On Sat, 3 May 2014 01:36:42 +0200, "Dimitrij Klingbeil"
wrote:

"John Fields" wrote in message
.. .

Download all the files into a single folder and then run ZCD7.asc.

John Fields


Done that and wondering: why is the sim taking so long? Three and a half
hours CPU time for a single run from start till finish at essentially 100%
CPU load. OK, that was less than 15 minutes real "wall clock" time (on a
dual Opteron system with 16 cores total), but still, that sim takes more
than plenty of time for what it does. I think, it dislikes LTSpice's
"alternate" solver in some strange way (or I've got some tolerance or
stepping parameters somewhere off-kilter).

BTW, it looks fine with a sinewave, but take a look at the version below -
with a really messy SNAFU mains like you'd sometimes get in places with a
lot of thin wires and "solar" inverters. Of course the waveform of the
"noise" is not realistic, it's just "something messy" made up in order to
simulate effects of non-deterministic stuff happening when a lot of poorly
filtered switchers "get their way", but still, some "interesting" reactions
from the circuit (slightly modified .asc below, all else being the same):

---
YOW!!! Interesting indeed!

With mains that dirty and that type of ZCD, I'd have to run the load
_and_ the ZCD from a ferroresonant transformer.

John Fields