Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
05/31/06 16:06
Read: times


 
#117400 - if supervisor ever fires you have a bug
It went off-topic so I moved it to chat.

Erik said:
The issue is that if your supervisor ever fires you have a bug. YOU breought up that the supervisor could allow you to get by with a "reasonably well designed" project because that when "reasonably well designed" was not good enough, the supervisor would save you. BULLSHIT


The issue is, life is more complicated than black&white, as you present it here. And I believe you know it very well.

No matter how thick your metal box is, how perfect the inputs filters are, now much ferrite you used, how perfectly you welded down the lid and how thick the gold on it is, there are always circumstances which will throw your device off. I mentioned RF, but "curcumstances" are also mechanical influences (shocks, vibrations), heat/cold, humidity etc.

Every sensible designer designs to a certain set of requirements. Most often he is able to define a "typical" situation, maybe estimate likely deviations. Some of the requirements are set in standards, based on "normal" way of things.

But, sometimes, things get out of "standard", "normal", "typical". You can either moan, scream and complain, and eventually get out of business; you can make things in heavy metal and - being uncompetitive - get out of business; or you can make things as rugged as possible with as little effort/cost as possible and get a good reputation.

An example. Designing a PC mouse in pure metal, welded down and filled up with ferrite would indicate seriously damaged designer's sense. However, when one of my friends attempted to work with his laptop in the vicinity of a XX kW RF amplifier box, the cursor on the screen from time to time jumped wildly, opening and closing windows at random. He'd rather chose the mouse stay unresponding momentarily from time to time, I'd guess.

Another example. Your power supply is probably as perfect as possible (and you even provide your power supply, which is not always the case - e.g. in PCs). However, the electrolythics in it might suffer from a slow death, especially if subject to heat or similar (I know, there are higher-rated caps etc. but nobody can foresee how does it behave under xxxx current/voltage pulses suffering heat shocks etc.etc.). Consider one sopmewhere in feedback path, with randomly changing leak. The output of the power supply gets wild, emitting spikes and noise. The micro goes crazy and your display will display funny things, if the supervisor is too "bold".

JW


List of 13 messages in thread
TopicAuthorDate
if supervisor ever fires you have a bug            01/01/70 00:00      
   design for worst case, not typical            01/01/70 00:00      
      OT: where? why?            01/01/70 00:00      
         not allowed            01/01/70 00:00      
            ridiculous laws            01/01/70 00:00      
            Dont the police always wear sunglasses?            01/01/70 00:00      
         if it got to work            01/01/70 00:00      
            on radios and similar            01/01/70 00:00      
            Boxes above the raceway            01/01/70 00:00      
               poor english & too specific            01/01/70 00:00      
                  Sounds like fun            01/01/70 00:00      
   Lots of RF            01/01/70 00:00      
   If the supervisor fires....            01/01/70 00:00      

Back to Subject List