Home / Comparisons / 12,000 BTU wall-mounted heat pumps
DURASTAR Sirius Heat R-454B and the MIDEA28: spot the differences
Two 12,000 BTU wall-mounted units on R-454B, both ENERGY STAR cold climate certified and offered by AirGreen. Are they the same? Their AHRI certificates answer: laid one on the other, 27 of their 52 measured values are identical and 25 differ. This page shows which ones, then what each manufacturer publishes about cold weather, sound and installation.
Sirius Heat R-454B 12,000 BTU
DRAW12F2A / DRA1H12S2A · AHRI 215359252
See the product page
MIDEA28 12,000 BTU
MSEHU-H12B-2A / MO1HS-H12B-2A · AHRI 215471381
See the product pageSame AHRI and NEEP certified values: SENVILLE AURA R454B, DIRECT AIR HEAT XTREME PRO R-454B and BOSCH Climate 5000 MAX PERFORMANCE R-454B, in 12,000 BTU. Verified in NEEP on November 10, 2026.
The 52 measured values of the two certificates, one cell each. See the differences
same figure in both certificates different figures
What the electricity costs, in dollars, winter and summer
AirGreen estimate for the same 12,000 BTU/h zone in Montréal, at Hydro-Québec’s Rate D, taxes included. Both heat pumps are calculated in exactly the same way, with the same method as on all our comparison pages; only their certified efficiencies change. Here, the Region V HSPF2 is the same, 8.0: the winter amount is therefore the same for both, and the gap comes from summer, where the SEER2 is 24.0 and 25.4. Electric baseboards serve as the reference point.
How this amount is calculated
- Heat to deliver over the winter. A zone whose heat loss reaches 12,000 BTU/h ( kW) at −23.1 °C, Montréal’s design temperature: kW × 24 h × 4,167 degree-days ÷ (18 − (−23.1)) = kWh of heat for round-the-clock heating, all winter. Heating degree-days (base 18 °C) and design temperature: Natural Resources Canada, HOT2000 climate data, Montréal Intl station. With fewer hours per day (“Heating” slider), the heat is reduced in the same proportion, for the heat pump as for the baseboards: this simplification does not account for recovery after a temperature setback.
- Electricity for heating. Heat pump: heat × 3.412 ÷ HSPF2 Region V, the AHRI certificate value published by NEEP. Region V is the coldest in the standard, and its calculation includes the electric backup expected when the heat pump is no longer enough. Electric baseboards: 1 kWh of electricity for 1 kWh of heat.
- Electricity for cooling. 12,000 BTU/h ÷ SEER2 × hours per day × 60 days ÷ 1,000: the method of Hydro-Québec’s air-conditioning calculator, with the SEER2 seasonal efficiency from the AHRI certificate. The number of hours is a scenario you can set above, from 1 to 24 h. At 24 h, the calculation assumes rated capacity all day long: that is a ceiling, since a variable-speed unit slows down once the room is at temperature.
- Price. Hydro-Québec Rate D in effect April 1, 2026: 7.065¢ per kWh up to 40 kWh a day, 11.142¢ beyond; the average price, 9.104¢, is the one Hydro-Québec’s calculators use. GST and QST included. Network access charges, the same with or without a heat pump, are not counted.
AirGreen method v1.1, November 9, 2026. An estimate is used to compare units under the same conditions. The actual bill depends on the house, the insulation, the settings and the use; this amount is not a promise.
Two certificates laid one on the other: 27 values overlap, 25 do not
For each AHRI certificate, NEEP publishes a capacity, a COP and an electrical power at minimum, rated and maximum output, at six test temperatures, plus four seasonal ratings. That makes 52 values per unit. Each cell lays two of them one on the other: a single figure when both certificates say the same thing, two figures when they differ.
Differences found: 25 of 52
Values compared: SEER2, EER2, HSPF2 for Regions IV and V, then capacity, COP and electrical power at 35 °C and 27.8 °C in cooling, and at 8.3, −8.3, −15 and −30 °C in heating. NEEP gives no rated value at 27.8 °C or at −30 °C. Sources: AHRI certificate 215359252 (NEEP record 389518) and AHRI certificate 215471381 (NEEP record 387358), read on November 10, 2026.
Below −15 °C: same certified points, one more figure from DURASTAR
NEEP certifies both units at −15 °C and at −30 °C, with the same maximum capacities: 11,500 and 5,900 BTU/h. In between, DURASTAR publishes a capacity at −25 °C in its submittal; MIDEA publishes none in its brochure, its submittal or its service manual. Where a unit has no figure, the table gives the two published values on either side of that temperature.
