On 5/30/2011 4:51 AM, Robert Green wrote:
....

If their meter is bad or wired to a parasitic load, remotely monitoring a
sub-meter won't help a bit. It seems everyone here is making a lot of
assumptions based on facts not entered into evidence....

Again, this ignores the symptom as posted earlier--over 17 months the
average is reasonably consistent w/ two extreme outliers (which are both
early in separate calendar years).

If the meter were faulty that wouldn't be the kind of error on would
expect; one would expect continuously high (or may getting higher) usage
indication if it's over-metering. Or, the parasitic/stolen load is
intermittent. Or some other effect nobody's thought of.

Yet again, I conjecture the "normal" readings are estimated and the
large ones are an annual "cactchup" to match reality when the meter is
actually physically read again. Certainly not conclusive but there's
bound to be a reason and it's unlikely the OPs simplistic technique is
going to get anywhere in resolving the issue.

As others have said, a physical photocopy of a couple of the monthly
bills (obviously w/ personal data elided) posted at a hosting site would
be quite an interesting exhibit.

A monitoring of what can be will start, but in a complex he'll still
have quite a few loads that can't be on the kill-o-watt meter (sp?); as
far as I know they're only available for single-phase loads and his main
loads likely are the lights/heat/etc that I would presume are 3...

There seems a fixation on this demand meter but
a) don't even know if it is, and
b) even if were, doesn't make any difference other than $$ amounts;
doesn't explain/have anything to do w/ the above usage data that aren't
the billing amounts but the used kwh.

--