# Using too much electricity?

| Posted in Energy Efficiency and Durability on

I just completed a 1043 sq ft story and a half cottage with a 351 sq ft garage apartment over a detached single car garage. I have two mitsubishi hyper heat splits in the cottage and one in the garage apartment. The cottage has a 27KW ecosmart tankless and the garage apartment has an 18KW. The garage apartment is never used and my usage is between 60 and 74 KWH per day. I’m gearing up to call the utility and the electrician to check things out. Is this atypical for an all electric home this size?

## Join the leading community of building science experts

### Replies

1. | | #1

How cold is it outside?

2. GBA Editor
| | #2

Andre,
We don't have enough information to answer your question. As David points out, it would be good to know more about your climate (your location) and the weather.

It sounds like the garage apartment is unoccupied, but the house is occupied, and that you are reporting that the total electricity usage for both buildings is about 2,000 kWh per month -- resulting in a monthly bill of \$240 if you are paying 12 cents/kWh.

That might be reasonable -- or not. It depends on whether the buildings have high levels of insulation or code-minimum insulation, and whether the buildings have a high rate of air leakage or a low rate. It also depends on what you are doing in the house: lots of cooking? Lots of hot showers? Lots of TV watching and computer use?

3. Expert Member
| | #3

Typical non-water heating non-heating power use for a 2-3 person family is between 2-3kwh/day. The hot water for a family that size could run between 10-15kwh/day in winter. Lets' assume you're using 60kwh /day for just the mini-splits, the rest for other things.

The energy content of 60kwh = 204,720 BTU, or about 0.2 MMBTU. Assuming you're getting a coefficient of performance ( COP) of at least 2 out of the mini-splits that would be more than 0.4 MMBTU of heat being delivered into the house. Over 30 days that would be more than 12MMTU / month, which would not be unusual for an older 2x4 framed ~1400' house if the average outdoor temps (not the average low temps) have been running 15-20F. That would be a bit piggy for a way-better-than code house that size at those sort of outdoor temps. It would be AWESOME for a house that size if the outdoor temps had averaged below 0F during that period.

It would be piggier still if the the outdoor temps have been averaging 35-45F. At outdoor temps that warm your average COP will be 3+ , for more than 18 MMBTU of heat month of fairly mild late-winter/early-spring type weather (or temperate maritime zone winter weather.)

During winter the hot water heating energy use is "in the noise" of the as-used heating efficiency of the mini-splits for a cold climate, unless you're taking "endless showers". (Are you running an average COP of 2.2, or is it closer to 3.7? The difference in kwh used at those efficiencies is a multiple of the hot water power use numbers.) Cooking energy use would normally be an order of magnitude below hot water energy use.

If you can give us a zip code or city location (for heating degree day data purposes) and the EXACT meter reading dates and the power used between those dates we can convert that to kwh per heating degree-day, (after subtracting off an assumed factor for other uses), which is probably the more relevant number. During cold weather the kwh/HDD will go up in part due to the higher space heating loads, but also due to the lower efficiency of the mini-splits at lower outdoor temps and higher operating speed.

During early spring the kwh/HDD ratio could easily be half what it is in January, since the mini-splits will be loping along at min-speed most of the time, and the outdoor temperatures are higher, since both lower speed and higher outdoor temps improve the operating efficiency of the mini-splits. At full speed at +40F a Hyper Heating Mitsubishi will have a COP of about 2.5. As temps drop to -10F the full-speed COP slowly drops to 2. But at lowest speed the COP will be more like 4+ @ 40F, and could easily be 3 @ +10F if the thing is sized to cover the full load at -10F. See figure 4 in this document:

Note that the COP at the "rated capacity" is much higher than at full speed, but the actual loads will be low enough at 40F that., If sized correctly, at 40F it would be running even more slowly and higher efficiency than the "rated capacity", but might be running above the "rated capacity" , delivering lower efficiency at the 99% outside design temp.

4. | | #4

I appreciate the responses. I'm in Memphis where it is 9 cents/KWH. Family of four. I will post meter readings and other specifics in about an hour.

