Hybrid Car – More Fun with Less Gas

Home Electric Consumption - Page 5

register ::  Login Password  :: Lost Password?
Posted by Neon John on October 1, 2007, 9:48 pm

Unlikely, with one exception.  Easy enough to check.  Open all breakers and
apply a
single known load (a 500 watt halogen lamp is a good choice.  So is a 100 watt
and count the disk turns.  Compute the indicated watts using the meter's Kh
and the time interval.

Based on both industry data and my own experience with a utility engineering
background, meters either work or they don't.  Calibration is very stable.  The
exception is if the drag magnet becomes demagnetized.  The drag magnet sets the
slope of the disk.  That is, how fast the disk turns for a given load.  The
are very stable but certain external forces can demagnetize them.  The most
usual is
a VERY close lightning strike.  This is an extremely rare occurrence, though,
and can
be easily checked with the above procedure.

How are you measuring the various loads?  Some of those figures seem quite high.
Particularly the water heater and well pump.  Just to get a better feel, I
the well pump into horsepower-hours - 5.4 approx.  If your pump is a 1 hp unit
it would have to run over 5 hours a day to consume that much energy, neglecting
losses, of course.  Do you run the water that much and/or could you have a leak?

The lighting load also seems quite high.  Is this a large house?  Outdoor
lighting, maybe?  Also, I don't see any mention of a washer or dryer.

I'd like to know some more details.  How large the house is, how many occupants,
you arrived at the figures, is the electric meter mechanical or electronic, etc.
kWh a day can't just disappear so there's a problem somewhere.

You might also state what instruments you have at your disposal.

John De Armond
See my website for my current email address
http://www.johndearmond.com  <-- best little blog on the net!
Tellico Plains, Occupied TN
If we aren't supposed to eat animals, why are they made with meat?

Posted by bscallywag on October 2, 2007, 10:19 pm

occupants, how

etc.  30

The house is 6000sq ft - 2500 basement, 3000 main level, 500 top
floor. 4 people (2+2 small)

The meter is electronic.

I have - calibrate - hand meter, mostly fluke ( 2% accuracy ).  A Rode
and Schwarts power meter that will deal with 220v split phase that I
can borrow on the weekend.
I also have a kilowatt meter and a clamp on wire power meter
( borrowed ) that is intended for monitoring home office etc usage.

The water usage is high because yes we did have a leak,  the poly. had
come off of the pit-less coupling.  That is now fixed,  but all my
data runs previous to that fix.

Measurement...  I have hour meters installed on the electric hot tank
elements, well, and HVAC compressor and ODU fan.
Measurement for them is hours x power (kw)

Cooking is on propane.

Cloth washing is cold water only, with air drying ( we don't own a
tumble dryer ).  And is lumped in in the other category.

Lightning strike was to the pole that 'my' transformer is on.  600ft
from the house.

Thanks for all the information so far.


addresshttp://www.neon-john.comhttp://www.johndearmond.com<--  best little blog
on the net!

Posted by dogbreath on October 2, 2007, 10:29 pm
 On Tue, 02 Oct 2007 15:19:32 -0700, bscallywag wrote:

occupants, how

etc.  30

OK, a fairly large house, but have you redone your measurements since the
water problem was fixed? If so, I'd first try to do a KWh reading
between the meter and the breaker box for a month and see if you can get
your reading to correlate to the utility meter reading. I have a sprinkler
system that got installed before we bought the house we are in, and it's
_prior_ to the breaker box with it's own breakers. Drove me jack batty
trying to find the controls for it.

at your disposal.

Posted by Neon John on October 3, 2007, 12:09 am

occupants, how

etc.  30

My first advice would be to call your utility and ask them to check your meter's
calibration.  Be sure to tell them that you've suffered a nearby lightning hit.
usual procedure is for a meter man to come out and swap out meters, then take
the old
one back to the meter shop for a calibration check.

If your meter were mechanical then I'd doubt there being any problem.  With the
electronic versions, I'm not so sure.  One of my utility clients is having a
very bad
experience with their new fleet of electronic self-reading meters.  Lots and
lots of
outright failures and LOTs of PO'd customers with suddenly huge power bills.
Many of
those revenue issues are the result of defective meters.

For a house that large I don't think your total usage is all that far out of
assuming you HVAC the whole place.  Your consumption can certainly be reduced but
it's going to take some work.

Back to your energy auditing. I see a whole bunch of problems here.  Let's walk
through 'em.

The first and probably the most serious problem is lack of uniform metrology and
methods.  You have several instruments of unknown accuracy and calibration
status and
you're using a disparate collection of measurement techniques, some of which just
won't work.

The first thing to do is constrain error accumulation.  The use of several
instruments with various characteristics and (presumably) without NIST-traceable
calibrations means that you have no control over error which I suspect is large.
Worse, since you must add the results of many measurements to compare with the
cumulative (revenue meter), you have totally uncharacterized and controlled error

The simplest and least expensive but most tedious method is to use only one
instrument of known accuracy.  Errors will still add up but if the instrument is
sufficiently accurate (1% FS is good, 0.25% is great) then the error accumulation
won't be too bad.