At −15 °C and at −30 °C, the two certified points fall in the same place; they are shifted by a few pixels so that both stay visible.
A submittal on one side, a brochure and a submittal on the other
DURASTAR submittal DRAW12F2A / DRA1H12S2A (form V1.2_0325); MIDEA Canada wall-mounted systems brochure (2025-08-20), whose feature table is common to the MIDEA35, 28 and 25 series, and MIDEA submittal for the EP Xtreme Heat series. Base pan heater: data declared by each manufacturer to NEEP.
Sirius Heat: heating and cooling advertised down to −30 °C; base pan heater included, 150 W declared to NEEP.
MIDEA28: heating and cooling advertised down to −30 °C; 150 W base pan heater declared to NEEP.
Same operating rule declared to NEEP for both: the pan heater runs for 5 minutes when it is below 3 °C outside for 30 seconds, after 25 minutes of operation.
Sirius Heat: A2L refrigerant detection system.
MIDEA28: A2L dispersal system.
Sirius Heat: wireless remote control with “Follow Me”; compatible with the DRSTAT101 wired thermostat. Wi-Fi is not mentioned in the submittal.
MIDEA28: wireless remote control, “Follow Me”; Wi-Fi control, through a Wi-Fi dongle or the wired controller with built-in Wi-Fi according to the submittal; optional wired controller and optional 24 V interface for a third-party thermostat; dry contact.
Sirius Heat: four published airflows, from 153 to 383 CFM.
MIDEA28: fan speed adjustable from 1 to 100%; five published airflows, from 147.2 to 441.4 CFM; horizontal and vertical swing.
Sirius Heat: Turbo, ECO, Active Clean and Sleep; auto restart; freeze protection at 8 °C.
MIDEA28: ECO, “Gear”, Turbo, sleep mode, silent mode, 8 °C heating, timer, child lock; intelligent sensor (“Intelligent eye”).
Sirius Heat: Active Clean mode; “Golden Fin” coating on the indoor and outdoor coils.
MIDEA28: automatic cleaning; removable fan wheel; TU1 copper indoor coil; humidity sensor.
A feature missing from a document is not necessarily missing from the unit: this page only repeats what each document writes.
Four published speeds from DURASTAR, five from MIDEA
DURASTAR publishes four speeds for the Sirius Heat, from “Silent” to “High”. MIDEA publishes five for the MIDEA28, from “Silent” to “Turbo”. Both give a level in dB(A), without the measuring distance.
How to read these figures. dB(A) measures a noise level as the ear perceives it: the lower the figure, the quieter the unit. Each dot is a fan speed published by the manufacturer, from the slowest to the fastest.
Why this is not a ranking. Neither document gives the measuring distance or conditions. The airflows are not the same either: at “High” speed, 294 CFM for the Sirius Heat and 323.7 CFM for the MIDEA28. A gap of a few dB(A) can come from the method or the airflow as much as from the unit.
Indoor unit, in the room
All published speeds
Outdoor unit, outside
A single published value per manufacturer
Measurements taken by each manufacturer using its own method: read them separately, not as a ranking.
Sources: DURASTAR submittal V1.2_0325 (“Sound pressure”); MIDEA Canada wall-mounted systems brochure (2025-08-20), p. 11. The MIDEA submittal rounds the same speeds to 20, 22, 33, 37 and 47 dB(A).
On the wall and outside: the published dimensions, overlaid
Is your wall wide enough?


Width of the unit only, to scale. The side clearances required by the installation manual are added; the installer confirms them on site.
The casings, one on the other
Sirius Heat in dashed line, MIDEA28 in solid line. Diagrams drawn from the dimensions published by each manufacturer, not photos; the circle is indicative.
Indoor units to scale


Width × height × depth per the manufacturers. DURASTAR publishes fractions of an inch (31-9/16 × 11-5/8 × 7-7/8 in), converted here to decimals. Photos scaled by width.
| Installation | Sirius Heat R-454B | MIDEA28 |
|---|---|---|
| Maximum piping length | 82 ft25 m; published minimum: 10 ft. | 82 ft25 m. |
| Precharged length | 25 ft7.5 m. | 24.6 ft7.5 m; the submittal rounds to 25 ft. |
| Additional charge beyond the precharged length | 0.16 oz/ft | 0.16 oz/ft |
| Maximum height difference | 49 ft15 m. | 49.2 ft15 m. |
| Liquid – gas diameters | 1/4 – 3/8 in | 1/4 – 3/8 in |
| Outdoor unit W × H × D, weight | 31.69 × 21.81 × 13 in, 72.8 lb805 × 554 × 330 mm, 33.0 kg. | 31.69 × 21.81 × 12.99 in, 72.75 lb805 × 554 × 330 mm, 33 kg. |
| Indoor unit W × H × D, weight | 31.56 × 11.63 × 7.88 in, 19 lb802 × 295 × 200 mm, 8.6 kg. | 31.30 × 11.61 × 8.86 in, 22.93 lb795 × 295 × 225 mm, 10.4 kg. |
| Electrical protectionPer the manufacturer | Recommended breaker 15 A (MCA 15 A) | Max. fuse 15 A (MCA 15 A) |
| Outdoor unit clearances | More than 4 in and more than 24 in depending on the sideFigure in submittal V1.2_0325; the side for each value is read on the figure. | From 4 to 24 in depending on the side and on the clearance of the other sidesMIDEA service manual, figure SC-1, p. 11; the side for each value is read on the figure. |
| Outdoor unit base pan heaterDeclared by the manufacturer to NEEP | Yes, 150 W | Yes, 150 W |
| Refrigerant and factory charge | R-454B, 35.3 oz (1.00 kg) | R-454B, 35.27 ozBrochure; the submittal writes 35 oz (1 kg). |
| Dehumidification | 1.4 L/h | 1.2 L/hEP Xtreme Heat submittal; the brochure does not print it. |
| Rated heating / cooling range | −30 to 24 °C / −30 to 50 °C | −30 to 24 °C / −30 to 50 °C |
Piping, charge and ranges: same figures in metres and grams in both documents (25 m, 7.5 m, 15 m, 15 g/m, 1.00 kg). The small gaps in feet and ounces come from the rounding in each document.
Eight questions about the Sirius Heat and the MIDEA28
Is the DURASTAR Sirius Heat a MIDEA28?
No, not in terms of the certificates. They have two distinct AHRI certificates (215359252 and 215471381), and 25 of their 52 measured values differ. What they have in common: the maximum heating capacities at the four test temperatures, the HSPF2 ratings for Regions IV and V, the R-454B refrigerant and the published dimensions of the outdoor unit. None of the documents consulted says that one is made by the other brand, and this page does not assume it.
Which one heats more at −15 °C and at −30 °C?
According to their certificates, neither: 11,500 BTU/h maximum at −15 °C and 5,900 BTU/h at −30 °C for one as for the other, with the same COPs at those capacities, 1.80 and 1.25. Between those two temperatures, only DURASTAR publishes a figure: 7,500 BTU/h at −25 °C, in its submittal, not certified.
So where are the differences?
Mostly in cooling and at low output. The SEER2 is 24.0 for the Sirius Heat and 25.4 for the MIDEA28; the EER2, 13.2 and 13.4. At 27.8 °C, all six cooling values differ. In heating, it is the values at minimum output and a few rated values that change. The spot-the-differences board shows them one by one.
Does it change the electricity bill?
In our estimate, winter comes to the same amount, because the Region V HSPF2 is the same, 8.0. Summer differs a little, depending on the SEER2. The cost block does the math with your hours of use.
Why does the DURASTAR submittal say 5,800 BTU/h at −30 °C, and this page 5,900?
The DURASTAR submittal (form V1.2_0325) prints 5,800 BTU/h at −22 °F. AHRI certificate 215359252, as carried by NEEP, shows 5,900 BTU/h. When two sources disagree, this page follows the certificate and shows the other figure in the Extreme cold section. The same rule applies at −8.3 °C: the submittals print a rated capacity of 10,000 BTU/h (DURASTAR) and 9,800 BTU/h (MIDEA), and NEEP, 12,000 BTU/h for both.
Do the two units have the same dimensions?
Outside, yes according to the manufacturers: 805 × 554 × 330 mm and 33 kg for both outdoor units. Inside, no: the Sirius Heat indoor unit measures 802 × 295 × 200 mm and weighs 8.6 kg; that of the MIDEA28, 795 × 295 × 225 mm and 10.4 kg.
What about the SENVILLE AURA, the DIRECT AIR HEAT XTREME PRO or the BOSCH MAX PERFORMANCE?
In 12,000 BTU on R-454B, their certificates carry the same AHRI and NEEP certified values as that of the MIDEA28, field by field. What is written here about the certified values of the MIDEA28 therefore also applies to those three certificates; the indoor units, remote controls and documents remain those of each brand. See SENVILLE AURA R454B and DURASTAR Sirius Heat.
Are they eligible for grants?
