understanding dehumidification performance at part load in...
TRANSCRIPT
ASHRAE Region V CRC August 12, 2005
Hugh I. Henderson, Jr., P.E.CDH Energy Corp.Cazenovia, NY
Understanding Dehumidification Performance at Part Load in
Commercial Applications
DOE/NETL Project #DE-FC26-01NT41253
Overview
o DOE Research Project Results Ø field tested various commercial and residential
cooling systemsØ detailed laboratory tests of DX and CHW coilsØ Findings: latent capacity degrades at part load
conditions
o Implications for equipment manufacturers & design engineers?Ø Calcs at sensible or latent design conditions are
not enoughØ Staging is key
Steady-State Cooling Coil Performance
o Cooling coil performance is well understood at steady-state conditions Ø Coils provide both sensible cooling and
moisture removal (latent cooling) Ø SHR: sensible heat ratio…..fraction of total
cooling that is sensible Ø Colder coil surfaces remove more moistureØ Reduced air flow provides more moisture
removal (for DX coils)Ø Higher condenser/outdoor temperatures
increase SHR
40 45 50 55 60 65 70 75 80 85Dry Bulb Temperature (F)
0.000
0.005
0.010
0.015
0.020
Abs
olut
e H
umid
ity R
atio
(lb
/lb)
SHR: 0.80
Sat
. Tem
p =
42.
4Total: 2.0 tons (coil)Power: 2.5 kW EER: 9.6 Btu/Wh
Typical AC Coil Performance
-
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
800 900 1000 1100 1200 1300 1400 1500 1600
Airflow (cfm)
SH
R (
-)
80/7280/6780/62
Catalog Data for 3-ton RTU
o SHR is decreases with… Ø Lower air flow cfm/tonØ Higher WB (actually RH)
Nominal Rating Point: SHR ~ 0.71
DX Part Load Performance
o AC cycles compressor ON and OFF based on a space thermostat
o The portion of time the coil operates (i.e., the runtime fraction) is longer when cooling loads are greater
RTF = ON(ON + OFF)
o How do sensible and latent capacity vary under cyclic conditions?
Sensible and Latent Capacity With Continuous Supply Air Fan Operation
COIL1_TEST_4B_10B_16B_22B 08/30/02 07:42:04 Cycle #1 (Comp ON time: 45.0 minutes)
0 20 40 60 80 100time (minutes)
-20
-10
0
10
20
30
Cap
acity
(M
Btu
/h)
Compressor
Sensible
Latent Addition
Latent Removal
Off-Cycle Evaporation is Adiabatic Process:
Sensible ≈ Latent Transition pt
Sensible Heat Ratio (SHR) vs Efficiency Debate
q Some in the industry assert that high efficiency AC have lowered SHRsØ ARI studies of industry test data have shown steady-state SHRs
remain near 0.75 …. unchanged from the past
q Part-load latent degradation may explain some of this disconnectØ newer equipment may use larger coils and lower suction
temperatures to achieve the same SHRØ steady-state performance is unchanged….but part-load
characteristics may be affected
Lab & Field Test Results
&
Eng. Model Development
Latent Degradation Concepts
0 20 40 60TIME (minutes)
La
tent
Cap
acity
Delay Time (to)time for first condensate to fall from pan
QL
Qe
Useful Moisture RemovalMo
Evaporation
twet = Mo/QL
γ = Qe/QL
Mo ≈ Evaporation
LHR Degradation Model
o Borrowed approaches used to develop part-load efficiency degradation function for SEER test procedure
o Used same part-load assumptions (Cd):Ø AC at startup described by a time constantØ Cycling rate driven by thermostat curve
o Additional latent assumptions:Ø Coil surfaces hold a fixed amount of moisture (Mo)Ø Off-cycle coil is like an evaporative cooler
0.0 0.2 0.4 0.6 0.8 1.0Runtime Fraction (-)
0.70
0.80
0.90
1.00
1.10
SH
R
Latent Degradation Model with:
twet = 720 secγ = 1.07Nmax = 3.6 cyc/hτ = 75 seclinear decay
Measured Data:
• averaged into 4 hour intervals
• entering RH: 57% to 63%
• entering air temp: 68F to 72F
• loop temps: 65F to 90FSteady State SHR = 0.76
Continuous supply air fan operation
Comparing LHR Model to Field Tests
o Henderson (1998) compared the model results to field data collected on a 3-ton residential geothermal heat pump
0
20
40
60
80
100
120
140
160
180
2 3 4
Number of Rows
Occ
uren
ces
Packaged
Air Handlers
Evap Coils
Equipment Survey
o Reviewed specifications for 500 commercial and residential AC units
o Goal was to determine Ø range of common coil
geometriesØ Variation by equipment type
o Typical DX AC coil is 3-rows, 15 fpi
0
20
40
60
80
100
120
140
160
180
200
10 11 12 13 13.5 14 14.5 15 16 17 18
Fin Spacing (fpi)
Occ
uren
ces
PackagedAir Handlers
Evap Coils
Field Testing
q Six field test sites were recruited:
Residential (5 units at 4 sites)
Ø 5-ton single-speed residential DX system in Connecticut
Ø Two single-speed DX units (2.5-ton and 3-ton) at a Virginia residence
Ø 3.5-ton residential DX system in Florida (single-speed condensing unit with variable speed air handler)
Ø 3-ton residential DX system in Florida (two-speed condensing unit with variable speed air handler)
Commercial (3 units at 2 sites)
Ø 10-ton commercial rooftop DX unit in Boston (2 stage)
Ø Commercial constant-air-volume chilled water coil in Florida (3 ton)
Ø Commercial variable-air-volume chilled water coil in Florida (7 ton)
COIL2_TEST_4B_10B_16B_22B_25B 11/14/02 10:07:57 Cycle #2 (Comp ON time: 45.0 minutes)
0 20 40 60 80 100
time (minutes)
0
5
10
15
20
25
Cap
acity
(MB
tu/h
)
16.3 min
QC= 6.7
QS= 21.8
QL= 7.4
Integration Pt
11.9 (gamma = 1.60)
Integrated Moist (delay of 1.0 min)
Mass twet
Sens (lb & min): 1.98 17.0
Lat (lb & min): 1.97 16.9
79.9 F, 60.4 F dp, 51.5 % Run 4 1.5 hz, 76.6 psi, 967 cfm, 30.04 in Hg
Actual Laboratory Results
Condensate
Sensible
(+) Latent (psych)
(-) Latent (psych)
Constant Fan Cyclic Tests for COIL2 (80db, 60dp)
0.0 0.2 0.4 0.6 0.8 1.0Runtime Fraction (-)
0.8
1.0
Sen
sibl
e H
eat R
atio
(-)
Steady State SHR = 0.755 (based on condensate)
twet= 17.3, gamma= 1.50, exp
1st Cycle2nd Cycle3rd Cycle
Laboratory Testing – SHR Degradation
Constant Fan Mode
LHR Model matches measured data at nominal entering conditions
(80F db, 60.4F dp)
Constant Fan Cyclic Tests for COIL2 (75db, 64dp)
0.0 0.2 0.4 0.6 0.8 1.0Runtime Fraction (-)
0.50
0.60
0.70
0.80
0.90
1.00
Sen
sibl
e H
eat R
atio
(-)
Steady State SHR = 0.533 (based on condensate)
twet= 11.2, gamma= 0.49, exp
1st Cycle2nd Cycle3rd Cycle
Laboratory Testing – SHR Degradation (cont.)
Constant Fan Mode
Model also matched measured data at other entering conditions!
(75F db, 64F dp)
Auto Fan Cyclic Tests for COIL5
0.0 0.2 0.4 0.6 0.8 1.0Runtime Fraction (-)
0.6
0.8
1.0
Sen
sibl
e H
eat R
atio
(-)
Steady State SHR = 0.680 (based on condensate)
1st Cycle2nd Cycle3rd Cycle4th Cycle5th Cycle
Laboratory Testing – SHR Degradation (cont.)
AUTO Fan Mode
Coil 5 showed impact in AUTO Fan Mode!
Field Testing – SHR Degradation
Significant AUTO fan degradation!
Sweetser - Downstairs AC Unit
0.0 0.2 0.4 0.6 0.8 1.0Runtime Fraction
0.8
1.0S
ensi
ble
Hea
t Rat
io (-
)
Steady State SHR = 0.855
Auto Fan ModeConstant Fan Mode (08/18/02 to 09/04/02)
(Model: twet=15 min, gamma=2.2, Nmax=3)
Measured Performance Parameters
Size
Fin Surface
Area
Retained Moisture Mass
Cond Delay Time
twet
(tons) (ft2) (lb) (lb/kft2) (min) (min) Coil 1 (Slanted slab, 3 row, 13 fpi, plain fins, orifice)
2.9 243.8 2.1 8.6 13.5 16.5
Coil 2 – Normal Flow (A-coil, 3 rows, 15.5 fpi, lanced sine-wave fins, TXV)
2.4 237.8 2.0 8.4 16.3 17.0
Coil 2 – Low Flow (A-coil, 3 rows, 15.5 fpi, lanced sine-wave fins, TXV)
1.4 237.8 2.0 8.4 32.5 29.0
Coil 4 (vert. slab, 2 rows, 14 fpi, wavy fins, orifice)
1.8 137.4 1.9 13.8 23.5 18.5
Coil 5 (slanted slab, 4 rows, 12 fpi, wavy fins, orifice)
2.3 162.7 1.4 8.6 11.5 9.5
Coil 6 (A-coil, 3 rows, 13 fpi, wavy fins,TXV)
1.7 231.1 2.7 11.7 34.0 33.
Coil 6 – High Flow (A-coil, 3 rows, 13 fpi, wavy fins,TXV)
2.0 231.1 2.7 11.7 27.0 27.
Coil 8 – Chilled Water (vert. slab, 4 rows, 10 fpi, wavy fins,46°F CHW)
1.5 135.0 1.4 10.4 26.5 26.5
Notes: 1-
2- 3-
Cooling capacity includes sensible and latent cooling at nominal conditions. Nominal conditions correspond to ASHRAE Test A test point. Surface area is gross fin area (coil face area x coil depth x fin spacing x 2). Condensate delay time and twet are at nominal conditions.
- 2 4 6 8 10 12 14 16 18 20
Korte & Jacobi (1997)
Lab - Coil 1
Lab - Coil 2
Lab - Coil 3
Lab - Coil 4
Lab - Coil 5
Lab - Coil 6
Lab - Coil 7
Lab - Coil 8
Residential 3.5 ton
Rooftop 10 tons
CHW Coil 2.9 ton
Henderson (1998)
Coil Moisture Retention (lb per 1000 ft2)
Lab + Field: Retained Moisture
Estimated in Field
Lab Measurements
Moisture per total gross fin area (gross area = face area x depth x fin spacing x 2)
Heink - Humidity Inside vs. Outside
0 50 100 150
Outdoor Humidity (gr/lbm)
0
20
40
60
80
100
Indo
or H
umid
ity (
gr/lb
m)
Auto Fan ModeConstant Fan Mode
Space Humidity – Impact of Constant Fan
o Large penalty for constant fan operation
o Even for a dual-capacity AC unit
Impact on Space Humidity Levels
Room Humidity - July 13
50
55
60
65
70
75
80
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24Hour of Day
Spa
ce H
umid
ty (%
RH
)
TRNSYS Model of Hotel Room in Hawaii
Constant Fan
AUTO Fan
Chilled Water Coils
Chilled Water Coil Performance
CHW Performance at 400 cfm/ton
0 2 4 6 8 10Chilled Water Flow (gpm)
0.50
0.60
0.70
0.80
0.90
1.00
Sen
sibl
e H
eat R
atio
(-) 80F / 60Fdp
80F / 68Fdp
75F / 56Fdp
75F / 64Fdp
Nominal Operating Point
Chilled water coils behave similar to direct expansion (DX) coils …..even though they are controlled differently
• gpm varies to provide capacity
Latent Degradation - CW CoilsConditions: IN= 75F, 0.010lb/lb, TCH=44F, fpi= 8, rows= 4
0.2 0.4 0.6 0.8 1.0Total Capacity (tons)
0.6
0.8
1.0
Sen
sibl
e H
eat R
atio
(-)
LHR curve for DX EquipmentAssuming: 960s, 1.5, 45s, 3 cyc/h, linear, SHR_steady-state=0.75 (at 400 cfm/t)
CW Coil (ASHRAE CCDET Model)
What Can Equipment Manufacturers
& Design Engineers Do?
What Control Strategies Help?
Do fan delay strategies help?q No…not until about a 10 minute delay
Two-speed or staged capacity systems?q Yes…increases runtime fraction (depends on coil split &
fan control)
Reduced airflow during off cycle?q Yes…would minimize off cycle evaporation
Cycle the fan with the compressor?q Yes…whenever possible
Impact of Fan Delay
•Fan was turned off for 2-, 5-, and 10-minutes after the cooling cycling to “allow moisture to drain from the coil”.
• No significant “drainage” occurs
Impact of Fan Delay
0.0 0.2 0.4 0.6 0.8 1.0Runtime Fraction (-)
0.80
0.85
0.90
0.95
1.00
Sen
sibl
e H
eat R
atio
(-)
Steady State SHR = 0.823 (based on condensate)
10-min delay
5-min delay
2-min delay
No fan delay
Coil on/off times (min): 7/17.5 10/10 20/6.7 30/6
Impact of Proper AC SizingConstant Fan Cyclic Tests for COIL2 (80db, 60dp)
0.0 0.2 0.4 0.6 0.8 1.0Runtime Fraction (-)
0.8
1.0
Sen
sibl
e H
eat R
atio
(-)
Steady State SHR = 0.755 (based on condensate)
twet= 17.3, gamma= 1.50, exp
1st Cycle2nd Cycle3rd Cycle
“Normal” Sizing
50% Oversized
Normally Sized
Oversized
Impact of Equipment Staging
q Specify 2-Stage RTUsq Most RTUs spend majority of operating time at less
than 50% runtime…with no latent capacityq Continuous operation in first stage provides more
latent removal at part load..assuming face-spilt coils
q Operate in continuous fan only when requiredq Consider cycling fan with compressor when ventilation
is not requiredq Consider using lower air flow rate when cooling is off
Summary
q It takes 10 to 30 minutes after startup for moisture removal to begin
q A typical coil holds ½ to 1 lb per tonq With constant fan, latent removal disappears
below half loadq Chilled water coils have similar part load
performanceq To maintain latent capacity
q Cycle fan with the compressor (when possible)q Slow the fan down during off cycleq Add more capacity stages