A second option, the one I use in my energy auditing, is to use multiple
instruments, calibrated against a single lab transfer standard.  My lab standard
is a
GE 0.1% class watt-hour transfer standard.  I have this instrument NIST-traceable
calibrated occasionally but since I have a 20+ year history on the instrument, I
that barring physical damage, its calibration is stable and accurate.

I calibrate each instrument before an audit and then check it afterward.  This
precludes any damage, drift or other problems causing bad data to creep into my
results.  If there is a significant change then I redo that measurement.

You MUST measure actual accumulated watt-hours on each load.  Your hour meter
technique simply won't work because the load simply isn't stable enough over a
period of time.  Even with the water heater where the resistance is stable,
and daily voltage fluctuations invalidate a simple "elapsed time and single
reading" method.  With loads such as the AC, the actual power draw varies widely
according to the point in the operating cycle, ambient conditions and other

You can't do volts * amps because that doesn't take into account power factor.
the power factor varies with load on electric motors, for things like washing
machines and AC compressors, it's difficult to impossible to characterize "a"

The easiest, simplest and cheapest method of making these measurements is with
common revenue watt-hour meter.  Mechanical revenue meters are quite accurate
(usually better than 1%) and are extremely stable.  The turndown ratio (ratio of
maximum to minimum measurement capability) is second only to the very most
test instruments.  Used watt-hour meters are widely available on the net for
$0, frequently less than half that.

Here's my little stub of a web page on how I use revenue meters


I have in excess of 50 of these setups in my meter fleet.  Though I show a large
amp meter base in those photographs, I prefer the 100 amp or "CT" meter base
that is
round and of the same diameter as the meter.  Lighter and easier to handle.

A meter base isn't strictly necessary.  Several of my early setups consisted of
nothing more than a thin plywood inverted "T" with slots cut in the vertical
piece to
match the meter stabs (terminals).  The stabs have holes through which I passed
and nuts to which the wiring was anchored.  If the wood is varnished it is a good
dielectric even when moist.  The only real requirement is that the meter remain
nearly vertical.

I have a few of my meters set up with alligator clips and/or flying leads so
that I
can meter at the breaker panel.  I simply lift the hot and neutral wires (or
hots for
240 volts) of the branch being investigated and place the meter in series.

What you'd need to do is to get one or more of these meters and methodically
each load in your house.  You can use the spreadsheet on my site to keep track
of the

If that clamp-on meter you have is a true-RMS responding true watt meter
(voltage AND
current inputs) then you can do a quickie check of your revenue meter.  The
is to gain access to the incoming feed at a point where you can gather both legs
together.  I normally do that at the weatherhead.  Arrange the legs in a cross
pattern so that the current flow in each leg through the clamp meter is opposite
other.  That is, the two add.  Attach the potential leads between the legs (240
volts).  The meter reading is divided by 2 to get the actual total wattage.  This
configuration is identical to the way the revenue meter does it and handles both
and 240 volt loads.

Of course, you must observe proper electrical safety while doing this.  I use
standard utility hot gloves when making this measurement.  If, for some reason, I
don't have my gloves handy, I layer on many layers of non-latex examination
Typically 6-8 layers, depending on the brand of glove and how stiff they are.
those go leather work gloves to protect the insulating gloves from abrasion or

The service drop at the weatherhead is unfused other than the transformer primary
fuse so a fault can generate a few tens of thousands of amps of fault current.
that in mind.  I wear flame-proof (utility worker) pants and shirt and a face
when I'm messing with power drops.  My safety gear has never been tested but I've
seen the result of service drop shorts and it isn't pretty.

What you'll do is compare that meter's reading in watts to the revenue meter's
indication.  I outline how to determine watts with a mechanical meter on that web
page above.  An electronic meter is a bit different.  Normally it has an
either an LED or bars on the LCD display, that indicates watt-hours consumed.
The Kh
factor will tell you what each event is worth in watt-hours.  Typically with LCD
displays, the Kh is 1.  That is, each movement of the bargraph is one watt-hour.
indicators can be anything.  I've seen Kh values as high as 8.  That is, 8
for each flash.

At the same time you observe the clamp-on meter to verify that the load is
you count the events over a timed interval, multiply by the Kh factor and divide
the time interval of the measurement to get average watts.  I like to accumulate
least 25 events and preferably 100 to minimize quantization error. The two values
should agree to better than 1%.

John De Armond
See my website for my current email address
http://www.johndearmond.com  <-- best little blog on the net!
Tellico Plains, Occupied TN

*fas-cism* (fash'iz'em) n. A system of government that exercises a
dictatorship of the extreme right, typically through the
merging of state and business leadership, together
with belligerent nationalism.  -- The American Heritage Dictionary, 1983

Posted by Jim on October 3, 2007, 4:06 pm
     Something is very definitely wrong somewhere; we may have been a wee bit
premature in castigating you so soon. But! You are in the right place.

This Thread
Bookmark this thread:
  • Subject
  • Author
  • Date
please rate this thread