Variable Capacity Versus Fixed Capacity

An Experiment in the Impact of Duct Location

March 01, 2014
March/April 2014
A version of this article appears in the March/April 2014 issue of Home Energy Magazine.
SHARE
Click here to read more articles about HVAC

A new generation of fully variable-capacity HVAC units has come on the market, and they promise to deliver very high cooling efficiency. They are controlled differently than standard single-capacity systems. Instead of cycling on at full capacity and then cycling off when the thermostat is satisfied, they can vary their capacity over a wide range, thus staying on for nearly twice as many hours per day as single-capacity systems. These systems have a greater impact on conductive losses of the duct system, because cold air dwells in the ductwork (commonly located in the attic or other unconditioned space) for longer periods of time.


FSEC’s variable-capacity heat pump research is conducted in its Manufactured Housing Laboratory, which features simulated-occupancy and multiple-duct systems. (David Hoak)

Peak Cooling Energy 24-Hour Composite


Figure 1. Peak-demand profiles for hot summer days representing 13 SEER and 21 SEER units, each using indoor ducts. (Charles Witers)

Table 1. Average Indoor Temperature, RH, and Cooling System Run Time for All Experimental Periods, May 1–November 30, 2010

Table 2. Average Outdoor and Indoor Temperature, Indoor RH, and Cooling System


AHUs for 13 SEER, top, and 21 SEER, bottom, 3-ton heat pumps can be connected to indoor or attic ducts. (David Hoak)


Temperature, humidity, and airflow instrumentation at entry, pictured, and exit to AHU monitors system operation in real time. (David Hoak)

In order to evaluate the delivered performance, as well as the relative performance, of a 21 SEER variable-capacity heat pump versus a 13 SEER heat pump, Building America is conducting an ongoing study. The study, which is being conducted by Building America Partnership for Improved Residential Construction (BA-PIRC), consists of four phases. Phase 1 covers the impact of duct location on performance; phase 2 covers the impact of duct leakage; phase 3 covers the impact of system sizing; and phase 4 covers the impact of duct insulation. This article, and the report from which it is adapted, address phase 1 of the project.

As part of phase 1, the BA-PIRC team ran experiments in a highly instrumented, unoccupied 1,600 ft2 house—the Manufactured Housing Laboratory (MH Lab) at the Florida Solar Energy Center (FSEC). The performance was evaluated with two different duct systems: a standard attic duct system and an indoor duct system located in a dropped-ceiling space. The design cooling load of the MH Lab is 1.5 tons, excluding any loads created by ductwork. The installed heat pumps are 3-ton units.

The Experiment

The two heat pumps were installed in the MH Lab house with the air handler units (AHUs) located side by side in a conditioned utility room. Each AHU could be attached to either the attic or the indoor duct system. Instrumentation was installed to record the energy use of the various heat pump components and other appliances, and temperature and humidity sensors were installed to record the heat pump system operation and environmental conditions indoors and in the attic. Weather data of temperature, relative humidity (RH), solar radiation, rainfall, wind speed, and wind direction were recorded. Static pressures and airflows were measured in real time for each air handler and plenum.

Automated controls were implemented to activate various internal loads (both sensible and latent heat) to simulate occupancy by a three-person family. The oven, dishwasher, and showers were automatically cycled on and off at prescribed times to provide realistic internal loads as if the house were occupied.

The 21 SEER heat pump, which has a variable-capacity compressor (varying in speed from 15 Hz to 60 Hz and in capacity from 40% to 118% of nominal) and variable-speed air handler fan operation, has two cooling modes: standard control and humidity control. In standard control, the system airflow varies generally in proportion to cooling capacity, with the exception that at lower-capacity levels the airflow rates remain relatively high, moving air at about 560 CFM per ton. In humidity control mode, the airflow rate periodically drops precipitously during lower-capacity operation for a few minutes at a time, moving only about 190 CFM per ton at the lowest system capacity.

A total of six experimental cooling configurations were examined: (1) a 13 SEER unit with attic ducts, (2) a 13 SEER unit with indoor ducts, (3) a 21 SEER unit with attic ducts, (4) a 21 SEER unit with indoor ducts, (5) a 21 SEER unit with attic ducts and RH control set to 45%, and (6) a 21 SEER unit with indoor ducts and RH control set to 45%. The attic duct system was reasonably airtight with no return leakage from outside the conditioned space. Supply duct leaks represented about 1% of system airflow. Heating experiments were also implemented using configurations 1 through 4.

Statistical analysis was used to develop best-fit lines and equations that characterize the relationship between daily cooling and heating energy use and outdoor-minus-indoor temperature. Least-squares, best-fit regression equations were developed. Coefficients of determination (r2) values were in the range of 0.85–0.97. Statistical analysis was also performed to examine cooling and heating peak demand. Least-squares, best-fit regression equations were developed to characterize the relationship between peak-hour energy use and the differential temperature of outdoor-minus-indoor temperature.

Experiments were run for both cooling and heating seasons, examining seasonal energy consumption and peak demand for both heat pumps and both duct systems.

Cooling Seasonal Savings

Based simply on SEER ratings, the 21 SEER unit should save 38.1% in seasonal cooling energy, compared to the 13 SEER unit. Experiments were performed using both standard control and RH control (45% set point) for the 21 SEER unit. Simulations were implemented using the best-fit equations and TMY3 data for Miami, Orlando, and Atlanta. The MH Lab experimental results are not much different from the simple SEER comparison, considering that experimental data are not constrained to rated conditions.

  • With indoor ducts, analysis found that the 21 SEER system produced about 36% in seasonal cooling energy savings compared to the 13 SEER system in Miami and Orlando (about 28% in Atlanta) when using standard control, and about 33.5% in seasonal cooling energy savings compared to the 13 SEER system in Miami and Orlando (about 25% in Atlanta) when using RH control.

  • With attic ducts, analysis found that the 21 SEER system produced about 36.5% in seasonal cooling energy savings compared to the 13 SEER system in Miami and Orlando (about 33% in Atlanta) when using standard control, and about 34% in seasonal cooling energy savings compared to the 13 SEER system in Miami and Orlando (about 31% in Atlanta) when using RH control.
Cooling Peak-Demand Savings

Based simply on EER ratings (95°F out, 80°F entering conditions), the 21 SEER unit should reduce cooling peak demand by 16.7%. Experiments were performed using both standard control and RH control (45% set point) for the 21 SEER unit. In the MH Lab experiments, the 21 SEER unit greatly exceeded expectations. Peak-demand savings were calculated based on an outdoor temperature of 94°F, which is very close to the summer design temperatures for Miami, Orlando, and Atlanta.

  • With indoor ducts, analysis found that the 21 SEER system produced 45.0% in cooling peak-demand savings compared to the 13 SEER system when using standard control, and 37.1% cooling peak-demand savings compared to the 13 SEER system when using RH control. Figure 1 shows a daily composite of cooling power for the indoor duct system comparing the 21 SEER and 13 SEER units.

  • With attic ducts, analysis found that the 21 SEER system produced 22.7% in cooling peak-demand savings compared to the 13 SEER system when using standard control, and 19.6% cooling peak-demand savings compared to the 13 SEER system when using RH control.
Seasonal Heating Savings

Based simply on the heating season performance factors (HSPF) of 8.0 and 9.6 for the 13 SEER and 21 SEER units respectively, the 21 SEER unit would be expected to save 16.7% in seasonal heating energy. In the MH Lab experiments, the 21 SEER unit considerably outperformed its ratings.

  • With indoor ducts, analysis found that the 21 SEER system produced on average about 40% in seasonal heating energy savings compared to the 13 SEER system in Miami, Orlando, and Atlanta.

  • With attic ducts, analysis found that the 21 SEER system produced on average about 26.5% in seasonal heating energy savings compared to the 13 SEER system in Miami, Orlando, and Atlanta.
Heating Peak-Demand Savings

Based simply on manufacturer coefficient of performance (COP) ratings (rating at 42°F delta temperature and 21 SEER medium capacity), the 21 SEER unit should reduce heating peak demand by 4.7%. Based on the experimental data from the MH Lab, the 21 SEER unit greatly exceeded expectations. Peak-demand savings were calculated based on an outdoor temperature of 30°F.

  • With indoor ducts, analysis found that the 21 SEER system produced 23.8% in heating peak-demand savings compared to the 13 SEER system.

  • With attic ducts, analysis found that the 21 SEER system produced about 21.5% in heating peak-demand savings compared to the 13 SEER system.
Impact of Duct Location

Experiments found substantial reductions in seasonal and peak-demand energy consumption when switching from attic to indoor ducts. The MH Lab house has a medium-color asphalt shingle roof. On summer days, the peak attic temperature reaches about 125°F, or 35°F warmer than outdoors, and the daily average attic temperature is about 96°F, or 14°F warmer than outdoors. The following cooling and heating seasonal savings were obtained based on regression analysis performance at a daily temperature of 82°F.

  • Switching from attic to indoor duct system produces about 17% seasonal cooling energy savings for the 21 SEER unit and about 11% seasonal cooling energy savings for the 13 SEER unit.

  • Switching from attic to indoor duct system produces 16.7% seasonal heating energy savings for the 21 SEER unit and about 10.8% seasonal heating energy savings for the 13 SEER unit.

  • The impact of locating ductwork indoors is much greater at the peak summer hour (94°F outdoors) and somewhat greater at the peak winter hour (30°F outdoors).

  • Switching from attic to indoor duct system produces 38.8% cooling peak-demand savings for the 21 SEER unit but only 14.0% cooling peak-demand savings for the 13 SEER unit.

  • Switching from attic to indoor duct system produces 14.9% heating peak-demand savings for the 21 SEER unit and 12.3% heating peak-demand savings for the 13 SEER unit.
Comfort Control

Both the 21 SEER and the 13 SEER system maintained comfortable conditions, with space temperature averaging about 76.5°F and indoor RH below 60%. Table 1 shows average measured indoor and outdoor conditions over several months of the cooling season for six experimental configurations. The main thing this table shows is that indoor humidity is lower with the oversized 13 SEER system than it is with the 21 SEER, but all configurations resulted in acceptable humidity levels. Table 2 is the same as Table 1 except that it shows only days with outdoor dew point temperatures of 70°F or greater.

During hot and humid weather, the 13 SEER system consistently produces lower indoor RH than the 21 SEER system. The 13 SEER system produces about 49% RH with either the attic or the indoor duct system. The 21 SEER system in normal control mode produces an average of 52% RH with the attic duct system and 55% with the indoor duct system. The 21 SEER system in humidity control mode (set to 45%) produces 51% RH with the attic duct system and 53% with the indoor duct system.

System run time is approximately twice as great for the 21 SEER system as it is for the 13 SEER system. This is because the 21 SEER system operates at or near minimum capacity most of the time. On a typical summer day, the 21 SEER system runs for about 16.5 hours while the 13 SEER system runs for 9 hours with the attic ducts and 7 hours with the indoor ducts.

Conclusions

In nearly all respects, the 21 SEER heat pump meets or exceeds performance expectations relative to the 13 SEER heat pump when operating with indoor ducts. Measured seasonal cooling performance falls short of estimations based on simple comparison of SEER rating by a small margin, but the other results for peak cooling performance, seasonal heating performance, and peak heating performance show that the 21 SEER unit outperforms its ratings—in some cases by a large margin.

Cooling Performance of the 21 SEER Heat Pump

While the SEER ratings of the two heat pumps would indicate expected cooling energy savings of 38.1% for the 21 SEER unit, actual seasonal savings were approximately 36% based on regression analysis and TMY3 calculations for the attic duct system. It should be understood that the systems were not confined to operating under specific rating conditions, and that differences can be expected. When the 21 SEER unit was operated in the RH control mode (45% set point) with attic ducts, actual seasonal savings were approximately 34%.

In terms of peak cooling performance, the 21 SEER greatly exceeded its ratings when examined at 94°F outdoor temperature. Based on their energy efficiency ratio (EER) ratings (13.0 and 11.8 respectively) and assuming that each system was operating at full capacity, the 21 SEER unit would produce an expected peak-demand reduction of 9.2% compared to the 13 SEER unit. In actual practice, results from the MH Lab found a cooling peak-demand reduction of 22.7% with the attic duct system, an approximate 250% level of outperformance. When the heat pumps were using indoor ducts, the MH Lab found a cooling peak-demand reduction of 45.0%, an approximate 500% level of outperformance. The key factor appears to be equipment oversizing. While the MH Lab house has a design cooling load of about 18,000 Btu/hr, the installed 3-ton units are actually oversized by 100%, excluding load associated with duct condition losses. Because the 21 SEER unit is greatly oversized, it can operate at or near minimum capacity during the hottest hours of hot summer days, and the 21 SEER unit operates much more efficiently at minimum or near-minimum capacity.

Heating Performance of the 21 SEER Heat Pump

In terms of seasonal heating performance, the 21 SEER unit greatly exceeded performance expectations. While the HSPF ratings of the two heat pumps (9.6 and 8.0) would indicate expected heating energy savings of 16.7% for the 21 SEER unit, actual seasonal savings were approximately 26.5% based on regression analysis and TMY3 calculations. When the heat pumps were using indoor ducts, the seasonal heating savings were an even more robust 40%.

In terms of peak heating performance, the 21 SEER greatly exceeded its ratings when examined at 30°F outdoor temperature. Based on manufacturer-expanded performance data, COPs for the 13 SEER and 21 SEER heat pumps are 3.01 and 3.15 respectively, when operating at 42°F delta temperature and assuming that the 21 SEER unit is operating at intermediate capacity. The indicated COPs of 3.01 and 3.15 suggest that the 21 SEER unit should only produce peak-demand reduction of 4.7% at 30°F ambient temperature. In actual practice, results from the MH Lab found peak-demand reduction of 21.5%, an approximate 500% level of outperformance. When the heat pumps were using indoor ducts, the MH Lab found peak-demand reduction of 23.8%, again an approximate 500% level of outperformance. There is no obvious explanation for this positive performance gap.

Savings from Indoor Ducts

Conductive duct losses from ductwork to the attic affect the performance of the heat pumps in both cooling and heating operation. Conductive losses of the attic ductwork create a larger energy penalty for the 21 SEER heat pump than they do for the 13 SEER heat pump, because the 21 SEER unit operates at a low capacity nearly twice as many hours per day as the 13 SEER unit. (Note that most of the losses associated with the MH Lab attic duct system are conductive losses; because there are no return leaks, air leakage of the supply ducts represents about 1% of the system airflow, and the AHUs and return ducts are in the conditioned space.) Attic temperatures during typical summer weather have a daily average of about 96°F and an average afternoon peak of about 125°F. When using the MH Lab’s essentially leak-free attic ductwork, the 21 SEER and 13 SEER heat pump systems respond as follows when going from the attic duct system to the indoor system.

  • On a typical summer day, cooling energy decreases by 16.8% for the 21 SEER and 11.2% for the 13 SEER unit.

  • On a peak summer afternoon (94°F), peak cooling energy decreases by 38.8% for the 21 SEER unit and 14.0% for the 13 SEER unit.

  • On a typical central Florida winter day, heating energy decreases by 16.7% for the 21 SEER unit and 10.8% for the 13 SEER unit.

  • On a peak central Florida winter morning (30°F), peak heating energy decreases by 14.9% for the 21 SEER unit and 12.3% for the 13 SEER unit.

Therefore, there are significant benefits to locating the ductwork indoors, especially for the 21 SEER unit and especially during the peak cooling hours.

While this work characterizes the effects of duct conductive losses, it does not indicate the effects of duct air leakage. Experiments on the effects of duct leakage have been carried out, with results that will soon be available to the public.

learn more

DOE’s Building America program has published a technical report—Energy Savings and Peak Demand Reduction of a SEER 21 Heat Pump vs. a SEER 13 Heat Pump with Attic and Indoor Duct Systems—that discusses the findings of this study in detail. Download a PDF of the complete report.

The phase 1 results of the 21 SEER testing show that oversizing variable-capacity heat pumps reduces seasonal and peak energy consumption.

If it turns out that considerable oversizing of variable-capacity systems is as beneficial as the findings in this report suggest, this will require alterations to widely accepted sizing guidelines, utility incentive criteria, and building code language. As this research unfolds, it may become clear that oversizing should be encouraged as best practice for variable-capacity, and perhaps two-stage, A/C and heat pump systems.

 

James Cummings is a program director at the Florida Solar Energy Center. Charles Withers is a senior research analyst at the Florida Solar Energy Center.

This article was adapted from a DOE Building America program report.

  • 1
  • FIRST PAGE
  • PREVIOUS PAGE
  • NEXT
  • LAST
Click here to view this article on a single page.
© Home Energy Magazine 2021, all rights reserved. For permission to reprint, please send an e-mail to contact@homeenergy.org.
Discuss this article in the HVAC group on Home Energy Pros!

Comments
Add a new article comment!

Enter your comments in the box below:

(Please note that all comments are subject to review prior to posting.)

 

While we will do our best to monitor all comments and blog posts for accuracy and relevancy, Home Energy is not responsible for content posted by our readers or third parties. Home Energy reserves the right to edit or remove comments or blog posts that do not meet our community guidelines.

Related Articles
SPONSORED CONTENT 2021 National Conference Registration is open! Learn more! Watch Video