5. | | #5

Info from utility:
"Selected Bill: "Last month:
1/08/2015" 12/06/2014" Bill Impact
Billing Days: 33 days 32 days \$7
Average Cost per kWh: \$0.0925 / kWh \$0.0908 / kWh \$4
Average Use per Day: 77.94 kWh / day 67.78 kWh / day \$30
Total Electric: \$237.86 \$196.99 \$40.87

12/5/14 = 3104 11/3/14 = 935
01/7/15 = 5676 12/5/14 = 3104

I have 2X4 walls with sprayed cellulose, foam sealed T's and caulked plates. The utility conducted a blower door test and the leakage was 5%. I do have an externally vent range hood (broan) and I recently noticed the doors have some gaps. Just picked up some new door sills. It was unusually cold for a couple weeks in December. We are more a "bath" family with a deeper soaking tub. (One tub in the main house.)

Dana - My feeling is the downstairs split is "piggier" than it should be. It's not keeping the space as warm as it should. It was serviced once for low refrigerant, and I'm trying to get checked out once more.

6. | | #6

My apologies, the table didn't format properly. There are three columns: 1/8/2015, 12/6/2014 and Bill Impact.

7. | | #7

More info: I have a 12na downstairs, a 9na upstairs and a 9na in the garage apartment.

8. | | #8

The typical "where did the electricity go?" investigation is frustrating. The best you can do is start collecting data going forward, to find out which appliances are using the electricity, and whether they're using more than they should.

The easiest thing to do, although not the best, is to put a clamp meter on everything in your panel and get an idea of what it uses when running. That doesn't track run times, though, so ideally you buy a TED monitor and start logging data, maybe a day or more per large load. You can get an old-school TED that does one 240-volt circuit for about \$20 on Amazon, or spring for one of the more expensive multi-channel models that puts data on your computer screen.

If you end up with a single channel, you can learn a lot by putting it on your service conductors and watching what happens. You need to be home a lot to make this work--watch the display and notice what's running and what the total load is. I did that for a while in our house and learned a lot.

9. | | #9

Great info Dana! Wish I'd had it during construction.

10. Expert Member
| | #10

Andre: Without a specified location there is no way to look up the heating degree-day information that correlates with that power use. If the bill also includes average outdoor temperature over the period that could ball-park it too.

If you don't want to be specific on a web forum about location, look up a weatherstation near you on degree-days.net, and download base-65F degree day spreadsheet (select daily, not monthly) for the periods in question. Use a spreadsheet tool to add up the daily HDD up and report the kwh/HDD ratio(s) for the specific days between meter readings (not billing dates.)

If the mini-split isn't really keeping up with the setpoint and running at full speed most of the time, you're probably averaging a COP between 2-2.5. Oversizing for the 99% heat load by 25-50% is usually favorable for overall efficiency, but beyond that it begins to fall.

11. | | #11

Hi Dana,

I posted earlier that it's in Memphis, family of four. 9 cents/KWH. Your input is greatly appreciated.

12. | | #12

Ok, hvac tech is here and downstairs unit was a little low on refrigerant. He's charging now. Dana, Martin - what impact do you think this had on electric usage?

13. Expert Member
| | #13

There are at least a couple dozen weather stations in the Memphis TN area on degreedays.net, and sometimes VERY local climates will vary significantly from what's measured 30 miles away:

Improper refrigerant charge would usually reduce the operating efficiency of the unit. A couple years back I read an account of an LG mini-split that had been severely under-charged operating at a COP of barely 1.0 in a temperate western WA location. The owner complained to the installer, who re-tested the unit and declared it "good" without making any adjustments. Since the unit happened to be one that was being monitored by the NEEA as part of their region-wide ductless heat pump assessment project, an engineering contractor for the NEEA ended up later testing and correctly assessed & corrected the problem, after which the power used by the unit (which was being separately metered using the NEEA's equipment) dropped by more than 2/3.

Few installers are that incompetent, but it only takes one...

• |
• |
• |
• |