Sanity Check: Mini-Split Power Consumption

mjacksonw

Hey all. We put in mini-splits about a year ago, and we're hoping to use the system for basically all our heating this winter in the Boston area.

We run our house on 250-300 watts typically (assuming HVAC is off), and we're seeing comparatively huge energy consumption heating kicked in. It might just be my unfamiliarity with enormous electricity bills, but it would be good to get a sense if what we're seeing is normal, or if there's something up.

On a reasonably cold day (43F high, 22F low, 30F average) last week, we used 98 kWh, contrasted with 10-13 kWh during a temperate week in September. At times on the cold night, the house was drawing over 4kW, which is just way different from what I'm used to – when I turned off the system for a moment, it went right back below 300W. Right now, it's 52F out, and the house has been hovering around 2.1-2.2kW.

It's a 1920s house (3 stories, 2300sqft) with original windows. We got things better insulated last year in the attic and blew insulation into the exterior walls, but I can't imagine anyone could use the word "tight".

System 1 –
Condenser: Mitsubishi MXZ3C24NAHZ2 (24K BTU)
Head 1: MSZFH09NA (9K BTU)
Head 2: MFZKJ12NAU1 (12K BTU)
Head 3: MSZFH06NA (6K BTU) - Currently off (guest room)

System 2 –
Condenser: Mitsubishi MXZ3C30NAHZ2 (30K BTU)
Head 1: MSZFH12NA (12K BTU)
Head 2: MSZFH06NA (6K BTU)
Head 3: MSZFH12NA (12K BTU) - Currently off (attic)

What's the best way to know if this looks roughly sane/expected? I don't necessarily need anything exact, just need to know that there's nothing terribly wrong, and that I'm getting approximately the efficiency I should expect from decent mini-split heat pumps.

mattmia2

If you look at the installation manuals for the condensers there are efficiency curves of the heat pump based on outdoor temperature. The efficiency falls dramatically as the outdoor temp falls. If the math in my head was right the last time I looked at one, at 0 f the efficiency had fallen to a point where it was drawing as much power as an electric resistance heater producing the same output.

JUGHNE

What do you have for back up heat?

mjacksonw

Our backup is a 1940s oil furnace and hot water radiators. It works, but it's darn inefficient. Just figuring that if we can almost always get away with electric, then there's no reason to rush a new furnace.

I guess I'm generally fine using oil in extreme conditions, but if we need it for 20F (not so uncommon), that might be a different story.

Zman

What is your electric rate? 98 KWh does sound pretty high for a colder day but nowhere near design. I have been doing research on heat pumps in colder climates for an energy code review committee. There is a lack of real-world data out there so your info is very interesting.

mjacksonw

Okay, so looking at the specs, the 24K BTU condenser has a COE of ~2.5 at 17F and ~4.2 at 47F.

If I'm doing this right, that'd be 24K BTU * 0.293 = 7,032 watts / 2.5 COE = 2,812 watts of input at 17F.

The 30K BTU condenser has about the same COE, so 30K * 0.293 = 8,790 watts / 2.5 COE = 3,516 watts.

So I guess if it's near 17F, and I'm seeing things averaging less than 6.3kW with all the heads on – or 3.8kW at 47F – then I'm doing okay.

At $0.22/kWh here, if I'm ridiculously oversimplifying and figure the high is 47F and the low is 17F and that we get a something like the average between the two, weighted toward the more efficient high temperature, we might average 4.5kW over the 24h, or ~108 kWh for the day. Which is in the neighborhood of my 98 kWh day last week.

Am I doing this right?

mjacksonw

@zman $.113 for supply, $.116 for delivery.

Zman

mjacksonw said:

@zman $.113 for supply, $.116 for delivery.

Ouch!!

You are generally doing the math right. It gets more complicated if you consider the deration at temp and the actual run time. You also need to check to see where the resistive defrost cycle is factored in.

Jamie Hall

Your math is right so far as I can tell. A somewhat sobering illustration of the problems with heat pumps in colder climates, eh?

JUGHNE

So 30 days of your reasonable cold day of 30 average you would be using 98 KWH X 30 X .22 = $646.80 per month. Of course you have to subtract your domestic house average use.
What heats your DHW?

IIRC you did mention in a previous posting about your oil bill for the boiler, how do these numbers compare?

Just a side question here; how is the comfort level of the heat pump compared to the hot water? I ask because I have installed numerous HP's and people get used to the comfort level provided. We have heated floors and freeze whenever visiting a HP home. Spoiled maybe?

Kjmass1

You can’t use COP like that because the condenser isn’t running at 100% all the time.

You could back in to it using heating degree days if you know your heat load at design temperature and even that is kinda a guess.

The guys Over at greenbuildingadvisor.com could break it down much better than I could. But if I had to guess it’s probably about break even over the course of a winter, with the heat pumps being significantly greener.

Jamie Hall

Heat pumps may or may not be "significantly greener". It depends very much on the power used to generate the electricity involved.

They do look nice and green to the homeowner and the consumer, but... you need to take into account the whole system.

mjacksonw

@JUGHNE:

What heats your DHW?

Stand-alone gas DHW tank. That's a project for another day.

IIRC you did mention in a previous posting about your oil bill for the boiler, how do these numbers compare?

It's tough to compare apples-to-apples. In January 2019 – avg temp of 28.5F – our electricity bill was $609 (figure ~$550 was heating), and we were using oil for the first four and a half days of that cycle. On the other hand, in the first 17 days of January 2018 – avg temp of 23.5F – we went through $556 of oil.

So if electricity costs ~2.5x heating oil per BTU, I guess the heat pumps are performing more efficiently than the oil furnace is on an output basis, though they might be closer to break-even on a dollar basis.

Just a side question here; how is the comfort level of the heat pump compared to the hot water? I ask because I have installed numerous HP's and people get used to the comfort level provided. We have heated floors and freeze whenever visiting a HP home. Spoiled maybe?

I will mostly say that my wife breaks out the blanket more these days than she did when we used the radiators.

If I could wave a magic wand and maybe ignore important principles of economics and physics, we'd have an electric water heater for the radiators. Oil is a pain in the butt – suppliers being iffy when demand "heats up", the tank taking up a lot of space, etc – and my dad died in a gas explosion this summer, so I'm a little reluctant to expand our use of gas here, even though we have it for DHW/dryer/range.

We're investigating solar now, so maybe a 7-10kW array would eat into the electrical expense to make it feel sane.

Kjmass1

I have a similar set up as you in Boston as well- 4 heads/2 condensers with a 1970s natural gas steam boiler that I use once it gets below 35 degrees. The HPs were designed for AC so the sole head on the first floor won't adequately keep up with 1200sf when temps get cold. It might once I insulate my basement, however.

Do you have an oil bill from a cold month last year? Gallons, price, days, K factor? Then you can get a better handle on heat loss previously and compare heat pump only.

Robert O'Brien

Jamie Hall said:

Heat pumps may or may not be "significantly greener". It depends very much on the power used to generate the electricity involved.

They do look nice and green to the homeowner and the consumer, but... you need to take into account the whole system.

Yes

kenjohnson

mjacksonw said:

Okay, so looking at the specs, the 24K BTU condenser has a COE of ~2.5 at 17F and ~4.2 at 47F.

If I'm doing this right, that'd be 24K BTU * 0.293 = 7,032 watts / 2.5 COE = 2,812 watts of input at 17F.

The 30K BTU condenser has about the same COE, so 30K * 0.293 = 8,790 watts / 2.5 COE = 3,516 watts.

So I guess if it's near 17F, and I'm seeing things averaging less than 6.3kW with all the heads on – or 3.8kW at 47F – then I'm doing okay.
....

Am I doing this right?

I think there may be something wrong with your equipment or install. The math doesn't match what the real-world performance should be, in my opinion.

If you carefully study the input and output power data in the manuals at different operating temperatures, I think you'll find that the 24 kBTU condensor is rated for more than 24 kBTU at 47 degrees F and/or the input power will be more reflective of COPs much higher than 2.5 - maybe getting closer to 4 or 5.

I'm in a similar climate. I have half the system as you and for the 6 weeks from October 1st to this weekend, I added about 250 kWh to my electric bill for the entire six week period. If I was running at 40 or 50 kWh per day at any time in the last few weeks, I would have called the installer in for an assessment.

Kjmass1

There is no way you are using all 24k BTUs in these temps, so the math doesn't work out.

You need to figure out your heat loss by either doing a Manual J, or back in to it from an old oil bill.

For example- my home uses 800 BTUs per degree hour. So on a 20 heating degree day (65F base - 20F), the average temperature for 24 hours will be 45F. 800 BTUs x 20 HDD = 16,000 BTUs per hour.

At 45F my heatpumps should be will above a CoP of 4, maybe 5. So 16,000 BTUs / 3400 BTUs = 4.7 kWh / CoP of 4 = 1.175 kWh.

@ $0.23/kWh it'll cost me $.27/hr at those temps. That's a huge generalization, and it'll cost more to run as it gets colder.

My 80% steam boiler would cost 16,000 BTU / .80 = 20,000 BTU
20,000 / 99,999 (BTUs in a therm) = .2 therms * $1.50/therm = $.30/hr.

One thing to factor is your boiler loses at least 25% being the in basement, where as the heat pumps only have to heat the main floors.

george_42

I live in central pa and have geothermal in a 3000 sq ft home and I put a meter on my geo unit two years ago and have used total 18,000 kw