Folks,
Tin whiskers continue to generate considerable interest. People often suggest that their risk is great and yet unknowable. RPN may help to clarify the TW risk. What is RPN? It is the risk priority number from failure mode and effect analysis.
As this link tells us:
A failure modes and effects analysis (FMEA), is a procedure in product development and operations management for analysis of potential failure modes within a system for classification by the severity and likelihood of the failures. A successful FMEA activity helps a team to identify potential failure modes based on past experience with similar products or processes, enabling the team to design those failures out of the system with the minimum of effort and resource expenditure, thereby reducing development time and costs. It is widely used in manufacturing industries in various phases of the product life cycle and is now increasingly finding use in the service industry.
RPN is an important part of FMEA. It is the product of three numbers that range from 1 to 10. The first number is the severity (S) of a possible fail. A “10” would be given if the failure injured someone, “7” would be assigned if the failure caused a high degree of customer dissatisfaction, whereas a “2” would be given if the failure has only minor negative effects.
The second number is occurrence (O) of a fail. The highest rating is a “10,” which would be a failure every day (reminds me of Windows ME!) or one fail in 3 events, whereas a “7” would be a failure every month or one in 100 events. A “2” is a six sigma fail rate.
The last number is detection (D) of a potential fail. A”10” would suggest that the detection of a potential fail is either not performed or not possible. A “7” is a manual detection approach that may not be reliable, whereas a “2” is 100% effective potential failure inspection.
So obviously a product with a RPN of 10 x 10 x 10 = 1000 is a disaster, its failure is dangerous, frequent and incapable of being detected beforehand. Industry rules of thumb suggest that and RPN of 200 needs to be addressed and an RPN of 75 is usually considered acceptable.
Let’s look at a “ball park” RPN for tin whiskers. We will assume the application is a critical IC in a PC. Let’s assume that a severity rating of “S” of 8 (failure renders the unit unfit for use) is reasonable. TW are hard to inspect for future fails, so detection, “D,” could be as high as a 10. At this point we are at 8 times 10 equals 80 for both. A bad start.
Occurrence (“O”) for TW failure modes is dramatically different. When trying to assess the occurrence of TW fails, one is often directed to NASA’s web page . Many reference this web site that lists a little more than a score of TW fails. What escapes me is that people don’t seem to appreciate the rarity of less than 100 fails in decades of data collection. Surely TW fails are not common. I could find no report of a failure of a RoHS compliant product anywhere on the internet. So it would be hard to rate “O” any higher than a “2.” I suspect that the reason few TW fails have apparently occurred is due to TW mitigation techniques that are widely practiced.
I would expect that “modern” process defects like the head-in-pillow or graping defects could have a much higher RPN than TW, if assembled without proper process controls and materials. However, there is little need to worry about these defects either, if you use the right solder paste and practice some assembly process precautions.
Score of 160?… getting close to 200 (need to address.. rule of thumb?)
And, for some of the reasons you gave.. we don’t really know the occurrence rates in real life…
The lab tests don’t always track with real life… They are generally intended to accelerate a aging / use of a product.. addressing a known phenomenon.
Until we really understand the root cause of tin whisker growth.. I find our present tests results – questionable.
I have seen product that had nearly 100% failures due to tin whisker growth…
Granted.. it involved the bright tin plating of the shielding between rf sections of a radio.
Not the normal SMT process issue…
But the whiskers were nearly an inch long! ( very weird, waving a pencil in what appeared to be open space – removed shorts.. but you could see them under magnification.
The other aspect of the use of RPN scoring… it very much depends on the product and it’s application.
Yes, this is addressed in the scoring system.. but I question the scales as explained.
In the radio mentioned above… it was for communications in Aircraft.. generally rescue/police/coast guard helicopters… how does that fit on a scale of 1-10?..
a dead pc rates 8? (your example)..
Dead people only rate 2 points more?… is there some log scale involved here?
There is no getting around it.. ROHS, while legally exempt from Military or other high reliability applications… will HAVE to be used in these applications.. because no one will be making components for any other process in the near future… at any price.
example:
Why was a process that allowed tin whisker growth allowed on a satellite (that failed because of it)?
The only answers I can think of :
– The only component choices available for the application required ROHS.
– Or someone involved, didn’t know their job (used ROHS when alternatives were available)
Don’t know which applies for this example (was it ever questioned?)
But I do know the choices for future high rel electronics applications … will require the use of ROHS based components…. or very different processes that will require years to qualify.
I am very concerned about how this will be addressed.
It effects our security …..(cost and reliability of military/police) ,
economy……. (expected life of our infrastructure),
and environment …..(how we address product life/reuse, and pollution)
And I don’t see balanced, informed discussions taking place in Europe or USA on this topic.
Just over simplified , knee jerk reactions ….. with little regard for un-intended consequences
Don’t worry about it .. science will fix it!… doesn’t always work.
sorry.. will get down off the box now….
I like the comparison to head-in-pillow defects. Both Sn whiskers and HIP are particularly nasty because it will pass the functional and visual tests but eventually fail in the field. A product that is functional leaving the factory but can fail quickly in the field is really scary.
Got to agree with John.
Euro MP’s with too much time and money trying to justify their existance. The same people who wanna ban Tetrabromobisphenol-A and created a standard for Cucumber straightness!!!!
Does any one have a handle on the faulure rate of units soldered with ROHS components and ROHS solder?
Seems like all commercial electronics should be failing left and right.
Secondly, If I use ROHS components and solder them with tin lead solder, how much does the likly hood of a failure from a tin whisker decrease?
I agree with this statement:
“There is no getting around it.. ROHS, while legally exempt from Military or other high reliability applications… will HAVE to be used in these applications.. because no one will be making components for any other process in the near future… at any price.
The only thing is there are companies which will re-tin your components with SnPb solder.
Now what?