 Location: Redruth, Cornwall, UK. Posts: 2,573	Re: Tin whiskers in transistors other than the AF11x series. Hi. We've been talking about cases of tin whiskers in European devices, notably Philips/Mullard but have there been any recorded cases of tin whiskers in the Japanese TO-1 style germanium transistors? Symon.

29th Apr 2016, 11:38 pm	#29
Lucien Nunes Rest in Peace Join Date: Apr 2008 Location: London, UK. Posts: 2,508	Re: Tin whiskers in transistors other than the AF11x series. I'd turn it very slowly in a lathe, to avoid any metal particles getting in that might be mistaken for whiskers, while minimising centrifugal force that might collapse them. I don't think there's anything harmful inside.

30th Apr 2016, 1:02 am	#30
paulsherwin Moderator Join Date: Jun 2003 Location: Oxford, UK Posts: 27,943	Re: Tin whiskers in transistors other than the AF11x series. I agree that there's unlikely to be anything toxic inside. For an amateur investigation you should be able to hold the transistor in a small vice and cut the top off with a Dremmel cutting disc or equivalent. This will contaminate the transistor, but it should be easy enough to distinguish between swarf/dust and tin whiskers using a microscope or even a strong hand lens.

30th Apr 2016, 4:27 am	#31
CambridgeWorks Nonode Join Date: Jun 2010 Location: Spalding, Peterborough, Cambridgeshire, UK. Posts: 2,858	Re: Tin whiskers in transistors other than the AF11x series. Do not assume only old parts affected. I am aware of problems on tin plated screening, like often used on rf screening on pcb modules. I once saw one metal screen with whiskers up to abou 6 mm long on something that was manufactured about 15 years previous. All whiskers that I have seen usually grow at an angle away from the surface. Some intermittent faults have been caused by these whiskers. Low current areas around cmos are easily affected as well. With the miniaturised equipment today, expect to see more problems. __________________ Apprehension creeping like a tube train up your spine - Cymbaline. Film More soundtrack - Pink Floyd

30th Apr 2016, 8:12 am	#32
turretslug Dekatron Join Date: Nov 2011 Location: Surrey, UK. Posts: 4,395	Re: Tin whiskers in transistors other than the AF11x series. Not to mention the use of unleaded solder- alloying with lead suppresses tin's tendency to grow whiskers. Critical applications were exempted but consumer items went unleaded. A cynic might venture that modern stuff has a short life-cycle anyway, so whiskering would assist in culling any survivors and make even the stubborn buy new things. Last edited by turretslug; 30th Apr 2016 at 8:18 am.

30th Apr 2016, 11:24 am	#35
dseymo1 Retired Dormant Member Join Date: Dec 2007 Location: Shropshire, UK. Posts: 3,051	Re: Tin whiskers in transistors other than the AF11x series. It's difficult to see why there would be a difference between PNP and NPN devices. If I understand correctly, the actual semiconductor structure isn't invoved in the whisker growth.

30th Apr 2016, 12:01 pm	#36
ukcol Rest in Peace Join Date: May 2003 Location: Harlaxton, Lincolnshire, UK. Posts: 3,944	Re: Tin whiskers in transistors other than the AF11x series. OK, I've opened up my suspect AC128 and pictures are shown below. I am at the limits of my photographic skills and the technical spec of the camera but I think I can see whiskers. Attached Thumbnails Last edited by Brian R Pateman; 30th Apr 2016 at 1:07 pm. Reason: I ment to show TWO pictures.

30th Apr 2016, 12:17 pm	#37
paulsherwin Moderator Join Date: Jun 2003 Location: Oxford, UK Posts: 27,943	Re: Tin whiskers in transistors other than the AF11x series. You may well be right. It can be difficult to spot them if the grease is opaque. It may help if you use a single LED torch to shine a bright light into it.

30th Apr 2016, 12:35 pm	#38
ukcol Rest in Peace Join Date: May 2003 Location: Harlaxton, Lincolnshire, UK. Posts: 3,944	Re: Tin whiskers in transistors other than the AF11x series. Hi Paul, I edited my post in which there is a better picture. My reason for editing should read "I meant to show TWO pictures". But the 30 minute editing time has expired so I can't correct it.

30th Apr 2016, 1:37 pm	#39
mhennessy Dekatron Join Date: Jul 2007 Location: Evesham, Worcestershire, UK. Posts: 4,244	Re: Tin whiskers in transistors other than the AF11x series. That's an excellent couple of pictures, Colin - definitely whiskers there I've had a few of these open themselves up, thanks to hairline cracks in the can - a gentle tug is all you need to pop the can off. Sadly, they didn't have whiskers, but I kept one to photograph (must get around to that!). I think the goo inside looks like silicone grease, but I can't really tell for sure. Amazing to think that it is the only medium to conduct the heat away from the junction...

30th Apr 2016, 2:23 pm	#40
ms660 Dekatron Join Date: Apr 2011 Location: Cornwall, UK. Posts: 13,454	Re: Tin whiskers in transistors other than the AF11x series. It might not be the end of the problems, tin is allotropic, it starts below 13 centigrade, white tin turns to pest tin, the resistivity of pest tin is many times higher than white tin so far as I'm aware. I used to mine this stuff, the most abundant ore is Cassiterite (SnO2) When crushed it can be reduced with charcoal in the Rayburn (or owt else) to produce the metal, if you've got any lead kicking around it should be possible to make yer own solder Lawrence.