Eligibility depends on the official list in force on the date of purchase. This page promises no amount; the current list is checked at the time of the quote.
The two certificates in full
| Certified value | Sirius Heat R-454B | MIDEA28 |
|---|---|---|
| Seasonal efficiency | ||
| SEER2 | 24.0 | 25.4 |
| EER2 | 13.2 | 13.4 |
| HSPF2 Region IV | 10.5 | 10.5 |
| HSPF2 Region V | 8.0 | 8.0 |
| Rated capacity at 35 °C | 12,000 BTU/h | 12,000 BTU/h |
| Cooling, same conditions | ||
| 35 °C: capacity min. / rated / max.BTU/h | 3,600 / 12,000 / 13,500 | 3,800 / 12,000 / 13,500 |
| 35 °C: COP min. / rated / max. | 8.79 / 3.86 / 3.35 | 8.57 / 3.91 / 3.35 |
| 35 °C: power min. / rated / max.kW | 0.12 / 0.91 / 1.18 | 0.13 / 0.90 / 1.18 |
| Cooling in mild weather | ||
| 27.8 °C: capacity min. / max.BTU/h | 3,600 / 14,300 | 4,100 / 15,100 |
| 27.8 °C: COP min. / max. | 8.79 / 6.98 | 9.24 / 6.51 |
| 27.8 °C: power min. / max.kW | 0.12 / 0.60 | 0.13 / 0.68 |
| Heating, same conditions | ||
| 8.3 °C: capacity min. / rated / max.BTU/h | 3,800 / 12,000 / 16,000 | 3,800 / 12,000 / 16,000 |
| 8.3 °C: COP min. / rated / max. | 4.84 / 3.63 / 3.35 | 4.84 / 3.74 / 3.35 |
| 8.3 °C: power min. / rated / max.kW | 0.23 / 0.97 / 1.40 | 0.23 / 0.94 / 1.40 |
| −8.3 °C: capacity min. / rated / max.BTU/h | 5,200 / 12,000 / 12,800 | 5,300 / 12,000 / 12,800 |
| −8.3 °C: COP min. / rated / max. | 2.34 / 2.44 / 1.75 | 2.22 / 2.51 / 1.75 |
| −8.3 °C: power min. / rated / max.kW | 0.65 / 1.44 / 2.14 | 0.70 / 1.40 / 2.14 |
| −15 °C: capacity min. / rated / max.BTU/h | 4,000 / 11,500 / 11,500 | 4,100 / 11,500 / 11,500 |
| −15 °C: COP min. / rated / max. | 1.80 / 1.80 / 1.80 | 1.72 / 1.80 / 1.80 |
| −15 °C: power min. / rated / max.kW | 0.65 / 1.87 / 1.87 | 0.70 / 1.87 / 1.87 |
| Coldest point — same temperature for both, not highlighted | ||
| Test temperature | −30 °C (−22 °F) | −30 °C (−22 °F) |
| Capacity min. / max. | 2,000 / 5,900 BTU/h | 2,000 / 5,900 BTU/h |
| COP at minimum / maximum capacity | 1.07 / 1.25 | 1.17 / 1.25 |
| Power at minimum / maximum capacity | 0.55 / 1.38 kW | 0.50 / 1.38 kW |
| Listings and refrigerant | ||
| ENERGY STAR / cold climate (NEEP) | Yes / Yes | Yes / Yes |
| ENERGY STAR directory, markets listed | United States, Canada | United States, Canada |
| AHRI certificate · NEEP record | 215359252 · 389518 | 215471381 · 387358 |
| RefrigerantGWP per the IPCC (AR4); ASHRAE 34 class | R-454B · GWP 466 · A2L | R-454B · GWP 466 · A2L |
higher marks the higher value of the same measure, under the same conditions. It is not a recommendation. Per-point capacities, COPs and power values are not highlighted: the order changes depending on the point read.
Both are on the AirGreen website.
Undecided between the two? Give us the room size, the planned outdoor location and your current heating: our team will look with you at what fits.
Where each figure comes from
Data verified November 10, 2026. Performance: AHRI certificates and the NEEP database. Dimensions, sound, piping, refrigerant charge and features: according to the manufacturer. When a source contradicts the AHRI certificate, the certificate prevails.
Trademarks. DURASTAR, Sirius Heat, MIDEA and MIDEA28 are trademarks of their respective owners. Logos are used only to identify the products compared. AirGreen is a dealer for both brands; this page is not produced, sponsored or endorsed by the manufacturers.
Photos: AirGreen and manufacturers, for identification.
Does a figure look wrong to you? Write to us: any reported value is checked again.
