Author Topic: tin whisker  (Read 16306 times)

0 Members and 1 Guest are viewing this topic.

Offline bkubicekTopic starter

  • Newbie
  • Posts: 2
tin whisker
« on: September 11, 2013, 06:05:41 am »
Hi!
I think that some more info on tin whiskers, and these things could be quite interesting.
http://nepp.nasa.gov/whisker/video/index.html
I stumbled on this topic, while watching this NASA soldering tutorial:

It would be quite cool if the showed soldering methods are still "up to date".
 

Offline Stonent

  • Super Contributor
  • ***
  • Posts: 3824
  • Country: us
Re: tin whisker
« Reply #1 on: October 17, 2013, 03:43:25 am »
Hi!
I think that some more info on tin whiskers, and these things could be quite interesting.
http://nepp.nasa.gov/whisker/video/index.html
I stumbled on this topic, while watching this NASA soldering tutorial:

It would be quite cool if the showed soldering methods are still "up to date".

Those NASA soldering videos make me want to tear my hear out when watching.

"Looks like Bob on the other side of the room didn't wash his hands after coming out of the bathroom, we're going to have to wipe and sterilize our solder again"
The larger the government, the smaller the citizen.
 

Offline Monkeh

  • Super Contributor
  • ***
  • Posts: 7992
  • Country: gb
Re: tin whisker
« Reply #2 on: October 17, 2013, 03:50:33 am »
Hi!
I think that some more info on tin whiskers, and these things could be quite interesting.
http://nepp.nasa.gov/whisker/video/index.html
I stumbled on this topic, while watching this NASA soldering tutorial:

It would be quite cool if the showed soldering methods are still "up to date".

Those NASA soldering videos make me want to tear my hear out when watching.

"Looks like Bob on the other side of the room didn't wash his hands after coming out of the bathroom, we're going to have to wipe and sterilize our solder again"

And this attention to detail is what keeps spacecraft, whether manned or otherwise, working..
 

Offline Stonent

  • Super Contributor
  • ***
  • Posts: 3824
  • Country: us
Re: tin whisker
« Reply #3 on: October 17, 2013, 05:33:11 am »
And this attention to detail is what keeps spacecraft, whether manned or otherwise, working..

I wouldn't think they'd be hand soldering everything for that though.

Now, two things I did get from the video that I liked. The resistor bending tool and bending the pins in the direction of the trace.

Also did not think about stressing the component. I've been known to take a component that is slightly askew and melt the solder on one side and pull it down by the pin while leaving the other side soldered down to straighten it. I do see how that could be a problem now and will be more mindful of it.
The larger the government, the smaller the citizen.
 

Offline Monkeh

  • Super Contributor
  • ***
  • Posts: 7992
  • Country: gb
Re: tin whisker
« Reply #4 on: October 17, 2013, 05:34:39 am »
And this attention to detail is what keeps spacecraft, whether manned or otherwise, working..

I wouldn't think they'd be hand soldering everything for that though.

There's more hand work involved than you might think, it's not as if they produce large quantities of some of this stuff.. Plus repair (kinda worthwhile when the board cost six figures), modification..
 

Offline Stonent

  • Super Contributor
  • ***
  • Posts: 3824
  • Country: us
Re: tin whisker
« Reply #5 on: October 17, 2013, 05:41:08 am »
I hear they're looking into that new fangled 74 series logic now.    :-+
The larger the government, the smaller the citizen.
 

Offline AndyC_772

  • Super Contributor
  • ***
  • Posts: 4228
  • Country: gb
  • Professional design engineer
    • Cawte Engineering | Reliable Electronics
Re: tin whisker
« Reply #6 on: October 17, 2013, 06:46:45 am »
Interesting that he bends the resistor leads one way to retain the part in the board, then after cropping, bends them a completely different way. He even applies quite a bit of twisting force with that wooden stick during the clinch process, which I can't see being a particularly good technique for ensuring the component and its end caps don't get over-stressed.

