*>>Do any of you have any standard torture tests that test your products
*>>against common levels of noise and spikes?  I am not interested in formal
*>>EMC testing at a lab, just something that can be homebrewed.   We have a

I used a fluorescent fixture ballast without the lamp and with just the
starter and the capacitor across the starter as load. The starters do not
last very long like this, but they are cheap. You can add a VDR across the
starter for extra effect. Makes 0.5 or more Joule pulses at as many volts
as you can think of (remove the input capacitor to the fixture). A spark
gap across the starter could prolong its life also (even better a gas
arrestor at ~400V firing voltage). Testing time is 1 hour or so (or until
the 60W ballast gets too hot to touch, whichever happens first).

For spark susceptibility testing I used a gas ligher piezo element (to
jump sparks into the ground/case), and also an ionizer that made 10kV from
a 9V battery and pulled >10mm sparks. Output caps were 0.01uF I think
(also 0.5J like the coil - the ionizer obviously did not have 0.01uF
output caps - I added them).

I think that if one makes an electronic ignition it should be made to
withstand sparks to ground from the output coil (can be 20J). This is
*hard*. I doubt that most series auto electronics will withstand this in a
modern 'computerised' engine. I'd probably test house alarm systems to the
same standard ...

Peter

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu