Mini-Split Heat Pumps are Quirky, Especially their Capacity

Mini-Split Heat Pumps are Quirky and Still Effective

The name "mini-splits" generally refers to heat pump systems with one outdoor unit and one or several smaller capacity air handlers. The efficiency of their operation comes via the inverter compressor and variable refrigerant flow technology, and they are available for single- and multi-family homes, as well as commercial buildings of any size. Here's a general overview of what the systems are and how they work, in case you're not familiar. Compared to conventional heating and cooling equipment, they're very quiet and very efficient. For most folks, ductless mini-splits are what they've heard of, but there is also a concealed ducted option that many of our clients choose to avoid the "wall wart".

Over the past several years, I have gotten to know mini-splits pretty well, and like the 2001 Land Rover Discovery II that we drove for 10 years, with it's backwards dashboard and odd noises, mini-splits can be rather quirky. Unlike the Discovery II, though, most of the quirkiness of the mini-splits is good, not just quirky! At this year's Building Science Summer Camp, Bruce Harley, Technical Director at Conservation Services Groupdiscussed his experience and the quirkiness of the combination ducted and ductless multi-split heat pump system installed in his Vermont home. Below are some of the key takeaways from his presentation with a little commentary from me:

VRF Mini-Split Heat Pumps indoor and outdoor unit

IMAGE LEFT - Mitsubishi ducted mini- or multi-split heat pump air handler in encapsulated attic
IMAGE RIGHT - Mitsubishi 4-ton multi-split outdoor unit

Key Takeaways from "The Quirkiness of Heat Pumps", a Presentation by Bruce Harley
(Click title to see entire presentation):

  • MYTH BUSTER: "Air source heat pumps work in hot and cold climatesMost mini- and multi-split heat pumps are rated and have proven to deliver heating down to -20⁰ C / -5⁰ F, albeit at lower capacity." All of our homeowners from Maryland, Rhode Island, Toronto, Michigan, etc. have all reported back to us that they're 2-3 year old mini-split heat pumps are keeping them plenty warm in the winter. For the commissioning I did of a Providence, R.I. home we worked on (images below), I stayed in the home, itself. It was the dead of winter (17 degrees!). If it didn't work I would have suffered right along side him. No one suffered!

VRF Mini-Split Heat Pumps Providence Rhod Island LG Squared

IMAGE LEFT - Mitsubishi outdoor unit mounted on exterior wall of Providence, R.I. home
IMAGE RIGHT - (Left to Right) Gordie (happy homeowner), Ray Bolvin (installer), and Me

VRF Mini-Split Heat Pumps Providence Rhod Island LG Squared2

IMAGE LEFT - Mitsubishi ceiling suspended ducted air handler.
IMAGE RIGHT - Mitsubishi outdoor unit with access panel removed during commissioning

  • ADVANTAGE VRF: "Efficiency of VRF heat pump is better than even the manufacturers tout or than they are rated." - B. Harley. Between the energy monitoring, homeowner testimonials, and review of our own energy bills, (we also have a ductless mini-split heat pump system in our condo) we find the systems are from 20-40% more efficient that promised. WooHoo!
  • MYTH BUSTER: "Temperatures setbacks do not save" energy. - B. Harley. Think "set it and forget it" for most of the year, and "use only modest setbacks for when you're away.- B. Harley. We recently got our July 2014 electric bill, and compared to July 2013 when we experimented with setbacks as opposed to "setting it and forgetting it." We found, after factoring in a difference in weather, etc, a 25%-30% savings by not setting back temp.

Electricity Bill Condo with Mini-Splits_with set backs

July 2013 Electricity bill with set backs

Electricity Bill Condo with Mini-Splits_set it and forget it

July 2014 Electricity bill when we set it and forgot it

  • ADVANTAGE VRF: "Efficiency in mild weather is highly dependent on user settings and direct solar gain, but neither impact annual cost or energy." - B. Harley
  • QUIRKINESS: Most manufacturers have hand held remote controllers that are not necessarily intuitive. When Bruce's system is in "energy mode", temperature control is restricted to 2 degree increments. Some manufacturers have more user-friendly controllers (see image below) for these types of systems. Unfortunately, because of their unique way of communicating, NEST thermostats are not fully compatible with any mini-split equipment.

Mitsubishi Honeywell MHK 1 Controller for Mini-Splits

IMAGE: Mitsubishi Electric MHK-1 Controller is one of the more user-friendly controllers

  • ADVANTAGE VRF: VRF heat pumps, in the heating mode, consume less than half the energy of resistance heaters. Not a big surprise.

Resistance Heat vs Heat Pump Bruce Harley

 IMAGE: Slide from Bruce Harley's presentation showing comparison between resistance heat and mini-split heat pump energy consumption.

  • QUIRKINESS: "Conventional concepts of "right-sizing" may not be relevant.- B. Harley. While conventional heat pump systems are selected based on the cooling load, VRF heat pumps can be selected the based on their heating capacity or cooling capcity, whichever is greater. In fact, it's highly recommended and practiced to avoid the need for auxiliary heat (e.g. resistance heat), which we see can be more than twice the cost. CAVEAT: I still recommend some back up heat. It may never be used, but a good safety net.

Quirkiness of Mini-Split (VRF) Heat Pump Capacity

On the topic of the HSPF (Heating Seasonal Performance Factor, a.k.a. heating efficiency) of VRF heat pump systems, Bruce Harley said, "Be careful about what manufacturers specs you use" and "Adjust for climate using bin analysis for actual equipment and application". This is extremely important! Especially when it comes to equipment capacity. In fact, knowing the rated versus actual capacity of the outdoor unit (heat pump) and the individual air handlers is one of, if not the most critical design considerations with VRF heat pump system.

Capacities are considered "Rated" if their published via testing performed by a third-party and approved by the Air Conditioning, Heating and Refrigeration Institute, or AHRI. You can search capacity and efficiency of all "rated" equipment on their website: AHRI regulates all 3rd-party testing for all types equipment based on standard rating conditions at full capacity.

As mentioned above, mini-split heat pump systems can have a single or multiple indoor units (as many as sixty four (64) ducted or ductless air handlers) connected to a single outdoor unit (heat pump), depending on building or home size, desired zoning, required heating and cooling, etc. Although equipment efficiency is better and costs can be slightly less to have one outdoor unit serving each indoor unit, it may be more practical to have a single outdoor unit with multiple indoor units connected. The more indoor units there are, the more it makes sense to use a single outdoor unit. Whatever the configuration, the actual capacity of each indoor and outdoor piece of equipment will vary. What factors change the actual capacity?

DESIGN CONDITIONS (Factors that effect system capacity)

  1. Climate of building or home
  2. Altitude of building or home
  3. System configuration
  4. Number of connected indoor fan coils
  5. Length of line-sets (a.k.a. refrigerant lines) and number of 90 degree bends in each line set between the indoor and outdoor units.
  6. Design and orientation of building or home
  7. Location of outdoor unit (can be installed on the interior or exterior of building or home with proper ventilation)
  8. Etc., etc. etc...

Below is a screen shot of the actual capacities based on design conditions (see "Factors" above), generated with a proprietary design software from Mitsubishi Electric Cooling and Heating, one of the leading manufacturers of VRF (mini-split) equipment. For simplicity, I'm only showing the first four indoor units, but there are 10 total. The project is a custom home on a small island near Great Exuma, Bahamas. Yes, it's hot and humid. All the time. Well, almost. Maybe one day of the year it drops below 79.9.

We decided to go with the CITY MULTI line from Mitsubishi, a commercial product that requires three-phase power, for this project, but everything I'm about to show applies to all VRF equipment. It doesn't matter if it's residential-type, commercial-type, any other manufacturer, or if the project is in Anchorage or Timbuktu.

VRF Mini-Split Heat Pumps are Quriky Capacity Bahamas 3

In this configuration, I selected a 10-ton (120,000 btu/h) outdoor unit to serve all but the guest house. The model number and actual capacities at design conditions (see "Factors" above) for the outdoor unit are in the top left corner of this screen shot. (blue = total cooling, red = total heating). All of the indoor fan coils (air handlers) are concealed ducted, and their model numbers are directly above each image (to the right) of the air handler (starting with PEFY). The two numbers after the dash represent the nominal capacity (24 = 24,000 btu/h, or 2-ton), and to the right are actual capacities based on manufacturer testing at the design conditions.

If I move the house to Fargo, North Dakota, with design temperatures at 71 degrees F in Summer and -23 degrees F in Winter, here's what happens to the capacities.

VRF Mini-Split Heat Pumps are Quriky Capacity Fargo ND

The AHRI certificate below shows the rated capacities of the outdoor unit is 114,000 btu/h cooling and 129,000 heating at their standard rating conditions which is different that design conditions most of the time. If you look back to the first screen shot, the actual capacity is higher for heating and lower for cooling. That has a lot to do with the design temperatures of rated vs. actual.

VRF Mini-Split Heat Pumps are Quriky Capacity AHRI

It is normal to have actual capacities differ from rated capacities with all types of equipment, conventional or mini-split, but now let's see what happens when I change some of the other design conditions, like number of air handlers, line-set length and number of 90 degree bends in the lines.

VRF Mini-Split Heat Pumps Adding Indoor units - Copy

In the screen shot above, I've added two more 8,000 btu/h air handlers to the Bahama configuration, still using a single outdoor unit. The outdoor unit (OU) capacity has increased by about 6,000 btu/h and the indoor unit (IU) capacities have decreased by as much as 1,500 btu/h. The reason the OU capacity goes up is that I did not have 100% connected capacity (nominal capacity of indoor units = nominal capacity of outdoor unit) in the original configuration. So, when there is less than 100% connected to VRF equipment, the capacity of the OU will match the total of the indoor units.

As for the indoor units, the more units that are connected to the OU, the more the capacity of the indoor unit decreases. How much it decreases also depends on the size of each added unit. This is fairly logical, since they all have to share the same OU, but it's critical to know so there's enough capacity to match the load of each zone.

VRF Mini-Split Heat Pumps change line sets

The above screen shot shows what happens when I changed all the line-set lengths from the original configuration (10 air handlers) to fifty (50) feet from some of them being over 130 feet, and changing all of them to have only one bend. See how much the outdoor unit capacity goes up? From 99 kbtu/h to 110 kbtu/h. So, the closer the air handlers are and the fewer bends, the more capacity the system will have. You see, it's not just temperature that effects capacity.

You can also compare the actual capacities of the indoor units with their rated capacities in manufacturers published performance data. Here is an example of the PEFY-P24NMAU-E3, the first indoor unit (air handler) shown in the configurations above. It also states the rated conditions. See the difference, and see why it would be important to know the capacity?

VRF Mini-Split Heat Pumps PEFY-P24NMAU Capacity

Two Very Important Caveats

  1. These actual capacities are at design conditions, and at full capacity, which means that it's the hottest or coldest day of the year, and all zones are calling for maximum heating or cooling. at lower (Summer) or higher (Winter) temperatures, which is most of the year, the system is NOT at full capacity. These are all the considerations that must be known when specifying equipment. The milder the weather, the less demand, the more capacity available. The above results show worst case. Screen shot below shows what happens when I lower the summer design temp by 20 degrees F. Also, don't forget if some zones aren't calling, other zones will have access to more capacity.VRF Mini-Split Heat Pumps are Quriky Mild Weather Capacity
  2. Look at the cooling capacity (blue) PEFY-P06NMAU-E3 air handler. It's the fourth air handler from the top. Number on the left is total cooling, and in parenthesis on the right is sensible (ability to lower air temperature). The difference is the latent capacity (ability to remove moisture load), which is very important in climates with humidity. The Bahamas are in one of those climates. A LOT of humidity. In the case of this particular air handler, it will has very little capacity to remove moisture. If you look at the screen shot before I lowered the outdoor temp (see below), it shows no latent capacity. This is because they have very little to begin with as low capacity units, so at extreme conditions, it's going to be even lower. Whole house de-humidification is necessary in cases like this, no matter how tight the home.VRF Mini-Split Heat Pumps are Quriky Capacity Bahamas 3

"If I Had More Time, I Would Have Written a Shorter Letter" - Mark Twain

I know this is a long post, but it's an important topic that needs to be known. I hope it's clear and useful to you and your life. If you have any comments or questions, please post them below. I'd like good discussion.

-written by Chris Laumer-Giddens

Views: 69209

Tags: AHRI, Chris, HSPF, Laumer-Giddens, Mitsubishi, air, and, bahamas, bill, bruce, More…btu/h, capacity, climate, commercial, commissioning, compressor, conditioning, cooling, design, ducted, ductless, efficient, electricity, energy, exuma, flow, great, harley, heat, heating, home, homeowner, humidity, hvac, indoor, inverter, island, maryland, mini, mini-split, monitoring, or, orientation, outdoor, providence, pumps, refrigerant, residential, rhode, right-sizing, setback, split, testimonials, tight, toronto, unit, variable, window, zoning


You need to be a member of Home Energy Pros to add comments!

Join Home Energy Pros

Comment by Bruce Harley on September 26, 2014 at 12:07pm

Almost, Brett.

1. Mini splits work great in cold climates.  Especially when you get the ones that are designed for cold climates (Mitsubishi Hyper-Heat or Fujitsu RLS2 models).   We've seen thousands installed in Maine, VT, and NH with very high customer satisfaction.  Other models still work in cold climates but you have to be really aware of the specs and match them with what you expect them to do.

2.  Don't do daily temperature setbacks in the cold part of winter or the hot part of summer with mini splits.  It wastes energy,  unlike with conventional systems where it saves energy.  AND it increases convenience.  Set and forget is GOOD with mini splits, in weather where you need the unit to be running most of the time.

Hope this clarifies. Thanks

Comment by Brett Little on September 25, 2014 at 8:47am

Trying to understand your myth buster here. Mini splits do or don't work in cold climates? Set and forget it is good or not good? (With just mini splits or everything?) 

Anyway my guess is #1 They do work in cold weather, #2 Set and forget is never good.  Help me out though!

Comment by Eric Kjelshus on September 23, 2014 at 9:17pm

Its a cold coil that takes water out of air.   I have found with very long run times the coil gets colder on inside coils.  so with no set back more water vapor is taken out.  I would like to see the discussion? 

Comment by Bruce Harley on September 23, 2014 at 10:56am

Great post, Chris. Mini split heat pumps can be an effective way to offset energy use and save heating costs, and carbon, in existing homes heated with oil, LP gas or electric resistance. They are also a perfect match for new or retrofitted homes with really low heating loads. At the same time, I would like to offer a few clarifications:

1. You use the term “VRF” (variable refrigerant flow) generally to describe mini split heat pumps throughout the post, but there’s an important distinction. Variable speed heat pumps, usually inverter-driven, are split into two classes: VRF (which are tested to AHRI 1230, are usually 3-phase and >65,000 btu/h, and have the expansion device at the indoor terminal), and residential “mini splits” or “ductless heat pumps” that are tested to AHRI 210/240, are always 1-phase and <65k btu/h, and have the expansion device at the outdoor unit.

2. I would also like to remind readers that not all “mini split” or ductless heat pumps are efficient—some residential units are not variable speed (which is a fundamental part of the design of high-efficiency units). Available products go all the way down to single speed, code minimum 13 SEER / 7.7 HSPF units. AHRI even lists a few units as variable speed, with code minimum efficiencies. The message: pay attention to what you’re buying!

3. A related comment about my first “Myth Buster”: I didn’t say that “Most mini- and multi-split heat pumps are rated and have proven to deliver heating down to -20⁰ C / -5⁰ F, albeit at lower capacity”. I simply showed a picture of one unit that was rated by the manufacturer to operate at -20C, and pointed out that the installer told the client to shut it off whenever the temperature dropped below -10C. This seemed like bad advice to me, since even at a reduced capacity the unit would still produce heat more efficiently than the electric baseboard in the house.

4. It’s also true that a few mini splits are designed to operate at much higher capacity and increased efficiencies at even colder temperatures. Both Mitsubishi and Fujitsu currently have models that are rated to provide steady heating output at COPs near 2, at or below -20C (-5F). These are the “cold climate” heat pumps that people should be thinking about using. CSG is currently involved in an effort (through NEEP) to define a standard specification for cold climate heat pumps – stay tuned for developments. Of course it’s possible to use any system that’s sized for the design load at design conditions; one big advantage of variable speed designs is that the unit is not inherently oversized for more typical conditions, which saves a lot of energy. But the cold climate units will inevitably produce heat at a higher efficiency under the coldest conditions, and they also operate at lower fan speeds in nominal conditions (saving energy and noise). So I consider them a preferred option whether new or retrofit, in IECC climate zones 5 and higher.

5. Where you quote me saying the “Efficiency of VRF heat pump is better than even the manufacturers tout or than they are rated” puzzles me… and I would generally be very surprised that any systems out there would be “20-40% more efficient than promised”, unless you were using the AHRI rated nominal efficiency as your expectation, and you lived in a heating climate that is much milder than the AHRI test conditions—let’s say, Atlanta GA… this gets into the issue of understanding manufacturers’ performance data and the AHRI ratings, which I’ll discuss below.

6. The Myth Buster that temperature setbacks don’t save energy is correct… with the caveat that I was talking about heating in a cold climate. I’m sure it’s true for cooling in a hot climate as well, as you point out below. But remember the following quote “Efficiency in mild weather is highly dependent on user settings and direct solar gain, but neither impact annual cost or energy [very much]” (note the addition). By utilizing manual setbacks (really, just turning the unit off when not needed) I realized a huge increase in efficiency in the shoulder season (October in Vermont), it just didn’t affect the annual performance very much. By the same token, I don’t just leave the air conditioner “on” all summer long, because we don’t have much cooling load. The wisdom of “set and forget” really applies to the more solid heating or cooling loads; during a time when the house can “coast” for half a day or more at a time with little impact on comfort, it is probably better to just shut it off until needed. (@ Sean: the result that “set and forget” saves energy applies specifically to variable-speed heat pumps. The main reasons are that typical recovery times (early morning for heating, mid- to late-afternoon for cooling) are at the most extreme daily outdoor temperatures, coupled with the fact that during the recovery a variable-speed unit will run at or near the highest speed.  Both of these conditions result in the lowest operating efficiencies during the recovery time).  I also don’t recommend the “auto” changeover setting (heat or cool any time as needed).  Best option: set and forget for consistent cold or hot weather, and use as-needed in mild shoulder periods.

7. Capacity, sizing and manufacturers ratings: This one is a bit intricate.  First, AHRI ratings are not tested by a “third party” – the majority of manufacturers test their own equipment according to the test standards, and report the results to AHRI. It’s really just a standardized listing of equipment performance at a specified set of test conditions. It’s important to understand the test conditions. For heating, the standard conditions are 17F and 47F outdoors. The efficiency rating (HSPF) assumes an extrapolated performance at other temperatures, a particular climate (approximately Pittsburgh, PA), and a particular ratio of sizing to the design load.  So it requires the inclusion (in calculations) of some resistance heat at colder temperatures—even when the machine doesn’t include a resistance heater. These test procedures and assumptions (as well as other details) are biased against equipment that maintains a higher capacity and efficiency in colder temperatures, and that is sized not to use any resistance heat at all.

Your examples below about the variation of performance with line set length and system configuration are good, and these are all based on the manufacturer’s detailed performance data that are available to designers and installers (and sometimes to the public). My concern is not (as I think you imply) that the manufacturers may publish unreliable numbers if they are not third-party “rated”—in fact, it’s almost the opposite. AHRI rated specifications are only for general comparisons between systems, and are useless for designing a system—for all heat pumps and air conditioners, and even more so for mini splits. It’s important to delve into the manufacturers’ performance data to make sure you know what the published, factory-tested capacities and efficiencies are at a range of temperatures down to and including your outdoor design temperature (for heating). The tricky part for variable speed units is that some tables may show the “nominal” compressor speed at the lowest outdoor temperatures--highlighting the higher efficiency, at the expense of lower capacity; while other tables will show the high-speed operation with higher capacity but at a lower efficiency. (Correcting inconsistency in reporting performance data is one of the aims of the NEEP effort).  If you’re designing a system to provide adequate heating at the proper design temperature, you will want to use the high-speed compressor data. It’s OK that the system runs at the highest speed under those conditions, because it’s not for many hours annually. And although having backup heat can be a good thing in a retrofit situation (where you may be trying to offset a lot of oil or LP gas, and where the backup system is already in place)—it’s just as possible to design a mini-split system to provide 100% of the heating load without backup if that’s your objective. Just be careful to look at the manufacturers’ engineering data, and apply that to your local climate and the load you’re trying to meet.

8. Finally, about the humidity loads: Air conditioners really are not dehumidifiers. Cooling takes moisture out, but a dehumidifier has an evaporator coil to condense the water and a condenser coil that delivers “reheat” to maintain a (roughly) constant temperature while reducing humidity. In humid climates, internal and solar gains provide the reheat and you end up dehumidifying when you cool the space. If there isn’t enough reheat, you have to over-cool the space to get the humidity out—and that’s where the big difference lies between central AC systems (typically very oversized) and mini splits. With a CAC it’s easy to turn the temperature down a few degrees and quickly find the balance point that offers comfort. With mini splits, the combination of lower latent fraction AND lower total capacity leaves little margin for error. It’s even more dramatic when the mini split sees part load conditions.  The reduced refrigerant flow helps achieve the super high efficiencies by creating a de facto “oversized” coil, improving sensible heat transfer but having the opposite effect on latent removal. The point is that air conditioners are not dehumidifiers; mini splits just happen to sacrifice more latent capacity in the interest of high efficiency. There’s no shame in providing an efficient dehumidifier along with a very efficient air conditioner, it’s still cheaper to operate than a code-minimum CAC.

Thanks again, Chris.  I hope this addition is helpful. 

- Bruce



Comment by Sean Lintow Sr on September 22, 2014 at 11:08am

Quick question Chris - the set & forget (for temperature setbacks) is that with any or just leaving the temperature at one set temp? Any other changes to explain the difference?

Comment by William B. Ramsey on September 18, 2014 at 10:08am

Excellent post, thank you!

I live in a humid climate and the mini-split heat pump I have will not dehumidify the home adequately during the hot humid days of summer.  It does a great job of maintaining the temperature however.  These systems gain their efficiency through the variable speed compressor.  The units remove sufficient moisture when the compressors are operating at full speed in Dry Mode, but at the expense of efficiency.   However Dry Mode lacks temperature control with the Mits Mr. Slim unit that I own.   I resolved the humidity issue by operating a dehumidifier during the summer months.  

I contacted Mits and suggested a change to these systems to include a humidity sensor which would increase compressor speed (as needed) for additional de-humidification in summer.  Efficiency would decrease slightly but comfort would improve significantly.  The Mits folks didn't seem interested in my comments.

Forum Discussions

Origin of guarded blower door testing

Started by Kim Tanner in General Forum. Last reply by Eric Kjelshus 11 hours ago. 9 Replies

Electro Magnetic Fields (EMF)

Started by Barry L NewDelman in General Forum. Last reply by Daniel Cullen Aug 10. 5 Replies

When is it too windy to use blower door?

Started by Brandon Walton in General Forum. Last reply by Iain Walker Aug 10. 17 Replies

Latest Activity

Eric Kjelshus replied to Kim Tanner's discussion Origin of guarded blower door testing
"That same house that meet highest sealing rate in early 1980 we went back and super sealed it and…"
11 hours ago
Jill Cliburn posted an event

National Solar Conference and NZE Decathlon at University of Colorado

October 9, 2017 at 6pm to October 12, 2017 at 7pm
11 hours ago
tedkidd replied to Kim Tanner's discussion Origin of guarded blower door testing
"Brendan, Eric, great stuff!"
12 hours ago
tedkidd liked Kim Tanner's discussion Origin of guarded blower door testing
12 hours ago


  • Add Photos
  • View All

© 2017   Created by Lawrence Berkeley National Laboratory.   Powered by

Badges  |  Report an Issue  |  Terms of Service