Offline NickS

  • Supporter
  • ****
  • Posts: 55
  • Country: au
Re: tin whisker
« Reply #7 on: October 18, 2013, 06:06:04 am »
Also did not think about stressing the component. I've been known to take a component that is slightly askew and melt the solder on one side and pull it down by the pin while leaving the other side soldered down to straighten it. I do see how that could be a problem now and will be more mindful of it.
I do that too and couldn't care less. Never had a failed resistor from it. :)
Mind you I'm not building spacecraft.....
 

Offline ElectroIrradiator

  • Frequent Contributor
  • **
  • Posts: 614
  • Country: dk
  • More analog than digital.
Re: tin whisker
« Reply #8 on: October 18, 2013, 07:54:13 am »
Do keep in mind that components on a spacecraft may be subjected to very serious levels of thermal extremes. A component assembly may have to survive repeatedly being frozen down to the temperature of liquid Nitrogen. This could happen for instance if there is a temporary fault condition with the attitude control, or something similar. Same thing for literally being baked in the high noon Sun for a while... >:D
 

Offline Stonent

  • Super Contributor
  • ***
  • Posts: 3824
  • Country: us
Re: tin whisker
« Reply #9 on: October 18, 2013, 07:02:56 pm »
Do keep in mind that components on a spacecraft may be subjected to very serious levels of thermal extremes. A component assembly may have to survive repeatedly being frozen down to the temperature of liquid Nitrogen. This could happen for instance if there is a temporary fault condition with the attitude control, or something similar. Same thing for literally being baked in the high noon Sun for a while... >:D

Possibly the temperature of solid nitrogen too.
The larger the government, the smaller the citizen.
 

Offline ddavidebor

  • Super Contributor
  • ***
  • Posts: 1190
  • Country: gb
    • Smartbox AT
tin whisker
« Reply #10 on: October 18, 2013, 07:37:07 pm »
The fact is if a satellite turn around earth 40 times a day, you have 40 extreme temperature changes.

For everyday. For 15 year.
David - Professional Engineer - Medical Devices and Tablet Computers at Smartbox AT
Side businesses: Altium Industry Expert writer, http://fermium.ltd.uk (Scientific Equiment), http://chinesecleavers.co.uk (Cutlery),
 

Offline grumpydoc

  • Super Contributor
  • ***
  • Posts: 2905
  • Country: gb
Re: tin whisker
« Reply #11 on: October 18, 2013, 08:23:56 pm »
Quote from: bkubicek
Hi!
I think that some more info on tin whiskers, and these things could be quite interesting.
http://nepp.nasa.gov/whisker/video/index.html
I stumbled on this topic, while watching this NASA soldering tutorial:

It would be quite cool if the showed soldering methods are still "up to date".

The NASA video was interesting but does anyone outside NASA call a PCB a "PWB"??

 

Offline Monkeh

  • Super Contributor
  • ***
  • Posts: 7992
  • Country: gb
Re: tin whisker
« Reply #12 on: October 18, 2013, 08:59:01 pm »
Quote from: bkubicek
Hi!
I think that some more info on tin whiskers, and these things could be quite interesting.
http://nepp.nasa.gov/whisker/video/index.html
I stumbled on this topic, while watching this NASA soldering tutorial:

It would be quite cool if the showed soldering methods are still "up to date".

The NASA video was interesting but does anyone outside NASA call a PCB a "PWB"??

Sometimes, it's an old term.
 

Offline Kjelt

  • Super Contributor
  • ***
  • Posts: 6460
  • Country: nl
Re: tin whisker
« Reply #13 on: October 19, 2013, 07:53:21 am »
Pffff now at least I know why those products from nasa are so expensive, they take ages to assemble  ;D
Come on use a semi cleanroom, gloves, bent it the right way the first time and you,re done in 10 seconds.
Are they using SMT nowadays or is that still too risky? If you have such thermal shocks, I wonder how SMT will hold up.
 

Offline Neilm

  • Super Contributor
  • ***
  • Posts: 1546
  • Country: gb
Re: tin whisker
« Reply #14 on: October 19, 2013, 01:37:23 pm »
Quote from: bkubicek
Hi!
I think that some more info on tin whiskers, and these things could be quite interesting.
http://nepp.nasa.gov/whisker/video/index.html
I stumbled on this topic, while watching this NASA soldering tutorial:

It would be quite cool if the showed soldering methods are still "up to date".

The NASA video was interesting but does anyone outside NASA call a PCB a "PWB"??

IEC61010 still refers to PCBs as PWB.
Two things are infinite: the universe and human stupidity; and I'm not sure about the the universe. - Albert Einstein
Tesla referral code https://ts.la/neil53539
 

Offline walshms

  • Regular Contributor
  • *
  • Posts: 183
  • Country: us
Re: tin whisker
« Reply #15 on: October 22, 2013, 04:40:32 am »
Now, two things I did get from the video that I liked. The resistor bending tool and bending the pins in the direction of the trace.
US Military Wiring Specification WS-6536 revision C; can't remember for certain, but I think it was an Air Force spec.  Stringent as hell, but very highly reliable, especially in high vibration/high stress environments.  I used to assemble power supplies that had to meet the requirements of that spec and they were beasts... and yes, everything was hand soldered, inspected, tested, reworked as needed, cleaned within an inch of its life, tested a second time and then potted with a re-enterable compound before being tested yet again.

Quote
Also did not think about stressing the component. I've been known to take a component that is slightly askew and melt the solder on one side and pull it down by the pin while leaving the other side soldered down to straighten it. I do see how that could be a problem now and will be more mindful of it.

Amazing how easily some things can be damaged.  When it's got to be as reliable as it can possibly be, you don't mess around.
 

Offline walshms

  • Regular Contributor
  • *
  • Posts: 183
  • Country: us
Re: tin whisker
« Reply #16 on: October 22, 2013, 04:46:42 am »
The fact is if a satellite turn around earth 40 times a day, you have 40 extreme temperature changes.

For everyday. For 15 year.

The International Space Station circles the earth about every 92 minutes, roughly 16 orbits per day.  Satellites higher up orbit slower... so it's not quite 40, but the extremes are indeed extreme.  Think in terms of up to ~200C swings, depending on the emissivity of the surface in question.  That's why most of the stuff up there has heaters built in.
 

Offline walshms

  • Regular Contributor
  • *
  • Posts: 183
  • Country: us
Re: tin whisker
« Reply #17 on: October 22, 2013, 04:51:53 am »
Are they using SMT nowadays or is that still too risky? If you have such thermal shocks, I wonder how SMT will hold up.

Sure; SMT is commonly used now, but the requirements are probably just as painful to meet.  Proper centering on the pads.  Solder fillets on component legs that must show the entire outline of the component leg (no excess) and make a smooth transition to the pad.  Each and every one.  No nicks on component surfaces, or the component legs.  The slightest scratch is enough to cause an inspector to reject a board.

Think that's expensive?  Trust me on this -- it's far more expensive to send something up there that fails because the time wasn't spent on the ground to ensure it was right.  That's why teams spend up to 10-15 years preparing hardware before sending it where it can't be repaired...
 

Offline Kjelt

  • Super Contributor
  • ***
  • Posts: 6460
  • Country: nl
Re: tin whisker
« Reply #18 on: October 22, 2013, 06:54:09 am »
Sure; SMT is commonly used now, but the requirements are probably just as painful to meet.  .........
Probably also gluedots to keep the components mechanically fixed without having mechanical stress on the solder joints?
They probably still use leaded solderpaste?  :)
 

Offline Stonent

  • Super Contributor
  • ***
  • Posts: 3824
  • Country: us
Re: tin whisker
« Reply #19 on: October 22, 2013, 07:32:54 am »

I think given the poor reliability that RoHS has added to electronics, I'm sure of it.
The larger the government, the smaller the citizen.
 

Offline ElectroIrradiator

  • Frequent Contributor
  • **
  • Posts: 614
  • Country: dk
  • More analog than digital.
Re: tin whisker
« Reply #20 on: October 22, 2013, 07:39:14 am »
They probably still use leaded solderpaste?  :)

Yes, they do. Solder containing lead is still allowed for use in critical applications like aerospace, military and medical equipment.
 

