Well I have to admit that I'm stumped with an energy audit I've been working on for the past two months. Perhaps someone can point me in the right direction...
My client contacted me concerned that his electric bills were atrocious. They were, to the tune of 34,680 kWh per year (average monthly at 2,890 kWh, and a baseline of 2,347 per month!). And after my first visit to his home, it was very obvious to me that, although he has a big house (3700 sq-ft), he and his family of four were energy frugal, being careful to shut off lights when not in use, using power strips for entertainment centers, etc.
He was particularly concerned about the phantom consumption when he and his family would leave for several weeks at a time. When they did he'd shut off everything but the alarm system. And even then he'd see a bill of 1900 kWh for that month, and for what?
So I dove in deep, taking this on as a personal project, since he'd had an auditor from the utility company come out previously and they couldn't find anything. One of the first things I did was to install a PowerSave EnviR monitor on the two mains into the main circuit panel and let that thing collect data for exactly two weeks. I compared this kWh total to what the utility company's meter read and found a discrepancy of about 500 kWh in favor of the utility company. This closely paralleled the additional 10k kWh per year that I could not account for in my projected analysis. Hmmm.... found it! (or so I thought)
So I got on the phone and was soon in touch with the supervisor of the electric meter department. I told him my findings so he and I got together at the client's home one day so that he could test their equipment, which consisted of a transformer/transducer system which reduces the incoming current from the main transformer at the street. After a good hour of testing he informed me that their equipment tested okay. Hmmmm.... bummer.
After many hours of monitoring individual circuits, extrapolating the long-term consumption, I still can't account for this additional 10k kWh per year anomaly. I even looked for hidden conduits coming off the wiring gutter under the transformer, assuming a neighbor might be "borrowing" some power for their growing operation. Nothing. Absolutely nothing.
And now that I've been so thorough with my investigation and extrapolation, I'm left scratching my head, wondering if, even though the utility company tested their equipment as good, perhaps it is not over a long term. I honestly don't know where to go from here... Suggestions?
Dennis, that's a ton of useful info and it will take me a bit to fully assimilate those articles/studies. I sure do appreciate your input. However, I'm disappointed to know that under low loads the inaccuracy will be in the favor of the homeowner rather than the utility. Makes perfect sense though.
I'm still in favor of having the utility company come out again to test their equipment, knowing what I know now. I want to pay close attention to exactly what they are testing this time to be sure they check all components.
If you look at the original posting, you will see that I had done some exhaustive research into their loads and was able to account for everything. That total didn't match the utility company's numbers though, which is what set me off on this wild goose chase. At some point I will have reconciliation, but until then, I've sort of adopted this project for a learning experience, as it's important for me to know what's going on in that mysterious black box called a meter.
Thanks again for all the great research and information!
Rod, I did see earlier that you had calculated the loads. The key is even though a consumer device may not have the accuracy... you can still use it to find the "phantom loads". You need to use that spread sheet of expected loads and try to place them on a circuit in the breaker panel... then it may be necessary to monitor & move from circuit to circuit over a period of several days. You should see the culprit -- if you leave the data logging going long enough to catch the actual problem. My guess is the phantom is not really going to be a small load... but it will be a big intermittent load . For example I've seen compressor stalls on HVAC systems going from where they were adding 2kW or 3KW average... to suddenly 6kW or 7kW average -- as the condenser motor had seized up.. and was having problems and the poor compressor was getting clobbered... A water pump thats leaking water back at the well - or even a water tank that has a failed rubber bladder and is short cycling the pump will consume a lot more power.
Now back to that 10kWh anomaly a year...if the total use is 38,000kWh, 10% is 3800kWh, 5% is roughly 1800kWh/yr... then 0.3% is likely to be above 10kWh/yr... that 10kWh is well into the noise level for many of the best logging equipment -- when you consider the size of the load... in fact a reasonable error might be up around 100kWh/yr.
On the other hand if you expected 10kWh max loads a year and sized the measuring equipment for the smaller loads/current.. then the error might be around 30Wh (0.3% measuring accuracy)
If the test equipment you are using is only rated at 3% accuracy... then it is reasonable to have a measurement error in the range of 1,100kWh/yr (roughly). That makes 500kWh/yr look pretty good...
