1
Residential Variable Capacity Heat Pump Field and Lab Testing: Final Results
20 November 2013
Ecotope
2
History• Inverter-driven compressor tech making increasing inroads
(ductless systems now at 20,000 plus installs in NW); BPA has incented central ducted systems via PTCS on provisional basis
• Advisory committee formed to guide research plan– Range of experience on committee– Interest in zoned systems, delivery temperatures – Timeline/budget limited research scope
• Study summary– Heating season only– ‘Detailed’ (full capacity/COP) vs non-detailed sites – Combination of field/lab info (EPRI) – ASHRAE 152 duct model exercised plus SEEM for full season
estimates
3
Major Findings – Expectations/Reality
• VCHP performs 25-30% better than a single speed 7.9 HSPF heat pump on an annual basis (note nominal average HSPF for this product ~12).
• Duct losses increased ~5% over a single speed heat pump
• Overall system offers improved performance over single speed base cases
• Heat pump sizing still matters• Auxiliary heat lockout control still matters
4
Field Deployment• Planning/refinement occurred through late 2012• Field deployment set for late February, 2013 (six sites);
central OR chosen for likelihood of remaining cold weather (about 24 potential sites)
• Critical points measured– 5 minute metering for electricity usage (including air handler
circuit power- CFM mapping)– One-time measurements of house heat loss (including blower
door test (system sizing, heating season simulation)), duct area/R-values, duct leakage, system CFM (at least 3 points so CFM could be correlated with power), heat pump control settings
5
Site Pictures
6
Site SummariesSiteid Location OD
unit size (tons)
House Type
Heated Floor Area (ft2)
Occs PTCS HP?
PTCS Ducts?
Strip heat lockout temp.
91001 Powell Butte, OR 3 Site 2694 2 Yes Yes not set
91002 Powell Butte, OR 2 Mfd 1394 2 No No 35
91004 Bend, OR 3 Site 2663 2 Yes No not set91005 Bend, OR 3 Site 2515 2 Yes Yes not set91009 Redmond, OR 4 Site 1860 2 No No 35
91010 Redmond, OR 4 Site 2660 2 No No 35
7
Duct Leakage/Air Handler Data
Site ID
Both sides duct leak to
out at 50 Pa*
(SCFM)
Supply duct leak to out at 50 Pa*
(SCFM)
Reference‡ Air Handler Flow(CFM, SCFM)*
Reference‡ supply
static pressure
(Pa)
Reference‡ return static
pressure (Pa)
91001 231*** 151*** 943, 1061 18 -84.5
91002 276 275 688, 774 16 -21
91004 n/a n/a 889, 1000 21.5 -48
91005 273 131 924, 1040 26 -72.5
91009 329 208 1340, 1395 50 -139
91010 148 131 1271, 1430 116 -158* Leakage and air handler flow results corrected to standard air (68°F and 1 atmosphere). The elevation of houses in the Bend area (about 3,000 ft above sea level) results in lower air density. More specifically, the density is about 88.9% of the density of standard air. The air handler flow values show both the local CFM and the standard SCFM.‡ “Reference” airflow corresponds to the supply and return static pressure measurements shown in the table. Typically this airflow represents the highest flow that could be attained using the User Interface (thermostat); this measurement was taken to make sure the air handler was not working against an extreme external static pressure (above 200 Pa) at its highest flow. No adverse static pressure conditions were found. All of these systems were set up by the installer in COMFORT mode (so maximum flows typically average 325-350 CFM/nominal ton of capacity).***These measurements taken at 45 Pa
8
Airflow & Fan Power
• Site 91009
0 100 200 300 400 500 600 700 8000
200400600800
1000120014001600
f(x) = 156.762972904162 x^0.33989972231259
Fan Power (W)
Airfl
ow (C
FM)
9
Air Handler Curves - All Sites
10
Balance Points (for 2 sites)
11
Heat Pump Balance Points
Site ID Peak total UA† (BTU/hr °F)
DHL base‡ (BTU/hr)
DHL w/DE††
(Btu/hr)
HP Balance Point
(°F)
Heat Pump Size
(tons)
91001 896 63,638 97,122 41 391002 283 19,260 26,814 12 291004 1126 72,632 72,632 36 391005 580 40,996 69,719 26 391009 502 37,452 72,367 20 491010 632 42593 66391 17 4
* All Design Heating Load (DHL) calculations based on 4° F design temperature.†Shell plus infiltration heat loss at heating design temperature‡Design Heating Load without duct losses at the design temperature††Design Heating Load with duct losses included
12
Detailed Field Results(834 monitored hours)
13
Capacity and COP Calculations
• Capacity determined by multiplying fitted airflow by coil temperature split
• True RMS power is measured at the panel for both indoor and outdoor units and converted to Btu/hr
• COP is ratio of capacity to input power at a given outdoor temperature
• Following 2 graphics do not include duct losses
14
Performance vs Outdoor Temp (1)(no duct effects included)
15
Performance vs Outdoor Temp (2)(no duct effects included)
16
Measured Field Performance (common five week period)
Site ID
Average Outdoor Temp. (F)
Total Input (kWh)
Total Output (kWh)
COP (avg)
ER heat output fraction
Metered Time Span
(hrs)
91001 38.4 716 1979 2.8 0.05 834
91002 39.3 277 627 2.3 0.09 834
91004 40.9 1185 2475 2.1 0.19 834
91005 39.4 861 1172 1.4 0.47 834
91009 38.2 662 1485 2.2 0.04 834
91010 37.3 706 1457 2.1 0.07 834
17
Compressor Response to Changing Heating Load
• Low compressor speed: input power < 1.25 x observed minimum input• Medium compressor speed: anything in between low and high• High compressor speed: input power > 0.87 x observed maximum input
18
Air Handler Response to Changing Heating Load
• Low fan speed: airflow < 1.25 x listed minimum airflow (comfort mode)• Medium fan speed: anything in between low and high• High fan speed: airflow > 0.87 x listed maximum airflow (comfort mode)
19
Operating Mode Summary• Fan and compressor speeds track one another
– Given a compressor speed, there is a preferred fan speed– High compressor outputs almost never pair with low and medium fan – Low compressor speeds pair with both low and medium fan
• Runtime Fractions:Compressor Only Operation
Compressor Speed
Fan SpeedLow Medium High All
Low 0.26 0.14 0.09 0.48Medium 0.10 0.20 0.09 0.39
High 0.01 0.01 0.12 0.13All 0.37 0.34 0.29 1.00
Compressor and Auxiliary OperationCompressor
SpeedFan Speed
Low Medium High AllLow 0.24 0.12 0.08 0.45
Medium 0.09 0.18 0.09 0.36High 0.01 0.01 0.18 0.19All 0.34 0.31 0.35 1.00
20
Rolling in the Ducts…
21
Duct System Inputs
• Duct leakage, normalized to fraction of air handler flow (which changes at different flows)
• Duct R-value• Duct area• Buffer space temperature (measured only for supply
buffer space)• ASHRAE 152 used to calculate distribution efficiency
𝐷𝐸=𝛼𝑠𝑢𝑝𝑝𝑙𝑦 𝛽𝑠𝑢𝑝𝑝𝑙𝑦−𝛼𝑠𝑢𝑝𝑝𝑙𝑦 𝛽𝑠𝑢𝑝𝑝𝑙𝑦 (1− 𝛽𝑟𝑒𝑡𝑢𝑟𝑛𝛼𝑟𝑒𝑡𝑢𝑟𝑛 )𝛥𝑡𝑟𝑒𝑡𝑢𝑟𝑛𝛥𝑡 𝑠𝑦𝑠𝑡𝑒𝑚
−𝛼𝑠𝑢𝑝𝑝𝑙𝑦(1− 𝛽𝑠𝑢𝑝𝑝𝑙𝑦)𝛥𝑡 𝑠𝑢𝑝𝑝𝑙𝑦𝛥𝑡𝑠𝑦𝑠𝑡𝑒𝑚
22
Average Duct Performance and Calculated Duct Distribution Efficiency (DE)
(common five week period)
Site ID
System Airflow (CFM)
Avg Supply Buffer Zone
Temp (°F) SLF RLF DE
91001 675 56.8 0.08 0.11 0.65
91002 550 55.2 0.13 0 0.82
91004 722 n/a 0 0 1.0
91005 766 55.9 0.07 0.17 0.61
91009 748 65.3 0.18 0.19 0.49
91010 1159 57.7 0.13 0.02 0.68
23
Modeling VHCPs – Annual Performance
• Annual performance estimated with SEEM• Simulation largely unchanged from September
2012 RTF presentation– Performance curve from catalogue data found to
work well enough for systems modeled– Field study found that the operating modes pair
fan and compressor speed (previous default assumption in simulation – remains unchanged)
24
Modeled VCHP and Duct System Efficiency (using SEEM for entire heating season)
Equipment Efficiency
Site Electric
Furnace, COP=1
Single Speed Heat
Pump, HSPF=7.9
Variable Speed Heat
Pump
91001 1.00 1.99 2.4691002 1.00 2.35 2.8491004 1.00 2.12 2.5991005 1.00 2.07 2.5391009 1.00 2.48 3.0591010 1.00 2.20 2.76
Average 1.00 2.20 2.71
Distribution System Efficiency
Site Electric
Furnace, COP=1
Single Speed Heat
Pump, HSPF=7.9
Variable Speed Heat
Pump
91001 0.72 0.67 0.6691002 0.78 0.70 0.7291004 1.00 1.00 1.0091005 0.69 0.60 0.5991009 0.64 0.52 0.5291010 0.73 0.66 0.64
Average 0.76 0.69 0.69
Overall Efficiency
Site Electric
Furnace, COP=1
Single Speed Heat
Pump, HSPF=7.9
Variable Speed Heat
Pump
91001 0.72 1.33 1.6191002 0.78 1.63 2.0491004 1.00 2.12 2.5991005 0.69 1.23 1.4991009 0.64 1.28 1.5891010 0.73 1.45 1.77
Average 0.76 1.51 1.85
25
Lab (EPRI) vs Field Results
• Agreement on capacity and COP measurements generally good; more divergence for some cases
• Lab tests done in EFFICIENCY mode• All sites in Oregon set up in COMFORT mode• Lab showed higher COPs as expected because of
EFFICIENCY vs COMFORT mode• Possibility of variability for individual pieces of
equipment, as well
26
Lab results Field results
Lab vs Field Comparison(2 ton system)
27
Overall Findings
• VCHP performs 25-30% better than a single speed 7.9 HSPF heat pump on an annual basis
• Duct losses increased ~5% over a single speed heat pump
• Overall system offers improved performance over single speed base cases
• Heat pump sizing still matters• Auxiliary heat lockout control still matters
28
Next Possible Steps
• Cost analysis. Collect data on installed costs and compare to annual estimates of savings.
• More complete set of SEEM runs- additional comparisons to base cases (HP and/or EFAF)
• Possible UES process (depends on cost analysis)
• Further discussion of how system could be commissioned (PTCS)