Offline AndyC_772

  • Super Contributor
  • ***
  • Posts: 4228
  • Country: gb
  • Professional design engineer
    • Cawte Engineering | Reliable Electronics
Re: tin whisker
« Reply #21 on: October 22, 2013, 08:23:59 am »
I should hope so. Lead leaching out of scrapped PCBs into drinking water supplies is less of a problem in space than it is on Earth!

Offline Stonent

  • Super Contributor
  • ***
  • Posts: 3824
  • Country: us
Re: tin whisker
« Reply #22 on: October 22, 2013, 09:22:21 am »
I should hope so. Lead leaching out of scrapped PCBs into drinking water supplies is less of a problem in space than it is on Earth!

Lead came from the earth in the first place.
The larger the government, the smaller the citizen.
 

Offline romantronixlab

  • Regular Contributor
  • *
  • Posts: 88
  • Country: us
Re: tin whisker
« Reply #23 on: December 08, 2013, 01:10:56 pm »
Since i have been reading information on tin whiskers why is formed etc. Ill post some pics for those lazy enough to search google and of course for the novices in the forum.

Removing lead from solder may seem a smart idea environmentally, but the resulting microscopic growths called tin whiskers could be just as problematic. - theguardian.com

Credited to the respective providers.
Tin Whisker Detached from RF Shield/Bracket which then bridged exposed conductor. nasa.gov


Millstone nuclear generating plant in Connecticut shut down when a circuit board monitoring a steam pressure line short-circuited. theguardian.com

nasa.gov
electronicstheory.com



Will think about it.
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 16281
  • Country: za
Re: tin whisker
« Reply #24 on: December 08, 2013, 03:22:29 pm »
Had it on my car as well. Took the unit apart and had a little work on it with a soldering iron to check joints on connectors, and scrubbed it with IPA. Seems to have fixed it for now. Was rather annoying to drive with all the warning lights lit, including for diesel heating, EGR, ABS, SRS, overheat and oil pressure. They stayed on with key off as well.
 

Offline Christopher

  • Frequent Contributor
  • **
  • Posts: 429
  • Country: gb
Re: tin whisker
« Reply #25 on: December 08, 2013, 03:55:55 pm »
Does thick conformal coating help at all?
 

Offline ddavidebor

  • Super Contributor
  • ***
  • Posts: 1190
  • Country: gb
    • Smartbox AT
Re: tin whisker
« Reply #26 on: December 09, 2013, 10:09:56 pm »
i suppose yes, how can tin wiskers grow up on the coating?

but i'm not an expert
David - Professional Engineer - Medical Devices and Tablet Computers at Smartbox AT
Side businesses: Altium Industry Expert writer, http://fermium.ltd.uk (Scientific Equiment), http://chinesecleavers.co.uk (Cutlery),
 

Offline KJDS

  • Super Contributor
  • ***
  • Posts: 2442
  • Country: gb
    • my website holding page
Re: tin whisker
« Reply #27 on: December 09, 2013, 10:17:10 pm »
Pffff now at least I know why those products from nasa are so expensive, they take ages to assemble  ;D
Come on use a semi cleanroom, gloves, bent it the right way the first time and you,re done in 10 seconds.
Are they using SMT nowadays or is that still too risky? If you have such thermal shocks, I wonder how SMT will hold up.

One of the worst tasks I've ever done is assemble three boards, each with 600 components, about 400 of which were 0402 resistors and capacitors to go on a military satellite. It took nearly forever to get it right, complete with a small air gap underneath each component to ensure that a slight board flex didn't stress the part. And the paperwork, don't get me started on the paperwork.

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 16281
  • Country: za
Re: tin whisker
« Reply #28 on: December 10, 2013, 04:28:30 pm »
You do know that the paperwork has to be larger than the unit mass. I only had to fill out one form per unit, and one on receipt, one on test, one on repair, one on verification and one for QC. Then the labels for the box, for the unit, for the stores and all the paperwork, in the style that stores wanted, complete with the spaces they wanted left blank for their side. If I took from stores a page for the order, a record in a register, a copy in a file for the unit, another in the general file and still another form to order the part. Carbon paper a good saving, as the forms all had to be original or carbon copy, not a photocopy. Best thing I ever did was to get a self inking stamp pad with the required data aside from my signature, so as to use on the forms. Stores looked at that and she wanted heir own.