I've gone through the process that you tried with the house... and you can indeed find all and accurately account for nearly every watt... if you are willing to take the time and pay the price to do so... it takes lots of data logging with many CT's sized appropriately for the circuit loads. The logged values can not be simply the hourly cumulative totals... or even the fifteen minute... you many need under minute measurements. Then you need not to get caught up in believing that the totals are precise -- they are generally not -- ESPECIALLY if you use the off the shelf consumer logging devices... TED's for example. You'd get a good feeling for the numbers but not precise.
I'd start looking at the 75A, 50A, 30A circuits first.... use CT's that appropriate for the circuit size and capture the data in second intervals if you can.... the small intervals will let you detect motor starts... etc. Then go through the 20A circuits and compare against the items on that circuit...
Then on to the 15A, etc...
My guess is before you hit the 20A circuits you'd find the big items that are causing the unexpected high loads during the summer time.... and for me... that keeps coming back to motors... most likely pumps or big fans (attic fans for example).
Since this is getting long I'll break response into multiple pieces.
As mentioned I've gone through this before... shrinking the equivalent (gas + electric) from about 54,000kWh down to under 18,000kWh (gas+electic). Note the (gas+electric) means the BTU heat value of the gas was converted to kWh and added to the consumption. That wasn't a 54MWh electric only consumption. But we shrank the electric the fastest... the gas usage is still slowly shrinking..
For Rod, second part...
When I see really large electric services... my first reaction my be "what a waste", but when I stop and catch my breath and take the time to ask why... something more useful is generally discovered.
My house is roughly 2800 sqft. Family of four... Seattle has a reasonably moderate marine climate. NM can be hot (but they don't have the air conditioning - per your comments), the summer time use is high and it looks like their loads are for a predominantly heating climate. But why 300A or 400A service.... that's isn't necessarily a "given" for that size of a house a 200A would likely be more than sufficient. So see if they know why the large electric service was installed... is there a sub panel that it is feeding... with stuff that was missed. If the house was older (mine was a 60's vintage)... maybe there are lots of lights... I had counted 110+ lights in my house... even with only 55 on at a time... that's a big load.
Our house originally had a 125A service panel with about 20 circuits. When we went through the house about eight years ago.. I moved to a 200A service panel with 40 circuits. But then immediately had the main breaker downgraded to a 150A breaker. That panel feeds a 100A sub panel via an automatic transfer switch for the generator. That sub panel has lots of smaller circuits in it... typically 15A circuits. Seems convoluted but I wanted to be able to measure and monitor individual circuits... and I didn't like the old system in which if a breaker tripped nearly all the lights on one floor in the house would go dark. We chose to break up lots of big circuits into many smaller circuits. (and in case anyone asks... yes it was permitted and inspected by electrical inspector and work was done by a licensed electrician).
The point being the easiest way for to get the many circuits was to use a larger panel... with the larger bus bar. You don't find 125A panels, or 150A panels with 40 circuits... so again... why the big service... was it just to have lots of circuits - with a few heavy swingers... or was it because the original owners had a kiln, a horse barn, etc.?
Nowdays, if that same house with a large service was located in WA, OR, or perhaps CO... the question might be... how big are the grow lamps and fans... but at the same time - the house may have a family that likes to grow their own organic vegetables... tomatoes, carrots, peas, celery... and not the hemp products..
If the family are the original owners they might also be able to indicate if the consumption has stayed the same or climbed over the years...
Also how many yard lights do they have? If this is out in the country... with kids - they may have added extra yard lights that go on after dark such that kids can play basketball on a court by the garage. If the yard lights are mercury vapor, halogen, etc... that could easily add up.... and switching those to LED's saves a large chunk.
You might not be able to whittle away the 500kWh difference or 10kWh difference.... with lots of accurate measurements... it is often just to eyeball them and say... 400W x 8 hours a day *365 days = 1,100+ kWh... and you'll move that to a LED equivalent and save 700kWh/yr... it may cost less to fix the problem then to measure it.
Teton County and the Town of Jackson Adopted the 2014 IRC energy code.... With that, inspectors now require blower door testing so caulking plates, outlets and any opening is almost needed to pass the test... Funny though..... I have been a contractor here for 25 years and in all structures I have built. I have caulked those joints for at least 20 years.... A case of good caulk bought at Jackson Lumber and a half day of my time works out to 300 bucks... Max.... And I always have happy customers...