|
Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets. |
|
Thread Tools |
17th Dec 2013, 12:58 pm | #1 |
Heptode
Join Date: Jul 2012
Location: Derby DE1, Derbyshire, UK.
Posts: 626
|
Tin whiskers in transistors
Hi,
I've looked at most of the information on here about all the research that was done by NASA, regarding the tin whiskers in certain transistors, causing internal shorts presumably bcause the inside of the transistor shell is lined with tin. But is there a specific reason as to actually why they grow and what is causing it? That's what I'm still struggling to grasp. Has anyone got any further thoughts on this now?. Thanks, Paul. |
17th Dec 2013, 1:40 pm | #2 |
Heptode
Join Date: Aug 2006
Location: Sheffield, South Yorkshire, UK
Posts: 990
|
Re: Tin whiskers in transistors
Hi Paul,
There must be a specific reason as, for example the AF11x series which is usually quoted with this problem, was replaced with the electrically identical but physically smaller AF12x series which isn't affected. So the cause of the whisker growth must be understood by the manufacturer at least. Like you, I'm not sure what they actually changed between the two series of transistors to prevent the problem? Regards |
17th Dec 2013, 2:07 pm | #3 | |
Dekatron
Join Date: Nov 2003
Location: Fakenham, Norfolk, UK.
Posts: 4,246
|
Re: Tin whiskers in transistors
Quote:
Paul |
|
17th Dec 2013, 3:13 pm | #4 |
Dekatron
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 4,924
|
Re: Tin whiskers in transistors
I recently stripped a 1960's Pye Vanguard RT. Everyone of the NKT200 transistors had shorts from collector to case, but not one of the GET800 series did. The latter have a case which is copper/gold in colour. I still have to find out whether the NKT's are working, and if not, can be retrieved by zapping them.
|
17th Dec 2013, 3:19 pm | #5 |
Dekatron
Join Date: Mar 2008
Location: Worthing, West Sussex, UK
Posts: 5,185
|
Re: Tin whiskers in transistors
This tin whisker problem is not just confined to transistors, I have had this problem with potentiometers.
I repaired the problem by painting the inside of the casing, no more problems since. Mark |
17th Dec 2013, 4:08 pm | #6 |
Heptode
Join Date: Jul 2008
Location: Selby, North Yorkshire, UK.
Posts: 979
|
Re: Tin whiskers in transistors
My understanding (from the articles published) is that it's down to the purity of tin coatings. If various other elements are present (most notably lead), then the crystals do not grow, or grow very slowly so do not cause the issues. It must be pretty much down to the choice of materials in construction.
|
17th Dec 2013, 5:44 pm | #7 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,801
|
Re: Tin whiskers in transistors
It's the presence of tin and an electric field which seem to be necessary, and the absence of a number of inhibiting substances. Going lead free has stirred problems.
David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
17th Dec 2013, 5:58 pm | #8 |
Octode
Join Date: Nov 2005
Location: Surrey, UK.
Posts: 1,869
|
Re: Tin whiskers in transistors
I would suggest that as well as the tin and the e-field, the dendritic growths are due to the presence of the silicone grease filling in the can.
Does anyone remember growing "crystal gardens" in icinglass (sp?) egg preserver? As far as I am aware, this substance is a silicone compound, ie, a silicon - oxygen chain with an organic radical attached. As I am now at the limit of my "O" level Chemistry, I would be very interested to hear from a specialist . Leon. |
17th Dec 2013, 6:52 pm | #9 |
Nonode
Join Date: Dec 2012
Location: Kirk Michael, Isle of Man
Posts: 2,346
|
Re: Tin whiskers in transistors
Water Glass is a sodium silicate. The soda to silica ratio varies according to use. Used extensively in pottery casting slips. Preserving eggs during WW2 obviously had some use. I remember we had a stoneware container used for just that (70 years ago).
Les. |
17th Dec 2013, 7:17 pm | #10 |
Octode
Join Date: Nov 2005
Location: Surrey, UK.
Posts: 1,869
|
Re: Tin whiskers in transistors
Les, you're absolutely right. I confused isinglass (checked spelling) with water glass which is, as you point out liquid sodium silicate.
Any specialist electro-chemists out there who would care to comment on dendritic tin growth in silicone grease? Leon. |
17th Dec 2013, 7:25 pm | #11 |
Dekatron
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
|
Re: Tin whiskers in transistors
My understanding is that tin whisker growth is poorly understood anyway, although certainly alloying with other metals, such as lead, inhibits it entirely. Crystal structure at the microscopic level makes a lot of difference, but presence of an electric field, or chemicals at the surface, does not aggravate the problem.
Some of Nasa's work was helped by some transistor samples provided by Paul Stenning. |
17th Dec 2013, 7:53 pm | #12 |
Nonode
Join Date: Aug 2007
Location: South Bradford, West Yorkshire, UK.
Posts: 2,571
|
Re: Tin whiskers in transistors
I don't believe that the tin whiskers need an electric field to grow as I've had some unused NOS OC170s with shorts to the case.
There is another phenomena which similar but does require an electric field and the correct environmental conditions to grow. This is dendrites but the whiskers are not dendrites. At work we have experienced dendrite growth on lead free PCBs due to a combination of electric fields, board contamination and heat. Keith |
18th Dec 2013, 9:59 am | #13 |
Dekatron
Join Date: Nov 2003
Location: Near Leeds, West Yorkshire, UK
Posts: 4,609
|
Re: Tin whiskers in transistors
Indeed, most AF11?s are bewhiskered when they've been unused stock over the decades.
__________________
Mike. |
18th Dec 2013, 10:55 am | #14 |
Administrator
Join Date: Dec 2002
Location: Cardiff
Posts: 9,060
|
Re: Tin whiskers in transistors
The transistors I sent to NASA were new old stock and had never been powered (apart from brief test before sending) since manufacture. So no electric field is needed.
|
18th Dec 2013, 1:01 pm | #15 |
Dekatron
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,953
|
Re: Tin whiskers in transistors
The "whisker growth" effect can indeed occur in unpowered equipment: some years back I was supporting a piece of rather old datacomms gear that used several hundred tiny TO-5 cased DPCO relays for low-level RF switching, around fifteen relays to a board - when a board failed we swapped in a brand-new one from our fortuitous supply of shelf-stock and found it U/S. So we tried another... and another... turns out *all* our new boards had tin-whisker bridges in at least one of the little relays.
Just because a spare part has a 'good' test-certificate from 20 years ago doesn't mean it will work today. |
18th Dec 2013, 1:34 pm | #16 |
Dekatron
Join Date: Aug 2003
Location: near Reading (and sometimes Torquay)
Posts: 3,086
|
Re: Tin whiskers in transistors
I suspect that the process is caused by vibration and temperature cycling and where the original tin was very pure.
Many years ago I was on a metallurgy course and we got to play with some single-crystal ultra-pure tin. Is it seriously bizarre stuff!! You might expect that a bar of this stuff would be a bit like a bar of solder, but on trying to pick it up you find that it is initially utterly floppy. As you bend it about it makes a crackling sound as twinning occurs in the crystal structure. What is going on is that when very pure and crystalline it takes almost no force to get it to shear along crystal planes through the bar. You can see the slip planes clearly as the stuff slides apart, a bit like a pack of playing cards. After a while the structure gets more messed up and it gets a lot stronger. So my point is that in the pure crystalline state it would take very little force to cause the tin atoms to slide over each other and go where they were not wanted. |
18th Dec 2013, 2:46 pm | #17 |
Dekatron
Join Date: Dec 2007
Location: Haarlem, Netherlands
Posts: 4,185
|
Re: Tin whiskers in transistors
What are the opinions on heating the transistor case with a soldering iron, instead of zapping? In theory, this solution could last longer, but will the inner workings of the transistor survive?
|
18th Dec 2013, 3:24 pm | #18 |
Dekatron
Join Date: Aug 2003
Location: near Reading (and sometimes Torquay)
Posts: 3,086
|
Re: Tin whiskers in transistors
It will depend on luck and thermal inertia as I think the transistor's guts will start melting about 80C less than the tin whisker.
|
18th Dec 2013, 4:28 pm | #19 |
Heptode
Join Date: Jul 2012
Location: Derby DE1, Derbyshire, UK.
Posts: 626
|
Re: Tin whiskers in transistors
Hi,
Thanks for all the replies, It is all mind boggling and truly fascinating: how come the transistor survives a zapping? I would have thought that would have finished the transistor off quick sharp. Paul. |
18th Dec 2013, 5:21 pm | #20 |
Moderator
Join Date: Jun 2003
Location: Oxford, UK
Posts: 27,787
|
Re: Tin whiskers in transistors
Several forum members have tried this with success. It's also possible to heat up the case and remove the base, which then allows the grease and any whiskers to be removed exposing the junction. Searching the forum should find some old threads about it.
|