That was only for avionics, I only had to fill out modification forms once, that took a 3 month period of back and forth of piles of paper. The basic form was 4 pages with "Add as needed extra pages marked annexure XYZ" and whole sections marked as " Not affected by this modification" along with prices, suppliers and 3 quotes per part with lead times and guarantees as to availability for x months and a set of contact details and other details.
 

Offline KJDS

  • Super Contributor
  • ***
  • Posts: 2442
  • Country: gb
    • my website holding page
Re: tin whisker
« Reply #29 on: December 10, 2013, 05:00:39 pm »
You do know that the paperwork has to be larger than the unit mass. I only had to fill out one form per unit, and one on receipt, one on test, one on repair, one on verification and one for QC. Then the labels for the box, for the unit, for the stores and all the paperwork, in the style that stores wanted, complete with the spaces they wanted left blank for their side. If I took from stores a page for the order, a record in a register, a copy in a file for the unit, another in the general file and still another form to order the part. Carbon paper a good saving, as the forms all had to be original or carbon copy, not a photocopy. Best thing I ever did was to get a self inking stamp pad with the required data aside from my signature, so as to use on the forms. Stores looked at that and she wanted heir own.

That was only for avionics, I only had to fill out modification forms once, that took a 3 month period of back and forth of piles of paper. The basic form was 4 pages with "Add as needed extra pages marked annexure XYZ" and whole sections marked as " Not affected by this modification" along with prices, suppliers and 3 quotes per part with lead times and guarantees as to availability for x months and a set of contact details and other details.

I'd have been happy if it were larger than the unit, even by a factor of ten. The unit from memory weighed about 250g, that's about half a pound in old money.

As one of the designers, not only did I have to sort the assembly paperwork but all of the design details as well.

From memory there was the following

The design description, which includes a full description of the design and how it works, all its schematics, parts lists etc
The process description, so if soldering component to board using Sn63 solder and weller iron with bit type xyz was on the list of approved and controlled processes that line was easy. If it wasn't then there was more work.
The worse case analysis, which covers temperature, component and tolerancing variations
The rad hardness report.
The compliance matrix
The test plan, which explains what testing philosophy will be used to prove that the product meets each line item of the spec
The test procedure which details how to do the tests
The test report, which contains the results of the tests. I had one friend that had put together 9000 plots on a single, complex piece of equipment. He only had another 6000 to do to finish when the project was cancelled.

Each report not only had the essentials, but also a few dozen pages of boilerplate, plus indices of chapters, figures, tables. It was usually necessary to provide six copies of each on paper, not less than 90g/sqm, as well as a few more copies on CDRom.

and despite all that, I still prefer working space jobs to military ones, at least the paperwork is proactive rather than reactive, even if there is fractionally more of it.

Offline codeboy2k

  • Super Contributor
  • ***
  • Posts: 1836
  • Country: ca
Re: tin whisker
« Reply #30 on: December 11, 2013, 06:58:11 am »
Does thick conformal coating help at all?
i suppose yes, how can tin wiskers grow up on the coating?

but i'm not an expert

Apparently, it does help. Specifically the one they tested and I assume is approved for use. But they only needed a few mils. Anything less than that and the whiskers grew through and could penetrate outward or another outside whisker could penetrate inward.

Summary: "After >9 years, no whiskers have grown through the nominal 2 mil thick Arathane 5750 (formerly known as Uralane 5750). There are extensive (high # per area) whisker growths beneath the coating that are pushing/stretching the conformal coat, but none have managed to puncture the coating where the coating is a nominal 2 mils thick or greater"

From their testing here: http://nepp.nasa.gov/whisker/experiment/exp2/
photos here: http://nepp.nasa.gov/whisker/experiment/exp2/exp2photo.html#ccs

EDIT: one of the more interesting findings from that experiment was that the conformal coating actually increased the rate and number of tin whiskers under the coating. i.e. they appeared sooner and in greater quantities to the non-coated samples in the experiment.
« Last Edit: December 11, 2013, 07:07:39 am by codeboy2k »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf