# How to limit heat pump electrical load

| Posted in Energy Efficiency and Durability on

So how do I size a system that meets the heating demand as much as possible while living within the available power capacity of the townhouse complex?

A group of residents and I are working to blaze a trail for our townhouse complex to transition from Natural Gas powered heat and hot water to electric. We are located in Vancouver Canada. Climate zone 4 with less than 3000 HDD. We had an electrical study completed to understand our existing state capacity.

Without getting into the weeds on the Canadian Electrical Code, the load capacity calculation uses the MCA (minimum current ampacity) times the Voltage to calculate the power load.  If you look at the spec sheet of the heat pump used in this article:
Ducted Minisplits in My Conditioned Attic (Part 1) – GreenBuildingAdvisor
The heat pump selected was the MXZ-3C24NAHZ2
SB_MXZ-3C24NAHZ (mitsubishitechinfo.ca)
It has a rated MCA of 30Amps. Our townhouses have a 240W feed, meaning it would count 7200W (30x240W) towards the electrical load calculation.
The requirement for the 40 amp breaker seems to back that up that this unit could indeed draw that kind of current. What I am gathering is that the COP (coefficient of performance) drops to about 1 during extreme cold. IE you put in 7200W of electrical power and get out 7200W heat. Using the conversion factor of 0.293 W/(BTU/hr) this converts to 24,573BTU/hr, roughly the rated heat output of the system.

I’ve had an independent heat load calculation done that came in at 10 kW. Using a gas fireplace to suppliment space heating on cold days, I could electrify the heat and hot water IF the electrical load for the heat pump could be kept to 4000 W or less. But if I have to allow for 10,000 W, our complex wide electrical system will need to be upgraded. That means 75% of the complex will need to agree to spend \$10-\$20k per household to upgrade the electrical. ie: Not going to happen.

Is it possible to specify a heat pump with a minimum COP of 2 at its coldest operating temperature?

Thanks so much for your help. I really enjoy following this site.

## Join the leading community of building science experts

### Replies

1. | | #1

It sounds like you have 4000 W of available electrical capacity, I don't think you can do anything other than get a heat pump that draws 4000W, or about 16 amps. I think the question you should be asking is what is the best cold-temperature performance you can get. How cold is cold for you? There are Hyperheat units that have a COP of 2+ at 0F.

2. | | #2

The Mistubishi FS18NA is rated for 17,000 BTU/hr at -13F (COP 2+) and has an MCA of 16A. At 5F it does 23,000 BTU/hr. This is the kind of unit you should be looking at.

1. | | #4

Ok, so the approach is to identify units that are designed to run at really low temperatures so they will still have a decent COP at temperatures they actually see?
Around here, near record cold is 5F.

3. Expert Member
| | #3

A MXZ-3C24NAHZ is awfully large for a townhouse in Zone 4. For example this townhouse in Zone 6 needed only about 3/4 tons of heat:

For demand calculation you can use the max power draw of the unit which is much less than the MCA. Mitsubishi typically lists this for most of their unit, this is the best resource for technical information:

The 3c24 is somewhere around 4000W way less than the MCA in the data sheet.

You are also not in cold enough climate to need a hyperheat unit. In your 0C 99% design temperature climate, the non hyperheat unit produces about the same heat and for some reason it does so at a higher COP.

1. | | #5

I'll see if I can re-engage our Electrical consultant. It does seem odd to me that I would be able to install something with a 40 amp breaker (MCA 30), but claim it will never draw much more than 16. If that is the case, then why wouldn't it be installed on a 20 amp breaker? Or could you?

Also, thank you for the link, lots of good details in there.

1. Expert Member
| | #6

Lot of electrical sizing is to protect the wiring in the device, not necessarily related to the actual draw of the device. Most likely there is some internal fusing that needs a certain service size for it to clear properly in case something like the inverter fails as short.

Inverter based compressors don't have any of the inrush or locked rotor draws of regular induction motor. I've monitored power use on these and they match the power in the spec sheet pretty closely.

For demand calculation you don't use breaker or wire sizes, what matters is the actual power of the load. For example an 3000W electric baseboard adds only 3000W of demand even if it is on a 20A breaker.

If you are still near the limit on the main service based on demand calculation, your utility might be able to provide actual power data if there are smart meters or install a power meter and log for a year. This would let you use the real world peak power for service and load sizing instead of estimating.

4. | | #7

I was able to contact Mitsubishi Canada (They were great! prompt, knowledgable replies).
Their explanation was that the value provided represent the power required to run at a specific outside temperature and a specific inside temperature. They key part is that the power requirement will increase with a higher interior setpoint. Basically in the most extreme scenario, yes the unit could draw the MCA level of amperage on a continuous basis.
Again my problem is that our units don't have a lot of power to spare. But our climate is fairly mild (Record low is 5F, -15C) and we could probably allocate 4000 to 5000W per home.
Based on what I've seen this should be sufficient for covering most of our heating needs, if not all. The struggle is figuring out how to lock out the heat pump in some manner so that it doesn't draw more than 4000W or 5000W. According to Mitsubishi, they don't have anything that does that. Possibly this might be able to be achieved with control of the thermostat, I recall smart meters have a similar program with utilities to automatically reduce power use during peak periods.

1. Expert Member
| | #13

The Mitsubishi rep is wrong. The compressor in the unit is not large enough to consume anywhere near the MCA amount of power. To use 30A you would need a compressor about 2x larger, the compressor running full tilt under any conditions will never get there.

Heat pump demand is based on actual max power which is clearly listed in the datasheet.

In your climate the non-hyper heat version delivers about the same amount of heat but has about 1/2 the max power requirement which would be a better fit.

1. | | #15

Below is what the rep said:
"The MCA listed on the unit nameplate/submittal is the highest steady-state electrical current that the unit should see when operating correctly"
The other point made regarding the max power use:
"The values listed in the submittals are based off certain indoor and outdoor conditions. For example in Heating mode the maximum power input at -13*F will be 4480W when the indoor return air temperature is 70*F DB, 60*F WB. The power input would increase if the indoor return air temperature increased"
This does make some sense, that the higher the indoor return temperature, the more power would get consumed.
Its hard to tell if I'm getting incorrect answers or correct answers to incorrect questions.

1. Expert Member
| | #17

The full Mistu spec sheet from Mylinkdrive has the FLA (rated load amps) for both fan and compressor. They sum to 22.4A, so 5.3kW. Nowhere near MCA as it should be the case as MCA is calculated from rated load.

5. | | #8

What's the amperage available per house? I ask because I live in an old rowhouse as well, and a heat pump fit the 100 amp panel with no issues. I used this article, not sure if it applies in Canada:

Do you have duct work? I went with Mitsubishi's ducted hyper heat unit, which has an outdoor MCA of only 17A and equivalent heating capacity.

If you can't do ductwork, look at multiple outdoor units. For example, two of these get you the same capacity with a combined MCA of only 20A.

6. | | #9

Let me look into that one! 17amp MCA sounds perfect
The bottleneck isn’t the 100amp panel. It’s the feeder it shares with the 8 other townhouses. Through the magic of derating, the feed only has a capacity of 400 amps, so we need to be careful about how we allow new loads, especially hi draw continuous loads.
We do have ductwork for the lower 2 floors, but not the upper floors. Ductwork is pretty leaky though and the air return mostly just uses the joist space.

1. | | #10

Ah the joys of retrofits. The 12kbtu SUZ unit uses 14A, so easy enough to pair with the tiniest furnace you can find. 95% electric is often so much easier than 100% electric. Wild that the utility doesn't want more sales!

7. Expert Member
| | #11

You're trying to size off of the wrong number. MCA is Minimum Circuit Ampacity, and is used to size the wire, and usually the overcurrent protection (circuit breaker) too. The MCA is *NOT* the same as the maximum *continuous* load of the appliance, and cannot be, per code. You are only permitted to use 80% of a circuit's ampacity on a continuous basis, with "continuous" defined as 3 hours or more. That means that the maximum allowable continuous load on that 30A circuit would be 24A, or a little under 6kw.

The data on the unit implies that the maximum continuous load of the unit is somewhere around 4.5kw. Even if you round that up to 5kw, you're still well under the MCA, as you should be. One of the things that happens here is that there are standard circuit breaker sizes, typicall going 15A, 20A, 30A, 40A. A 20A, 240V circuit has a maximum allowable continuous load of a bit under 4kw, so it's too small for your unit. That means you have to go up to the next standard size, which is 30A, even though it's more than your unit really needs. From what you've shown in the data sheet, I think it's very unlikely that unit will ever use more than around 5kw, regardless of outdoor temperature -- it's going to run into a limit at some point, it can't just draw more and more power forever until the breaker trips. At some point, it will just not be able to provide any more heat, and it's load will level off, and per code that point cannot be over 24A on a 30A circuit, which is not even 6kw.

You should have demand factors to take into account between units in your condo complex. The basic idea is not everything will be running at the same time, so you don't need to provide large enough power feeds to support everything running at max, since it won't ever happen in practice. As a conservative guess, as someone who designs large power systems for a living, I'd be surprised if the average load of your condos during times of heavy use is much over 5kw per unit, which would be 40kw for the complex of 8 units, which is about 167 amps at 240v. A 400A service is supposed to be limited to about 320A continuous (the 80% rule mentioned earlier), which is about 76.8kw, or 9.6kw per unit. That's actually a pretty good amount of power, but remember that it's very unlikely that all the units will draw that much all the time, and you can go up over the 80% number for periods of less than 3 hours. If you assume a more normal load pattern with varying loads, you have about 96kw available, or 12kw per unit -- 50A at 240V.

I think you'll be fine with the unit you want to use. If you have serious concerns, consider installing some metering on your main power feed so that you can watch the actual load. Some smart meters that the utilities use can be made to cycle through some of this data for you to see. If you have a symbol a little like this: -[===]- at about the 4 o'clock position on your smart meter, that is a sensor for a magnet. The magnet will tell the meter to show you different info. It's not a problem -- you're not going to do anything to the meter -- the function lets you pass the magnet past that sensor to cycle through different displayed data, and ampere load is usually one of the things that can be displayed. That might help you gauge the actual load of your complex.

Bill

1. | | #16

I agree that the breaker size is not a method to figure out the max power requirement. I was told by an electrical engineer I hired that when doing the max load calculations you had to consider the maximum continuous current draw. The Engineer said it was the MCA as did the Mitsubishi rep. Doesn't mean they are 100% correct, but those are two pretty good sources.

In Canada Section 8-202 covers how to calculated the minimum Feeder capacity. Which, with the help of the engineer, we have been using to estimate the spare capacity in the feeders. That part has been fairly easy (with help from someone knowledgable of course). Figuring out what value to use for the heat pump load continues to confound me.

8. | | #12

I think you're looking at this wrong. Let me offer an analogy. Let's say you own a truck that has a rated capacity of 20,000 pounds, and there is a bridge that is limited to vehicles of under 10,000 pounds capacity. The question you're asking is akin to asking what modifications can you make to that truck to reduce its capacity, and the answer is there aren't any. If you want to cross that bridge legally you need a smaller truck.

Similarly, any permitting for the equipment is going to be based on the manufacturer's specification, and no modification you can make is going to change the manufacturer's specification. Some modifications might invalidate the listing and make the equipment illegal to install.

You need to figure out two things. First, you need to figure out what the actual limit on what you can install is. It sounds like there are two actors with an input on that. One is your local building department, and the other is your townhouse association. You need to talk to both. Then you need to find out if you can buy a piece of equipment that is under that limit that meets your needs.

9. | | #14

Interesting question.

I think Bill has it right, above, and you can most likely make this work within your constraints

For what it’s worth, here in the US the NEC treats refrigeration equipment quite differently than lightening and general use loads because heat pumps behave differently.

In Lighting and General Use, the breaker protects the conductor. But Table 240.4 G lists the exceptions to that rule, including refrigeration and AC. For that equipment, the overload protects the conductors, while the breaker is sized to not trip during inrush current and protects against ground fault and short circuit conditions. So the code will allow seemingly strange things—a 50A breaker and 10 gauge wire on a nameplate MCA of 30.1A for ex. (Mike Holt has a couple of lectures about this.)

Long story short, you can’t use breaker size to infer peak load with heat pumps. Actual peak loads are much lower as Bill says.

Another way at this is through COP. Mitsubishi’s FS18 one-to-one bottoms out at a COP of 1.9 at -13F. Not sure they publish that data point for all units, but in any case your worst case on any cold climate ashp is going to be much better than a COP of one.

If you can data log this it would make an interesting case study. I’d guess as we move toward electrification that these questions will crop up more and more.

• |
• |
• |
• |