Interesting stuff about RFI:

Heresay: One of the PDA manufacturers supposedly had problems with their 
product locking up if a cell phone was used close by.


Personal experience: In a previous job, I used to develop software for 
embedded controllers.  These programs had to survive real-world interference 
such as electrical shocks and failing power supplies.  The software had to 
continually verify that its own program was operating correctly.  This 
included things such as repeatedly telling an input that it was an input not 
an output, verifying that the CPU stack has not overflowed, verifying that 
each parallel software task was being executed every now and then, being sure 
that the EEPROM memory was kept intact despite impending power-loss, and 
repeatedly telling attached hardware what it should do again and again in 
case electrical noise zapped the communications wire between parts and 
confused something.

The consequences of not doing this right might for example involve a product 
which suddenly could not work any more if the power was cycled on and off 
with a particular timing, followed by a recall of all the products sold up to 
that point all over the world for reprogramming.  But it had to be that 
particular milliseconds timing of on/off power cycling to scramble the 
memory...

All of this is just software glitches from noise and power-loss even though 
the hardware continued to work.


Brian
http://www.bdphotographic.com

Reply via email to