DOE 2.2 - Work-Arounds to Hidden Problems:
eQUEST - A half hour to learn;
three years to master.
Building Energy Simulation Forum (BESF)
April 20, 2011
Portland, OR
Reid Hart, PE
Associate Director, Technical Research Group
Announcing: Final draft available
covers:
DCV Modeling of
–
–
–
–
–
Movie Theaters
Meeting/Lecture
Classrooms
Gym/Fitness
Retail
Includes prototype
eQUEST models
Demand Controlled
Ventilation (DCV)
Measure Analysis Guide
Research Funded by:
Bonneville Power Administration
Final Draft available at:
http://energytrust.org/Business/building-energy-simulation/
Final soon available at: www.peci.org &
www.bpa.gov/energy/n/commercial.cfm
DOE 2.2 – Work-Arounds to Hidden Problems:
eQUEST - A half hour to learn;
three years to master.
Learning Objectives:
• Understand correct modeling of outside air
economizers for single-stage dX cooling
• Distinguish commercial ventilation requirements and
modeling: What to do if you aren’t in California!
• Understand DOE 2 limits and proper modeling of
unoccupied fan operation
• List issues for proper VFD fan modeling
• Know how to model DCV
4
Understand Correct Modeling
of Outside Air Economizers
for Single-Stage dX RTUs
DOE 2.2 – Work-Arounds to Hidden Problems:
eQUEST - A half hour to learn;
three years to master.
5
Basic OSA Economizer Idea
6
Savings from Economizer Levels
Outside Air Economizer Potential Savings
200.0
Standard
Economizer
180.0
160.0
kWh / Ton
140.0
Premium
Economizer
120.0
100.0
Ideal
Economizer
80.0
60.0
40.0
Required
Cooling
20.0
0.0
37
42
47
52
57
62
67
72
77
82
87
92
97
Outside Air Temp - Eugene OR
Ideal economizer does not happen in RTUs.
Get most of savings with alternating integration.
7
102
Integrated Economizer for RTUs
• Thermostat with dedicated stage for
economizer.
– Single stage at thermostat requires
“either/or” operation with low (55 F)
changeover setting. Result: very little
savings.
– Dedicated first stage on thermostat
allows economizer to operate before
compressor.
• For dX cooling with minimal stages; must locate primary
control sensor in discharge air, not mixed air position.
– Mixed air sensors notoriously inaccurate (poor mixing)
– Otherwise DA too cold, economizer will be disabled
8
Three specific economizer
modeling issues:
• Maximum outside air is usually less than the
100% default included in eQUEST
• The minimum outside air position is often
greater than required
• Single zone packaged units with single-stage
dX cooling coils operate in alternating
integration while DOE 2.2 PSZ system
incorrectly models full integration
9
Max OA Fraction
• Code may call for 100%; do you get it?
Packaged Unit Available OSA Flow
Ecotope Study of 16 Packaged Units
6
Frequency
5
4
3
2
1
0
<20%
Average: 64%
Median: 66%
30%
50%
60%
70%
80%
90%
100%
OSA Airflow; Damper Full Open
• 70% is more reasonable for RTUs without
exhaust fans
10
Suggested Max_OA_Fraction
OA fraction improved with:
• Barometric relief damper
• Motorized exhaust damper
• Seals and good closure on return damper
• Powered exhaust fan
• Return fan with return damper seals
11
Configuration
MAX_OA_FRACTION
No relief damper
0.50
Barometric relief damper (most common)
0.55 to 0.75 (0.70 suggested)
Motorized relief/exhaust damper
0.60 to 0.80 (0.75 suggested)
Good seals on return damper
0.75 to 0.85 (0.80 suggested)
Powered exhaust fan activated at 50% OSA
0.80 to 0.90 (0.85 suggested)
Return Fan with good seals on return damper
0.80 to 0.90 (0.85 suggested)
Minimum Ventilation Rate
• MIN-OUTSIDE-AIR at system level
• ZONE keywords take precedence
OUTSIDE-AIR-FLOW, OA-CHANGES,
OA-FLOW/PER or EXHAUST-FLOW
• Office typically 7% to 13% required; Actual ~20%
5
4
3
12
e
0
M
or
80
60
40
30
20
15
OSA percent (bin up to)
10
n= 16
mean = 20.6%
10
5
2
1
0
0
Frequency
RTU Percentage OSA
Ecotope/EWEB 2001
Baseline?
• Use 20%? May be
appropriate for low
density buildings
• Prototypes use ASHRAE
90.1 area + 50% default
people base case;
matching most codes
Alternating Integration Fix
–
–
–
–
–
13
Economizer only
Econo + Stage 1 dX
Econo + Stage 2 dX
Min OA + Stage 1 dX
Min OA + Stage 2 dX
Alternating Integration
OSA=65; RA=75
250%
deg F
75
200%
65
150%
55
100%
45
50%
35
0%
25
Cool Stage
DA
MA
OSA%
1
342
5
76
8
1109
1112
1134
1156
1187
2290
2212
2234
2256
2278
3390
3312
3334
3356
3378
4490
4412
4434
4465
4478
5590
5521
5534
5556
5578
6690
6612
6634
6656
6687
9
• Rtus (PSZ) with staged
dX do NOT have full
integration
• DOE 2 says they do
• Economizer must
throttle back when dX is
on . . . Why?
• Need to model all
modes during each hour
OSA High Limit or
“changeover” sequence
When is it too hot to use economizer?
• Mode of Control
– Single (OSA only) fixed (snap disc)
– Single (OSA only) adjustable (analog)
– Differential or Comparative (OSA vs. RA)
• Type of Sensor
– Dry-bulb measures temperature only
– Enthalpy adjusts for humidity
– Separate dry-bulb and humidity sensors
14
Dry-Bulb Differential Changeover
• Dry-bulb analog
OA & RA
sensors
• Differential
changeover
logic
– Both OA &
RA sensors
pprovided
– NO snap
discs
• Single
changeover OK
with aggressive
setpoint
15
Differential
Changeover
• Differential or Comparative Changeover
– RA & OSA sensor
– Avoids confusion; no guesswork
– Adapts to loads; avoids callbacks
• How it works
– Compares temperature or energy content (enthalpy) of
both RA & OSA
– Allows economizer if OSA is cooler or has less
enthalpy or heat energy depending on sensor
– Maximize integration or time both the economizer can
function with cooling
16
Dry-bulb Changeover in the West
• ARI map shows where
evaporative cooling
works.
• Economizer climates in
West (ASHRAE 90.1):
– 95% dry
– 5% intermediate
– 0% humid
• Changeover Sensor:
– West: dry-bulb
– East: enthalpy
17
Enthalpy
Sensors
See Iowa Humidity Sensor Testing: NBCIP PTR: Duct Mounted Relative Humidity
Transmitters. April 2004; Rev. November 2004
www.energy.iastate.edu/Efficiency/Commercial/download_nbcip/PTR_Humidity_Rev.pdf
18
Modeling dX Cooling
Economizer in eQUEST
• DOE 2.2 will overstate economizer energy savings for
Packaged Single Zone (PSZ) and PVVT systems using
single stage DX cooling coils.
• DOE 2.2 models a fully integrated economizer strategy
instead of actual alternating integration.
• In actuality, a single-stage dX cooling unit must throttle
back the outside air during integrated operation.
• In order to simulate an alternating economizer strategy
in DOE 2.2 a work around has been developed.
– Adjust the high limit to reduce economizer savings
– Described in Appendix A of DCV modeling guide
19
High Limit Adjustment –
Low Density Occupancies
Adjusting DOE 2.2 PSZ from full integration to alternating integration
Low Density Occupancies such as offices
OAT
Balance:
OAT High Limit
Adjusted High Limit Input, °F
Reduction in High Limit, °F
57
52
47
57
52
47
Light
Med
Heavy
Light
Med
Heavy
75.0
71.3
69.6
68.6
3.7
5.4
6.4
72.5
70.2
69.2
68.2
2.3
3.3
4.3
70.0
69.1
68.2
67.2
0.9
1.8
2.8
67.5
66.5
65.7
64.8
1.0
1.8
2.7
65.0
64.5
64.2
63.7
0.5
0.8
1.3
62.5
62.1
61.7
61.2
0.4
0.8
1.3
60.0
59.8
59.6
59.3
0.2
0.4
0.7
57.5
57.5
57.1
56.7
0.0
0.4
0.8
55.0
55.0
54.8
54.5
0.0
0.2
0.5
Internal loads are characterized as light, medium and heavy.
20
Heavy:
Lighting at 2.3 Watts/square foot with high occupancy; Call center
Medium:
Lighting at 1.7 Watts/square foot; moderate occupancy; open office
Light:
Lighting at 0.7 Watts/square foot with low density occupancy
High Limit Adjustment –
High Density Occupancies
Adjusting DOE 2.2 PSZ from full integration to alternating integration
High Density Occupancies (with increased ventilation)
OAT
Adjusted High Limit Input
Reduction in High Limit
52
47
37
52
47
37
Hi Limit
Light
Med
Heavy
Light
Med
Heavy
75.0
70.7
69.5
67.8
4.3
5.5
7.2
72.5
69.8
69.1
67.9
2.7
3.4
4.6
70.0
69.1
68.4
66.4
0.9
1.6
3.6
67.5
66.7
66.2
64.7
0.8
1.3
2.8
65.0
64.6
64.4
63.6
0.4
0.6
1.4
62.5
62.0
61.7
60.8
0.5
0.8
1.7
60.0
59.7
59.5
58.9
0.3
0.5
1.1
57.5
57.3
57.1
56.4
0.2
0.4
1.1
55.0
54.9
54.7
54.1
0.1
0.3
0.9
Balance:
Internal loads are characterized as light, medium and heavy.
21
Heavy:
Retail with high lighting or appliance and people density
Medium:
Moderately full classrooms, meeting rooms, and lecture halls
Light:
Theatre or assembly with intermittent occupancy, low light levels
Development of Economizer
High Limit Adjustment Values
• Use a simplified bin method to find:
– Bin cooling loads and occupied hours
– OSA % for 53°F DAT with mechanical cooling
– Time of economizer only operation in each bin
• Percentage of full integrated economizer
delivered by alternating integration
• Match those percentage reductions in savings
to parametric DOE2 runs.
22
Integration Increases Savings
+Internal gains
+ Solar
- Heat loss
- Ventilation
• Balance Point:
cooling not needed
• Internal gains are
reducing
– Efficient lights
– LCD monitors
Snap Disc
2.50
Cooling Savings kWh/sf
• Cooling load:
Cooling Savings vs. Changeover Setpoint
at Various Building Balance Points (BP)
9 SEER dX, 70% Max OSA
Differential &
Aggressive
Analog
Integrated
2.00
1.50
1.00
0.50
0.00
50
55
60
65
70
75
80
Changeover Temp OSA
High
BP=25
Core
BP=32
Strip
BP=40
Mod
BP=45
Med
BP=55
Low
BP=70
• Save more with higher OSA changeover
– High changeover requires integration
– Little savings at snap disc changeover points
23
Impact of Changeover Temperature
on % dX PSZ Economizer Savings
Impact of Changeover Temperature
on dX PSZ Economizer Savings
For single stage dX cooling systems, reduce simulated
high limit (changeover) by about 2F to 6F (depending on
internal loading) below actual in DOE 2.2.
Share of Ideal Economizer Savings
100.0%
Internal
Gain Light
W/sf for
70% area
90.0%
80.0%
70.0%
60.0%
L: 1.0
50.0%
L: 2.0
40.0%
L: 3.0
30.0%
20.0%
10.0%
0.0%
45.0
50.0
52.5
55.0 57.5
60.0
62.5
65.0 67.5
70.0
Drybulb Economizer High Limit (changeover)
24
72.5
75.0
Portland Cooling Loads in
Economizer Range
60%
12%
50%
10%
40%
8%
30%
6%
20%
4%
2%
10%
0%
0%
OAT Bin
25
Percentage of Peak
Cooling Capacity
14%
47
.5
50
.0
52
.5
55
.0
57
.5
60
.0
62
.5
65
.0
67
.5
70
.0
72
.5
75
.0
Percent of Occupied
Economizer Bin Hours
Economizer Bin Cooling Loads
Portland, Oregon
PDX Econo
Bin Hours
Light Mech
Cool Load
Med Mech
Cool Load
Heavy
Mech Cool
Load
Light / Med / Heavy refer to internal cooling
load with Balance point of 57F / 52F / 47F
Economizer Time and Share of
Ideal by OAT Bin
Economizer Alternating Integration
Bin Share of Ideal Economizer
100%
Time Econo Only or Share of Ideal
Light Time Econo Only
Med Time Econo Only
Heavy Time Econo Only
90%
Light Share of Ideal
Med Share of Ideal
80%
Heavy Share of Ideal
70%
For a single stage dX cooling
unit, the maximum comfortable
(53 DAT) economizer % open
is shown as green dotted line.
Percent of time for economizer
only (no mech cooling) is
shown with the per bin share of
an ideal economizer cooling
that an alternating economizer
would provide.
60%
50%
40%
30%
20%
45.0
50.0
55.0
60.0
OAT, deg F
26
Econo Max w/ Mech
65.0
70.0
75.0
Light / Med / Heavy refer to internal cooling
load with Balance point of 57F / 52F / 47F
Economizer Time–Weighted
Effective dT by OAT Bin
Economizer Alternating Integration
Time-Weighted Econo Effective Sensible dT by individual OAT Bin
Effective Sensible Temperature
Difference, MAT - DAT, deg F
20.0
Fully Integrated Econo dT
18.0
Econo dT w/ 1 stage cool
16.0
Light Alternate Econo dT
Med Alternate Econo dT
14.0
Heavy Alternate Econo dT
12.0
For a single stage dX cooling
unit, the maximum comfortable
(53 DAT) economizer dT is
shown (green dotted) vs fully
integrated economizer (orange).
Time weighted dT are shown for
alternating integration.
Differential adjusted for 70%
maximum outside air.
10.0
8.0
6.0
4.0
2.0
0.0
45.0
50.0
55.0
60.0
OAT, deg F
27
65.0
70.0
75.0
Light / Med / Heavy refer to internal cooling
load with Balance point of 57F / 52F / 47F
Economizer Delivered Cooling
Sensible Effect by OAT Bin
Alternating Integration Delivered Cooling Effect of Ideal Economizer
Cooling and Hour Weighted Effect of All Bins
100.0%
Percent Ideal Economizer Effect
98.0%
Light Cooling
Weighted Share
96.0%
94.0%
92.0%
Med Cooling
Weighted Share
90.0%
88.0%
86.0%
Heavy Cooling
Weighted Share
84.0%
82.0%
80.0%
47.5 50.0 52.5 55.0 57.5 60.0 62.5 65.0 67.5 70.0 72.5 75.0
High Limit Changeover Setting
28
Light / Med / Heavy refer to internal cooling
load with Balance point of 57F / 52F / 47F
Sacramento Cooling Loads in
Economizer Range
29
Economizer Bin Cooling Loads
Sacramento, California
12%
50%
45%
40%
35%
8%
30%
6%
25%
20%
4%
15%
10%
2%
5%
0%
0%
SMF Econo
Bin Hours
Percentage of Peak
Cooling Capacity
10%
Light Mech
Cool Load
Med Mech
Cool Load
Heavy Mech
Cool Load
47
.5
50
.0
52
.5
55
.0
57
.5
60
.0
62
.5
65
.0
67
.5
70
.0
72
.5
75
.0
Percent of Occupied
Economizer Bin Hours
• Climate does
NOT matter for
adjusted high
limit inputs
• eQUEST
provides climate
adjustment
• Portland and
Sacramento
Adjustments all
within 0.75°F
OAT
Light / Med / Heavy refer to internal cooling
load with Balance point of 57F / 52F / 47F
High Limit Adjustment –
Low Density Occupancies
Adjusting DOE 2.2 PSZ from full integration to alternating integration
Low Density Occupancies such as offices
OAT
Balance:
OAT High Limit
Adjusted High Limit Input, °F
Reduction in High Limit, °F
57
52
47
57
52
47
Light
Med
Heavy
Light
Med
Heavy
75.0
71.3
69.6
68.6
3.7
5.4
6.4
72.5
70.2
69.2
68.2
2.3
3.3
4.3
70.0
69.1
68.2
67.2
0.9
1.8
2.8
67.5
66.5
65.7
64.8
1.0
1.8
2.7
65.0
64.5
64.2
63.7
0.5
0.8
1.3
62.5
62.1
61.7
61.2
0.4
0.8
1.3
60.0
59.8
59.6
59.3
0.2
0.4
0.7
57.5
57.5
57.1
56.7
0.0
0.4
0.8
55.0
55.0
54.8
54.5
0.0
0.2
0.5
Internal loads are characterized as light, medium and heavy.
30
Heavy:
Lighting at 2.3 Watts/square foot with high occupancy; Call center
Medium:
Lighting at 1.7 Watts/square foot; moderate occupancy; open office
Light:
Lighting at 0.7 Watts/square foot with low density occupancy
Distinguish Commercial Ventilation
Requirements and Modeling:
What to do if you aren’t in California!
DOE 2.2 – Work-Arounds to Hidden Problems:
eQUEST - A half hour to learn;
three years to master.
31
Ventilation Rate History
32
Source: Fred Kohloss; ASHRAE Summer Meeting 2003
Health or Stinky Air?
• Primary purpose is “acceptable indoor air quality”
– Primary unit in testing is “olf” (olfactory unit)
• Standard has a purpose to “minimize negative impacts
on health”
– Difficulty in proving objective healthy concentrations for
contaminants of concern in commercial
– Second hand smoke major restriction
– Most commercial occupancies do not have significant health
contaminant issues, with exception of dry cleaning, nail salons,
light manufacturing and perhaps retail and warehousing
33
Standard 62.1 Procedures
• Ventilation Rate Procedure (Prescriptive Path)
– For given occupancy, provide minimum ventilation when
occupied
• IAQ Procedure
– Provide a ventilation and air cleaning strategy that:
• Addresses contaminants of concern
• Is found acceptable based on survey or measurement
– Now requires site specific investigation and monitoring
• Natural Ventilation Procedure
– Limited to almost always require mixed mode
34
ASHRAE Standard 62.1
Ventilation Rate Procedure
• Determine an outside air ventilation rate and provide
when occupied
• Ventilation now has 2 components
– Area ventilation rate
– People ventilation rate
• These 2 components are used to establish the
total or “full” ventilation rate
– Full ventilation rate = Area + People
– There is NOT a requirement to provide the area rate to
unoccupied areas!
35
People + Area Based Rate
• Occupancy from
– Default tables (ASHRAE 62.1 or code); usually high
– Observation or architectural program
– Owner information
• Most codes set a floor for people basis to 50% of stated
table default people densities
• Less important to get people accurate if using DCV
• Example; Conference Room, 1000 square feet:
– Area: 1000 x 0.06 cfm/sf = 60 cfm
– People: 50 x 5 cfm/per = 250 cfm
– Total = 310 cfm or 26% of 1200 unit or zone cfm
36
Ventilation Codes
• Most energy codes require DCV in high density spaces
• Most PNW mechanical codes either
– Allow ASHRAE 62.1 as an alternate
– Or are modeled after 62.1
• Occupancy type lists are generally shorter than
ASHRAE; tied to fire occupancies
• California is different
– Ventilation is maximum of area or people rate
• Generally 0.15 cfm/sf area rate
• Generally 15 cfm/person people
– Requires pre-purge and continuous fan operation
– DCV has general 1000 ppm setpoint
– CO2 sensor in space with readout required
37
California is Different (of course)
Standard: ASHRAE 62.1
38
California Title 24
Requirement
(since 2001n)
(2008/10 Compliance Manual)
People; Area rates
Sum
Max (DOE 2.2 default)
High density cfm per person
down to 5
stay at 15
General area: cfm/sf
0.06
0.15
Ventilate when
“actually occupied”
“Usually occupied”
Intermittent fan
Short periods (~30 min)
Only 5 minutes / hour
DCV CR Setpoint
Match space (800-1800)
Always 1000 ppm
DCV Sensors
CO2 return air OK;
schedule, counters
CO2 only, in space, readout,
trended if DDC
Space purge
Not required (recommended 1 hour or 3 air changes
if non human contaminants) before occupancy
Multiple zone
Ventilation efficiency
Min for space; Vent at fan
Space & Zone OA Specs
ASHRAE 62.1 vs CA T24
Area Rate:
INF-FLOW/AREA
Unoccupied Cfm/sq ft; not
wind corrected
Enter unoccupied; use
INF-SCHEDULE to multiply
for occupied
0.0263 + 0.12 = 0.1463
0.0263 x 5.56 = 0.12
People Rate:
OA-FLOW/PER
(Not: OA-FLOW/AREA)
39
Space:
Zone (62.1):
CA T24:
Infiltration works for area as
OSA is OSA for SZ units
Prototype Minimum Ventilation
• The minimum ventilation requirements used for
the prototype models follow ASHRAE 62.12010 Ventilation Rate Procedure.
• Ventilation = People Outdoor Air Rate +
Area Outdoor Air Rate
People Ventilation
Requirement
(CFM/Person)
Area Ventilation
Requirement
(CFM/SQFT)
10
0.12
Gym
3.75
0.18
Lecture Hall
7.5
0.06
Movie Theater
10
0.06
Movie Theater
Lobby
5
0.06
7.5
0.12
Space Type
Classroom
Retail
40
Understand DOE 2 Limits
and Proper Modeling
of Unoccupied Fan Operation
DOE 2.2 – Work-Arounds to Hidden Problems:
eQUEST - A half hour to learn;
three years to master.
41
eQUEST Defaults at Work
Night Cycle Fan Control Default:
• “Stay Off” is default
• eQUEST adds 2 hours/day to fan schedule
Options:
• “Cycle on Any” runs fan full hour on ANY load
• No option for
– “on” during occupied and
– intermittent” during unoccupied
42
• Intermittent with a 24 hour fan schedule is
intermittent all the time
Solution: Use the Fan Schedule
• Hourly outputs
analyzed for
unoccupied load
• Monthly equivalent
fan hours determined
• Schedules built up for
prototypes
Not perfect,
but closer than
“cycle on any”
43
Class
room
Retail
Gym
Movie
Theater
Lecture
Hall
January
1
3
4
3
2
February
1
2
3
2
2
March
1
2
2
2
1
April
1
2
2
1
1
May
0
1
1
1
0
June
0
0
1
0
0
July
0
0
1
0
0
August
0
0
2
0
0
September
0
0
1
0
0
October
0
1
2
1
0
November
1
2
2
2
1
December
2
2
3
2
1
Longer Run Hours with VFD
• Similar process
• Lower speed
results in longer
hours
44
Class
room
Retail
Gym
Movie
Theater
Lecture
Hall
January
5
All
All
All
All
February
2
All
All
All
All
March
4
All
All
All
6
April
3
8
All
All
5
May
1
4
6
3
1
June
0
1
4
1
0
July
1
0
5
0
0
August
2
0
8
0
0
September
0
0
5
1
0
October
1
5
7
5
2
November
4
7
All
All
5
December
All
All
All
All
5
List Issues for Proper
VFD Fan Modeling
DOE 2.2 – Work-Arounds to Hidden Problems:
eQUEST - A half hour to learn;
three years to master.
45
FAN EIR-fPLR
EIR-fPLR:
– Energy Input Ratio
– as a function of
– Part Load Ratio
•
Dimensionless
– Other formulas or
inputs define 100%
energy input
– Systems finds CFM
for the hour
100%
80%
Fan Energy Input
•
FAN-EIR-FPLR Curves
eQUEST defaults
60%
eQ_FC-DD
eQ_FC-IV
eQ_VAV-VSD
40%
20%
0%
0%
25%
50%
75%
100%
Fan Output
•
EIR-fPLR curve quickly translates hourly fan output (cfm) to energy input
– Does not require complexity of a ductwork model inside DOE2 or TRACE
– Allows any separately calculated relationship to be input
•
Default curves in the system for:
– Forward Curve or Air Foil; Discharge Dampers or Inlet Vanes;
– Any VSD; VaneAxial
46
Losses occur for each
conversion of energy
Wire Losses
2%
Efficiency
Meter
Wire
Motor
Belt
Fan
System
80%
70%
52%
Loss
2%
20%
6%
30%
48%
kW
9.65
9.46
7.57
7.14
5.00
4.65
Motor
Efficiency:
80%
= Losses of
20%
Fan Belt
Losses
6%
Air Flow
Overall Fan System Efficiency
Multiple Conversions = Multiple Losses
Fan
Efficiency:
70%
= Losses of
30%
Meter kW
Motor Input kW
Motor Shaft kW
Fan Shaft kW
Air kW = Work
Losses
Overall system efficiency = [5.0 kW work] / [9.65 kW in] = 52%
47
• Losses occur for each
conversion of energy
Wire Losses
2%
Motor
Efficiency:
80%
= Losses of
20%
Fan Belt
Losses
6%
Air Flow
Overall Fan System Efficiency
Multiple Conversions = Multiple
Losses
Fan
Efficiency:
70%
= Losses of
30%
VSD
Efficiency
Meter
Wire
Motor
Belt
Fan
System
48
80%
70%
52%
Loss
2%
20%
6%
30%
48%
kW
9.65
9.46
7.57
7.14
5.00
4.65
Meter kW
Motor Input kW
Motor Shaft kW
Fan Shaft kW
Air kW = Work
Losses
What is impact of VSD?
• The old rule of thumb is added constant losses of 5-10% of peak input
• Newer or larger units closer to 2%-3% of peak
VFD Efficiency – A Research ?
VFD losses must be included in EIR-fPLR curve
•
•
Two available references for VFD
efficiencies were used in FanSysCalc
They have different part load curves, as
shown at right
The DOE-OIT method is based on a
table published in three references:
–
–
–
•
49
USDOE OIT. Ask the clearinghouse: Variable
speed drive part-load efficiency. January 2002.
Energy Matters. Office of Industrial
Technologies, US Department of Energy.
http://www.ornl.gov/sci/buildings/20070205_
AuthorsManual_Jan2007.pdf
Chan, Tumin. August 2004. "Beyond The
Affinity Laws." Engineered Systems Magazine.
http://www.esmagazine.com/Archives?issue=65
496
Rooks, JA & Wallace, AK. 2003. Energy
efficiency of variable speed drive systems. Pulp
and Paper Industry Technical Conference,
2003.
The Bernier method uses an equation
published in
–
Bernier, MA & Bourret, B. December 1999.
Pumping energy and variable frequency drives.
ASHRAE Journal.
Compare VFD Efficiency Methods
100.0%
VFD Efficiency
•
80.0%
60.0%
DOE-OIT
40.0%
Bernier
20.0%
0.0%
0
50
100
150
Flow gpm
Further Reading:
Development of a Correlation for System
Efficiency of a Variable-Speed Pumping
System (ASHRAE 2008)
Efficiencies of an 11.2 kW Variable Speed Motor
and Drive (ASHRAE 2001)
Non-Dimensional Pumping Power Curves for
Water Loop Heat Pump Systems (ASHRAE
1999)
Utilizing VFD for Building HVAC System
Performance (ASHRAE 2008)
Variable Flow in Chilled Water Systems-Benefits
and Controls (ASHRAE 2006)
VSD:
50
Size Matters
Sellers, David. A Field Perspective on Engineering. Dec. 18, 2010.
http://av8rdas.wordpress.com/2010/12/18/variable-frequence-drive-system-efficiency/
VSD:
51
Vintage Matters
Sellers, David. A Field Perspective on Engineering. Dec. 18, 2010.
http://av8rdas.wordpress.com/2010/12/18/variable-frequence-drive-system-efficiency/
Simulating Static Pressure Reset
in eQUEST/DOE2; TRACE
• Temperature reset
FAN-EIR-FPLR Curves
CCC FanSysCalc vs. eQuest defaults
– very easy in wizard
100%
• Static Pressure reset
• The CCC
FanSysCalc tool
– is a good source to
generate a realistic
EIR-fPLR curve
52
80%
ccc_Base
Fan Input
– requires a new
– FAN-EIR-fPLR
curve
60%
ccc_LoSP
ccc_ResetSP
eQ_FC-DD
40%
eQ_VAV-VSD
20%
0%
0%
25%
50%
Fan Output
75%
100%
Fan Cycling and VSD Control
• Each DCV prototype was modeled with three different
fan control methods:
– constant volume,
– fan cycling, and
– variable speed drive.
• The fan cycling and variable speed drive fan control
methods respond to an increase or decrease in space
temperature.
• Most packaged rooftop units are controlled such that
the fan will operate during night time hours to meet
heating and cooling loads.
53
Know How to Model DCV
DOE 2.2 – Work-Arounds to Hidden Problems:
eQUEST - A half hour to learn;
three years to master.
54
Demand Controlled Ventilation (DCV)
• Idea here is to provide
– area ventilation rate when occupied,
– then proportion people ventilation rate to actual occupancy
• Measure occupancy by:
–
–
–
–
55
Carbon dioxide (CO2) concentration
Counters or security system
Schedule approximation
Camera technology
Demand Controlled Ventilation
– Use CO2 as a proxy for adequate ventilation with
CO2 sensors
• Odor: ~700 above background
• Target 1000 to 1200 ppm
• For 2-10 V CO2 sensor where 2 V is 800 ppm, we want
damper to start at 4 V or 1100 ppm
• Damper is full open at 10 V or 2000 ppm
– For smaller areas:
• Use occupancy sensors
• Use scheduled ventilation
56
DCV Savings
M illion Btu's Post ECM
Baseline
Gym is a great application
– Go from 38% average OSA
to 15%
– Save $2200 per year,
mostly heat
0.0
100.0
200.0
300.0
400.0
500.0
Heating
Cooling
Fans
Lights Ocp
Lt Unoc/Ext
Equipm't Ocp
Eq Unoc/Ext
M illion Btu's Post ECM
Baseline
0.0
Heating
Cooling
Fans
Lights Ocp
Lt Unoc/Ext
Equipm't Ocp
Eq Unoc/Ext
57
100.0
200.0
300.0
400.0
Office saves less, but may pay
– Only need 10-15% OSA; might
reduce by 5%
– Save $153 per year
– Save more if base case is over
ventilated (typically is 20%)
CO2 Sensors
• Why are they used?
– Energy Savings:
• Only when not in OSA economizer mode
– Eliminate over-ventilation when occupancy is low
– Coordinated with OSA economizer maximizes
benefit of ventilation control and savings
– Establish transient space occupancy
• Alternate methods
– Schedules, camera visualization, counters
– VOC sensors (if co-related to CO2)
58
CO2 as Proxy for Air Quality
• Carbon Dioxide is JUST a proxy
– Does not account for other chemicals
– Does reflect space occupancy
– Is not a “contaminant of concern”
• What do CO2 levels mean?
–
–
–
–
59
Background outdoors 350 to 450 ppm
Equivalent to 15-20 cfm/person: 1000 ppm
OSHA limit for exposure > 8 hrs: 5000 ppm
Submarines: 7000 normal, 30,000 dive!
CO2 Target Is NOT 700 ppm!
• DCV CO2 concentration setpoint
– Earlier standard called for 700 ppm CO2 over ambient
as target level
– Ambient at 400 means 1100 ppm setpoint
• Actually is based on specific occupancy
• Explained well in the 62.1-2007 users manual
– Steady state CO2 CR
• Concentration lags occupancy
– DCV allowed to control for steady state
– Pre-purge not required; may be good
60
Find CR for your space
• CR is target CO2
–
–
–
–
Think OA% “area” & “full”, not minimum
Met varies due to activity
Large ratio indicates more DCV savings potential
Where possible, use actual people & CFM, not defaults!
ASHRAE 62.1-2010
Area Type
Art Classroom
Office - default
Office - open
Class (age 9+)
Retail Sales
Grocery
Call Center
Lecture Class
Movie Theater (actual)
Conference
Restaurant
Assembly
Rock Concert (dance)
61
Rp
Pz
20
5
7
35
15
8
12
65
77
50
70
150
100
sf/p
50
200
143
29
67
125
83
15
13
20
14
7
10
met
1.2
1.2
1.2
1.2
1.5
1.7
1.2
1.1
1.0
1.1
1.4
1.0
2.0
Rs
cfm/p cfm/sf
10
5
5
10
7.5
7.5
5
7.5
5
5
5
5
5
0.18
0.06
0.06
0.12
0.12
0.06
0.06
0.06
0.06
0.06
0.18
0.06
0.06
CR
ppm
824
874
994
1001
1050
1162
1206
1278
1563
1592
1643
1644
2800
*For all types: area is 1000 square feet, COa = 400 ppm, Ez is 80%, unit cfm 1.2 cfm/sf
OA% OA%
area
full Ratio
19% 40% 2.1
6%
9% 1.4
6% 10% 1.6
13% 49% 3.9
13% 24% 1.9
6% 13% 2.0
6% 13% 2.0
6% 57% 9.1
6% 46% 7.4
6% 32% 5.2
19% 55% 2.9
6% 84% 13.5
6% 58% 9.3
DCV: Step vs Analog Control
•
Step control:
– “full %” when approaching CR or concentration target
– Return to “area %” @ point between CR and 400 ppm
•
Analog control recommended by 62.1 User Manual
– Area rate at 400-450 ppm
– Area + people (full) rate at CR target ppm
– Note difference in damper position vs OSA%
•
Analog is how DOE 2 models DCV according to people schedule
OSA
DCV Ventilation Sequence
Set Area & Full Positions based on OSA% measurements
40%
OApos
30%
OSA%
OAareaPos
@ 400
20%
10%
OAfullPos
@ AQset
0%
62
0
500
1000
Space/Return CO2 PPM
1500
2000
CR as
Indicator of People Concentration
• CR good up to typical high density
– Curve goes exponential at maximum densities
– 200 people per 1000 sf is 5 sf a person – rock concert SRO
– This is movie theater or assembly occupancy
OSA
CO2 CR as indicator of People
1.0 met, 5 cfm/person, 0.06 cfm/sf
50%
200
40%
160
30%
120
20%
80
10%
40
0%
0
63
500
1000
1500
Space/Return CO2 PPM
0
2000
OApos
DCV OA%
People
Linear Control is Quite Conservative
• Compare DCV to prescriptive rate
– High at lower proportional occupancy
– Helps offset lag due to space volume
OSA
DCV Conservative cf. Ventilation Rate
Compare Vent OA% required and DCV OA% provided
50%
200
40%
160
30%
120
20%
80
10%
40
0%
0
500
1000
1500
Space/Return CO2 PPM
64
0
2000
OApos
DCV OA%
Vent OA%
People
DCV Integrated Fan Control (DCV-IFC)
Yes, ASHRAE Standard 62.1 allows fan cycling
– General thinking: commercial fans must be ON during the occupied period
– Section 6.2.6.2 of ASHRAE Standard 62.1-2010 allows short-term
interruption of ventilation if ventilation levels are maintained on average
CO2 PPM as % of Full Design
3.0 hr Avg OA % of prescriptive
•
•
Exceeds design
occupancy
•
Fan Cycling Ventilation (cycles shown to meet ventilation only)
Vent OA damper
% Design Occupancy
160%
140%
Allow ed maximum ventilation increases as
OA approaches space temperature
120%
•
100%
80%
60%
•
40%
20%
65
Note: Startup ventilation compliant w ith 62.1 users manual
is typically below prescriptive levels under CO2 sensor
Fan operates 35.1% of tim e
5P
M
4P
M
3P
M
2P
M
1P
M
PM
12
AM
11
AM
10
9A
M
8A
M
0%
Approach:
-Record CO2 at
fan off
-Provide % air
relative to comfort
-Operate time
needed to match
ventilation rate
-Duty cycle if off
for 30 minutes
with no heat or
cool
RTU DCV Ventilation Issues
66
Improve ventilation
CO2 Transmitter Accuracy
Fan cycling
Control setup and adjustment
Alterations & Code
Current RTU
Conditions
AC-1
Pre-Retrofit Poor Ventilation
Wed, Jan 14, 2009
Near & Far sensors in same space (Far adjusted)
2,700
2,400
2,100
1,800
1,500
1,200
900
Fan on
PPM Carbon Dioxide
Proxy for Air Quality
• TAB rare for RTUs
• OA minimum often
set much too high
• Specific study of
senior center
600
300
: 00
12
PM
0
1 :0
PM
0
2 :0
PM
0
3:0
PM
0
4:0
Fan On
CO2 outside
CO2 RA
CO2 Space Far - Adj
CO2 Space Near
CO2 Target
• Like here, fan is often in auto position (~40% RTUs)
• During January, little daytime fan operation
– Heating load offset by internal gains
• CO2 concentration almost triple desired level
67
PM
0
5 :0
PM
0
6 :0
PM
Note: Fan is off most of the day
because heat from occupants met
heating load
RTU DCV Brings Significant
Ventilation Improvements
•
Fan slows when not heating or cooling, but remains on
• CO2 sensor modulates damper to maintain proper ventilation rate
AC-1
•
Multiple Room CO2 Measurement Divergance
Wed, May 27, 2009
8.0
7.0
1,100
6.0
5.0
900
4.0
700
3.0
2.0
500
1.0
300
0
9:0
68
Fan Amps
PPM Carbon Dioxide
Proxy for Indoor Air Quality
1,300
AM
0.0
A
:00
10
M
A
:00
11
M
P
:00
12
M
Fan Amps
CO2 Return Air
CO2 Computer Rm adj
CO2 Billiard Rm
CO2 outside
CO2 Target
•
After retrofit, highest CO2
concentration within target
limits.
Single return air sensor
worked for multiple
rooms:
– Computer lab
– Billiard room
Single CO2 Sensor Does It
Easiest placement of the
sensor is in the return air
AC-1
Mon, Jun 15, 2009
69
PPM Carbon Dioxide
Proxy for Indoor Air Quality
1,100
8.0
7.0
900
6.0
5.0
700
4.0
3.0
500
Fan Amps
Concerns:
– Studies suggest putting
sensor in the breathing zone
(California T24 code
required)
– When multiple rooms are
served by one unit, an
imbalance in ventilation
quality may occur with
uneven occupancy
– With VSD, less air throw at
the diffuser may reduce
ventilation effectiveness.
Multiple Room CO2 Measurement Divergance
2.0
1.0
300
0
1:0
PM
0.0
0
2:0
PM
0
3: 0
PM
0
4:0
PM
Fan Amps
CO2 Return Air
CO2 Computer Rm - Adj
CO2 Billiard Rm
CO2 outside
CO2 Target
Testing results:
• For smaller RTUs, even with one room heavily occupied,
return air sensing can meet requirements with a slight
ventilation setpoint adjustment
• Testing multiple places in one room showed little
variation with lower airflow
CO2 Transmitter Accuracy
Bin Occurance (Bars)
35%
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
30%
25%
20%
15%
10%
5%
0%
5%
10%
15%
20%
25%
30%
35%
Line is % of Units with
Worse Accuracy
CO2 Transmitter Error @ 1100 ppm
40%
+/- error
Source: National Building Controls Information Program. PTR. Iowa EC & CEC. June 2009.
70
Source:
http://www.energy.iastate.edu/
Efficiency/Commercial/
download_nbcip/PTR_CO2.pdf
NBCIP/ Iowa Energy
Center Testing
No manufacturers had all three samples within specs!
• Count on being within 200 ppm, usually high
• Many sensors saturate above 1700 ppm
• Self calibrating sensors improve persistence
Closer Look at Accuracy Impact
CO2 Transmitter Error @ 1100 ppm
10
42.5%
37.5%
32.5%
27.5%
22.5%
17.5%
12.5%
7.5%
2.5%
0.0%
-2.5%
-7.5%
-12.5%
0
-17.5%
5
-22.5%
Frequency n=45
15
error boundary
Source: National Building Controls Information
Program. PTR. Iowa EC & CEC. June 2009.
Eliminating 3 of 13 manufacturers
keeps error within +/- 21%
• Sensors are generally high; (mean = + 12%)
• Relative accuracy will maintain with self-calibration
• Particular manufacturers are at extremes
71
Questions ?
Reid Hart, PE
rhart@peci.org
REFERENCES
Hart, R., Price, W., Morehouse, D. (2006). The Premium Economizer: An Idea Whose Time
Has Come. In Proceedings of the 2006 ACEEE Summer Study on Energy Efficiency in
Buildings. Pacific Grove, CA: ACEEE.
Hart, R., Price, W., Taylor, J., Reichmuth, H. & Cherniack, M. (2008). Up on the Roof: From
the Past to the Future. In Proceedings of the 2008 ACEEE Summer Study on Energy
Efficiency in Buildings. Pacific Grove, CA: [ACEEE] American Council for an EnergyEfficient Economy.
Hart, R. (2009). “Premium Ventilation Package Testing: Short-Term Monitoring Report – Task
7.” Portland Energy Conservation, Inc. (PECI) for Bonneville Power Administration (BPA).
www.peci.org (Resources, commercial retail).
Jacobs, P., Smith, V., & Higgins, C. (2003). Small Commercial Rooftops: Field Problems,
Solutions and the Role of Manufacturers. In National Conference on Building
Commissioning: May (Vol. 20, p. 22).
Maxwell, Gregory, “Product Testing Report: Wall Mounted Carbon Dioxide (CO2)
Transmitters” (Iowa Energy Center, June 2009),
http://www.energy.iastate.edu/Efficiency/Commercial/download_nbcip/PTR_CO2.pdf.
Nassif, N., Kajl, S., and Sabourin, R. (2005). “Ventilation Control Strategy Using the Supply
CO 2 Concentration Setpoint.” HVAC&R Research, 11(2), 239–262.
72
73