Monday, 20 December 2010

teach them to fish ...

I firmly believe in "give someone a fish and you feed them for a day, teach them to fish and you feed them for life". However, teaching takes time and effort, something I am willing to do. Unfortunately not everyone wants to learn: "just gimme the fish!". This is also true for some students (note I am NOT generalising here, there are many hardworking students out there). I recently came across one such case on a calculator forum. He (or she) wanted help with solving a differential equation on the calculator. I (and others) recognised the equation as coming from electrical/electronics field. However, it was evident from the original post that he had put it incorrectly into the calculator. After giving a few directions and hints to get him thinking, the response was "just tell me how to put it in the calculator". He has not come back on that particular subject. Later that same person posted a link about calculator software that showed solutions step-by-step. Unfortunately this was for another calculator, and he proceeded to badger forum members to write such software for him.

I've given up on trying to help that person. If there was indication of wanting to learn, I would gladly help - but there doesn't seem to be. You can lead a horse to water...

Tuesday, 7 December 2010

FMEA/FMECA on electronic circuits - can anyone do them well?

In this case, it stands for "Failure Mode and Effects Analysis" or "Failure Mode, Effects and Criticality Analysis". I work with critical electronic equipment, and to ensure that it works as it should, one of the tools is an FMEA. They can be very useful, once completed correctly, for use in data for Basic Events in Fault Trees, Maintenace Task Analysis, failure detection and isolation, logistic support analysis, unscheduled removal and availability analysis, and more.

BUT, only if it is done well. One of the biggest problems is the proliferation of effect descriptions. There must be UNIQUE and WELL described effects. So often I find the same effect described in various ways. I am currently working on one that was contracted out to a "professional" company specialising in electronics FMEAs. Well, they have (a) badly described effects. I continually have to look at the failed element in the circuit to see what they meant with the effect. (b) Describe the same effect differently. There are almost 4000 low level effects - I reckon it should have been below 1000. There are over 400 high level effects - I am working on reducing it to below 100. Some of these high level effects differ only by one synonymous word (e.g. "element number noted in A" and "element number listed in A"). This makes it so difficult to filter on specific effects - which is what an FMEA is all about!!!.

Is there anyone that thinks the way I do??? Perhaps I should start doing them for a living (in my current job I don't have the time).

my blog name

Why PickyB? Well, one of my bosses told me the other day "you're the pickiest bastard I know", which I guess is a reasonable description of my attitude towards things. They have to be right, why else bother doing them.