The former regulator is often just an ordinary mechanical
you see on lawnmowers/lawn tractors etc.
: Given the mechanical systems etc., regulation tends to be poor
: at extremely low loads. So, when testing voltages off a
: you should be driving at least 5-10% of the load rating in
: to get a reliable number.
===> Ewwww, you're kidding; really? Hey, I've got enough to do
without going around figuring out how stuff works right now! ;-}
Seriously, everything you said makes sense and is as I'd have
imagined it. But, you've got me real curious now about the speed
When you say "just an ordinary mechanical ... " do you mean
the old centrifugal type?
I'm too sick to go look at it today, but I don't recall
seeing anything resembling that. There is however a fairly
complex (to me at least) mechanical linkage system I've always
been curious about, and I make sure I keep it meticulously
cleaned and free of any buildup so nothing will get sticky.
As for small load, this Coleman seems to stay on target fine but
the voltage of course is highest at that point and of course it
drops with added loads, and you're also right, it's 120V even at
a moderate load and there seems to be a "flat spot" in the
response curve where it held 120V pretty well before and after
the moderate load. I call a "moderate" load 50% to 60% or so.
Aside: Just for grins, I went in a few minutes ago and checked
my UPS log. Even with all my testing, I'm still outstripping the
grid supplier (NIMO) by a w i d e margin! In the last two months
there have been switches to battery for overvoltage 7 times,
undervoltage 6 times, and outages 8 times.
The day I did the testing there were of course outages, and I
was surprised to see overvoltages, but then it turns out I had
the upper voltage trigger limit set to 125V, so that accounted
for those, probably also the ones on the grid; I don't get
mearurements, just alarm triggers. No undervoltages either, so I
couned that a plus, although very brief problems aren't counted.
I've no idea how many cycles they consider "brief" but it's
shorter than the computer PS hold-over time, so the system never
notices them. I -think- I recall that this PSU can miss up to 5
cycles without problems; but don't quote me. So, the UPS must
react in, what, something less than 5/16 Second dropout? Head's
too cloudy to do the math at the moment; that's probably wrong.