Central Valley Research Homes: Caleb House 2018 Cooling Season and 2018-19 Heating Season Evaluation of Ducted and Ductless Multi-Split Variable Capacity Heat Pumps

COOLING SEASON CONCLUSIONS 
Differences in projected normalized annual cooling energy use for the single speed reference HP, seven-head ductless multi-split VCHP, and two-head ducted multi-split VCHP systems were very slight. The ductless VCHP used the least cooling energy overall of the three tested systems, both during constant setpoint days of operation and on recovery days. Compared to the reference HP system, the ductless VCHP used 5.4% less energy on constant setpoint days, and 5.9% less on recovery days. Compared to the ducted VCHP, the ductless VCHP used 7.3% less cooling energy on constant setpoint days and 1.1% less energy on recovery days. These savings are substantially less than SEER ratings would lead decision makers to expect. 

When evaluating the cooling electricity demand of the three systems, there was no system that was clearly the best. This is not unexpected considering the relatively low EER values of the VCHP systems. During constant setpoint operation, and under moderate cooling conditions on recovery days, VCHP systems daily peak electricity demand was 59% to 87% of the reference HP demand under the same conditions. But when operating at the start of the recovery period, the electricity demand for the VCHP was 107% to 111% of the reference HP demand on hot days.  Comparing the electricity demand of the two VCHP systems with each other, the ductless VCHP appeared to draw less power than the ducted system at cooler conditions and slightly more power at the hottest recovery day conditions. 

An interesting note about system power draw was that the ductless VCHP system used the most energy during standby conditions. While the two ducted systems typically drew less than 25 W during standby, the ductless VCHP drew over 55 W in standby mode. This extra power was used by the individual ductless units, with wall-mounted units drawing about 2 W apiece, and the ceiling-mounted units drawing almost 10 W each during standby. While these values are small, over a year they accumulate to a large number of kWh.

The ductless VCHP delivered much better comfort to the house than the ducted systems during the cooling season. Room temperatures during ductless VCHP operation did not vary as much as they did when ducted systems cycled on and off. The ductless VCHP also kept temperatures in very good control in the rooms with the greatest solar loads, bedrooms 1 and 2. Despite adjusting manual dampers to balance ducted air flows to these rooms, temperatures were consistently at least 4°F above setpoint during reference HP and ducted VCHP operation on sunny afternoons. 

All three systems had no problems meeting comfort requirements for indoor humidity during the cooling season. Overall, the ductless VCHP delivered greater indoor comfort than the ducted VCHP and reference HP systems. It also used 5 to 6% less cooling energy than these systems. However, VCHP cooling electricity demand was potentially higher than reference system demand when temperature setpoints were lowered during evening peak hours. 

HEATING SEASON CONCLUSIONS
The HSPF 9.3 ducted VCHP system was estimated to use 1,446 kWh annually for heating, or 32% less than the HSPF 8.2 reference HP. The ducted VCHP used the least heating energy across all conditions in the relatively mild Stockton winter climate, with ADOT above 40°F and winter design temperature of 33ºF. Despite its efficient HSPF 9.8 rating, the ductless VCHP used the most heating energy of all three tested systems. It was estimated to use 2,250 kWh annually for heating, 6% more than the 2,114 kWh of the HSPF 8.2 reference HP. 

When comparing daily peak electricity demand in heating mode, the ducted VCHP had lower demand than the reference HP. The ducted VCHP reduced peak demand in the morning to 77-81% of reference HP demand at the same conditions, and reduced evening peak demand to 73-77% of reference HP demand. The ductless VCHP did not keep demand below reference HP levels. Ductless VCHP morning peak demand was 89% of reference HP demand on 45°F ADOT days, but 133% of reference HP on 55°F ADOT days. The ductless VCHP evening demand was 139-150% of reference HP demand over all conditions. 

Note that VCHP systems are able to provide winter heating without using electric resistance heat. Standard heat pumps are usually equipped with auxiliary strip heaters, which are poorly controlled, often operate when they’re not needed, and use large amounts of energy and power. The reference HP tested at Caleb did not have strip heaters, and backup ER heaters did not activate during constant setpoint operation. Therefore VCHP energy and demand savings for the VCHPs in this study may be conservative, since any savings from strip heat were not included. 

The ducted VCHP system kept indoor air temperatures most uniform and closest to the 68°F target over the entire day, in all rooms and all outdoor conditions. The reference heat pump was next best at keeping room temperatures close to 68°F. 

The ductless VCHP system significantly overheated the house, except for the master bedroom and master bath, by 2 to 7°F. Overheating was most pronounced in bedrooms 1, 2 and 3 on the second floor, and was larger on colder days and during colder nighttime hours. The ductless VCHP used the smallest available air handlers rated at 8,100 Btu/hr heating capacity in each of seven rooms. But the capacity of each indoor unit exceeded the heating loads of the rooms by a factor of up to 7.5. 

Studying the ductless VCHP air handlers more closely, they were found to operate in a manner that delivered unnecessary heat to the house at the expense of occupant comfort and energy use. Two behaviors were noted: 1) air handler power levels were mostly in “quiet” or “low” mode and did not drop to “standby” mode, indicating that they were constantly circulating air, and 2) supply temperature was elevated, so indoor unit fans were circulating warm supply air to rooms that were over setpoint. 

Looking at cycling behavior of systems during the heating season, the ductless VCHP compressor operated for far more hours per day (7.3 to 16.9 hours) than the ducted VCHP compressor (1.0 to 8.6 hours), even though the outdoor units were identical, of the same manufacturer, model and size. Calls for heating from the master bedroom continually drove ductless compressor operation and supplied heat to the master bedroom, as well as to rooms that were already overheated.

Over five days of ductless VCHP operation in February, electric resistance (ER) heaters were used to heat the master bedroom. This was done under the theory that the master bedroom, having the largest need for heating, was driving the ductless system operation and the subsequent overheating throughout the rest of the house. Using auxiliary ER heat in the master bedroom did decrease overheating by about 1°F in the rest of the house, but created a large energy penalty.

Keyword Search:
Heat pump, HVAC, VCHP, all-electric, electrification