From: muratlanne@gmail.com   
      
   "Richard Smith"  wrote in message news:m1a52o6uwz.fsf@void.com...   
      
   "Jim Wilkins"  writes:   
   ...   
      
   You were forced to overspecify everything... ?   
      
   "The pigs all had their snouts in the trough".   
   By this everyone can get paid for doing nothing.   
      
   -----------------------------------   
      
   Sometimes, depending on conditions. When trying something new available   
   products are rarely a close match, and industry required what appeared to be   
   overspecifying based on previous incidents, like a forklift puncturing an   
   electrical cabinet. That resulted in the requirement that all internal   
   hazardous voltage be run through grounded conduit so a short would pop a   
   breaker, and regardless of power level any exposed wiring outside enclosures   
   had to be too strong to break accidentally or intentionally by hand, 16AWG.   
   The external panel controls had to be oil-tight to protect them from   
   conductive or flammable coolant spray from whatever they were placed beside   
   and soapy water wash-downs with a hose. Oil-tight switch controls were   
   several times more expensive, and modular so their switching functions could   
   be changed with a screwdriver, such as single pole single throw to double   
   pole double throw. Lab equipment wasn't subject to those requirements but we   
   met them anyway since we were prepared for them. Delivery is quicker if you   
   have stock on hand and charge for its replacement. I do that for common   
   hardware at home too.   
      
   Idiot-proofing encourages hiring cheaper idiots as operators. The history of   
   infantry weapons and tactics demonstrates this when compared to artillery   
   which required more careful handling. An example today is singers dropping a   
   fragile and expensive professional microphone. Air hose nozzles became   
   vented and less effective because some idiot worker once pushed a high   
   pressure nozzle against another and punctured the skin.   
      
   Automotive electronics can be subject to nearly the temperature stress of   
   fighter jets, for example a snowplow in Alaska (or NH) starting cold and   
   ramming at full power into a snowbank, or being splashed with icy water. As   
   an apprentice I helped assemble a large expensive machine that rapidly   
   cycled a batch of GM HEI ignition modules between hot and cold limits while   
   testing their proper operation. If a loose or corroded battery clamp   
   disconnects the alternator can generate around 100V until the energy stored   
   in the rotor has dissipated and the voltage regulator regains control of   
   rotor current.   
      
   Aviation electronics had to at least stay in place during a shock of   
   somewhat over 30 g's (?) so they wouldn't fly into the cockpit during a   
   rough belly landing. That one was a mechanical engineer's problem to solve,   
   as was adequate cooling airflow at altitude which seemed excessive on the   
   ground. Some WW2 aircraft engines fed supercharged air into the distributor   
   to avoid sparks jumping at high altitude.   
      
   Submarines were/are? tested to the pressure equivalent of 10,000' altitude.   
   Diesels pull a vacuum in the hull if the snorkel closes from a large wave.   
   The radar stealth coating on WW2 U-boot Schnorchels didn't hide them from   
   the latest secret Allied radar, the boats were out until the war's end but   
   ineffective after spring 1943. Those are examples of conditions one might   
   not expect or design for.   
      
   New equipment intended to test other new equipment sometimes needed   
   modification as the specs changed, using components with higher ratings was   
   a valuable hedge. An example I can mention is the Apollo spacecraft flight   
   control computer that had more computation to do during the actual moon   
   landing than had been expected or tested for. It's supposed failure was a   
   watchdog timer lamp that flashed because the program couldn't cycle through   
   all tasks in a pre-specified time which had been initially assumed to be   
   more than adequate, and could be changed in hardware if the programmers knew   
   to ask. The program loop's final task was to reset the timer that kept the   
   light off, presumably unless the computer crashed. I heard that from the   
   Mitre engineer who had helped design and build the computer. He was on other   
   tasks when the computer was being tested.   
      
   I don't intentionally overspecify on home projects that I can easily change   
   unless the available material is more capable than needed or I don't know   
   the worst condition, such as the starting shock load of my electric hoist or   
   the maximum starting (locked rotor) current for an electric motor running   
   from a DC-AC inverter. Measured results vary widely depending on the random   
   phase at connection and the inverter's surge capacity is also unknown, and   
   indeterminate until I buy one.   
      
   I designed the tractor bucket loader for an impact that would lift the rear   
   wheels off the ground, but didn't know the weight capacity of the front   
   tires and axles so I machined an adjustable hydraulic pressure relief valve   
   and crept the pressure up until a tire failed. I collect used test equipment   
   to be able to measure these things.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)