My computer has broken. When is yours going to break?

Internal view of my dead computer

Internal view of my dead computer: the faulty capacitor is just visible at the top left corner of the panel marked 'G5'.

A couple of weeks ago, after five and half years of faithful service, my computer died. And today I painfully reconciled myself to its death, and ordered a replacement: cheaper, more powerful, and hopefully longer lasting. But it made me think…

I had used that computer pretty much every day for the last 2090 days. Each day it started with a happy chime, until one day it didn’t. The pictures on the screen became degraded and eventually the computer stopped responding. An internet search revealed the cause: six years ago a bunch of faulty electrolytic capacitors entered the market and were incorporated into a wide variety of computing equipment. The capacitors were manufactured with a trace impurity in their dielectric, and now, all these years later, this fault is causing failures in otherwise reliable equipment. This set me thinking about the longevity of all the computing kit that has become ubiquitous in our lives  – well my life at least. Presumably after some period: 10 years, 20 years or 100 years, it will all break. And as electronics becomes ever more reliable, and we trust it in more diverse environments, these failures will come as a more  of shock. And I was reminded of one of the more shocking of the possible failure mechanisms. It arises from the use of lead-free solder.

‘Conventional’ solder is mainly an alloy of roughly one third lead and and two thirds tin. Lead-free solder is typically made from alloys of silver, copper and tin. If I have understood what I have read correctly, the problem arises with coatings used on components which are soldered onto the printed circuit boards. These used to be coated with lead-tin solder, but are now coated with pure tin. However pure tin is a metal which can grow ‘whiskers‘ – growths of tin which are only a few thousandths of a millimetre wide, but which can grow several millimetres in length. Wow! that’s incredible! The factors affecting the growth of these whiskers are many and varied, so that in service, it is hard to predict when, where, and at what rate they will grow. But grow they will, and eventually they will give rise to failures, typically by breaking off and causing a short circuit. The phenomenon has already resulted in the verified loss of at least one satellite, the recall of pacemakers, and failures within nuclear power stations.

Now I don’t mean to alarm you: engineers are aware of this problem and have been for 60 years – that’s why we used lead-tin solder for all those years – the lead inhibits the growth of these whiskers. Indeed I know colleagues at NPL who are hard at work helping manufacturers to find appropriate solutions. And when there are safety critical systems, exceptions can be made. But in the end – for one reason or another, due to tin whiskers or not – all this electronic stuff will just stop working.

Tags: , ,

One Response to “My computer has broken. When is yours going to break?”

  1. Ruth Temianka Says:

    Really interesting and informative post!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: