NEEA XMP PUMPS RESEARCH Technical Work Group Meeting

NEEA XMP PUMPS RESEARCH Technical Work Group, Meeting # 5 April 17 th, 2019

AGENDA Time Topic Welcome Back, 10: 00– 10: 15 am Background and Introductions Details Welcome & Introductions Brief reminder on XMP background and TWG 10: 15– 10: 30 am Review of Final Sample Frame 10: 30– 11: 05 am Review of Analysis Results 11: 05 – 11: 15 am Case Studies Timeline and XMP Next 11: 15 – 11: 20 am steps 11: 20– 11: 30 am Discussion and Questions purpose, goals & objectives, expectations Goal of today’s meeting Review the data that will be included in the analysis Review the final disposition of the Sample Frame Review the pumps that will be incorporated into the analysis in the coming months Review the data management approach Review the Extrapolation Method Review current analysis Who Warren Fish, NEEA Geoff Wickes, NEEA Sarah Widder & Nate Baker, Cadeo Group Sarah Widder, Discuss the case studies that will be produced for Cadeo Group & the Initiative Warren Fish, NEEA Review next steps for Analysis and RTF work Discuss parallel goals and next steps for overall XMP program Sarah Widder, Cadeo Group & Warren Fish, NEEA Sarah, Warren, Geoff, and ALL 2

SUMMARY Have we met the target pump recruitment? YES, except: • Requires some modification of sample frame targets to reflect realities of pump market and application/data availability (specifically called out as necessary within the RTF Research Plans) • Need to complete the data identification with current outreach prospects Efficient C&I Pumps Met overall target for *most* sub-applications • Control-specific targets were harder, typically in sub-applications that are less common in the PNW Circulators Met target for most important sub-applications Met target for DHW, with adjustments to Single Family (not very common in PNW and well characterized in literature) and Adaptive Pressure Control Method Sample Frame for Commercial (similar to MF) to meet the target Do not have HH applications in MF, need to work on filling this with further recruitment. Also removing commercial targets due to overlap with C&I Heating sample frame 3

REVIEW OF FINAL SAMPLE FRAME Efficient C&I Sample Frame Application Definitions Pump Type Application Name Examples Efficient C&I Pump Commercial Cooling Tower condenser loop (NOT WSHP loop) Efficient C&I Pump Commercial Cooling WSHP cooling loop pump, chilled water pump Efficient C&I Pump Commercial Heating WSHP heating loop pump, Boiler HW pump to AHU coils Efficient C&I Pump Commercial DHW Circulation DHW recirc loop in C&I to end-uses, DHW recirc to bank of water heaters Efficient C&I Pump Commercial Pressure Boost Domestic water pressure boosting pumps/skid Efficient C&I Pump Industrial Process pumps, boiler feedwater pumps, or process cooling Efficient C&I Pump Agricultural Irrigation pumps Efficient C&I Pump Municipal Water Distribution/Treatment Effluent pumps, water distribution pumps 4

EFFICIENT C&I PUMPS Summary Pumps that have been identified Application Commercial Cooling Tower Commercial Cooling Commercial Heating Commercial Domestic Hot Water Circulation Commercial Pressure Boost Agricultural Irrigation Industrial - General Industrial - Boiler feedwater Operating Hours Energy Usage Variable Speed Constant Variable Constant Speed Baseline Pumps Identified: 203 Goal: 205 Current status: Primary Data Collection* Total Confirmed 8 48/30 Combine dual VL baselines 5 19 16 110/30 37/30 Complete Combine dual VL baselines 0 2 2/15 43 Changes 48 5 103 43 9 28 89 2 2 0 23 11 12 16 25/20 139 23 30 15 12 N/A 14 45/30 23 Complete Split into specific sub-apps 13 12 9 13/15 No variable speed data Industrial – Cooling/Condenser 30 22 8 0 16 30/15 Complete Municipal Water Distribution Treatment 36 11 25 0 0 36/35 Combine dual VL baselines 13 8 99% 0 0 *The Primary Data Collection is incorporated into the 20 identified sites for each sector and application Combine with DHW Circ Combine dual VL baselines 5

REVIEW OF FINAL SAMPLE FRAME Circulator Application Definitions Application Name Examples Domestic Hot Water Recirculation Heat Recovery Loops, Underfloor heating, Digester Loop Pump Residential and Commercial Domestic HW Sector Name Examples Hydronic Heating Single Family Residential Multi-Family Residential Commercial Detached/semidetached home or apartment units served by individual systems Apartment complex, Independent living facility; units served by central system Office buildings/University Buildings Control Type Control Method Definitions Single Speed (24/7)* Pump is not controlled, recirculation is constant Changes in pressure drive pump operation Adapts to changes in demand by Adaptive Pressure*** adjusting flow on a proportional pressure curve Controls system flow based on a Temperature**** temperature setpoint Controls operate pump only when Aquastat** needed to maintain a given return water temperature Controls “learn” when water is Learning Mode** being used and only operate pump during those time periods Pump operates when hot water is On-Demand** needed (as indicated by manual or occupancy-based signal) Pressure* Speed Controls Run-Hours Controls * applicable to Hydronic Heating pumps only ** applicable to DHW pumps only *** applicable to both Hydronic Heating and DHW pumps **** Temperature Control differs from Aquastat in that it adjusts the speed of the pump operation as opposed to the operating state of the pump

CIRCULATORS DHW Summary HH Summary Pumps Identified: 25 Goal: 45 Current status: Circulators that have been identified We also have literature to help fill in this sample frame 56% Hydronic Heating Application Pumps Identified: 35 Goal: 35 Current status: 100% Domestic Hot Water Single Speed Pressure Adaptive Temperature Pressure Single Family 21/5 10/5 8/5 6/5 Multi-Family 0/5 0/5 Commercial 5/5 0/5 Primary Data Collection* 24/7 Operation Adaptive Pressure Learning On. Demand 18/3 5/5 3/3 5 2/2 12/3 0/5 0/3 24/5 6 6** 18/3 0/5 5/3 17/5 2 7** 19/3 8/5 (1 com) *The Primary Data Collection is incorporated into the colored identified sites for each sector and application ** 6 are temperature-based demand, not manual or occupancy-based At last TWG, removed targets for DHW Learning and On-Demand for Multi-Family and Commercial due to limited availability and questions about applicability 7

DATA COLLECTION NEXT STEPS Collect outstanding pump data • ~30 Existing Data (estimated end of April) • ~45 Primary Data (estimated end of May) Incorporate into the Database • Using Data Management Process/QAQC Process Questions? Next: DATA MANAGEMENT STRUCTURE 8

DATA SUBMITTAL PROCESS Data is submitted via Data Submission Template 9

DATA MANAGEMENT STRUCTURE Data Submitted Operational Data Cleaning of Raw Data QAQC Raw Data Table Audit Data Standardization of Data - Time stamps - Units Standardized Data QAQC Standardized Data Table QAQC Database Audit Tables Annualization of Data -Based on known Calculations constraints by application Annualization Data QAQC Results (Annual Energy Usage, Load Profile, Op. Hrs, etc. ) Annualized Data Table XMP Database 10

DATA MANAGEMENT STRUCTURE Audit Data QAQC The Audit Data is reviewed and crosschecked with Manufacturer information to ensure accuracy This is to ensure the information that was submitted is accurate and aligns with the mechanical components of the pump submitted Common issues: The pump model number does not align with the pump class A pump class that cannot have more than one stage is listed as having more than one stage Also use this as a chance to standardize the information that was submitted so it can be uploaded to the Database *If any changes need to be made to the submission the data submitter is notified 11

DATA MANAGEMENT STRUCTURE Operational Data The Operational Data is stored in the database in 3 different forms: Raw, Standardized, and Annualized Raw Submitted Data Standardized Submitted Data as submitted Annualized Data to the research No analysis or manipulation Standardized units and timestamps 8760 hours of Variables calculated Operational Data if needed Standardized Units No extrapolation or interpolation XMP Pumps Research Database Weather normalized, if needed QAQC Report 1 QAQC Report 2 QAQC Report 3 12

DATA MANAGEMENT STRUCTURE Annualization Develop model on actual submitted data Review characteristics of actual and annualized data for reasonableness: Daily, weekly, monthly trends 13

DATA MANAGEMENT STRUCTURE Any Questions on Data Management, QAQC, or Annualization of Data? Next: DATA ANALYSIS 14

DATA ANALYSIS Different approaches are used for the Operating Hours Analysis and Adjustment Factor Analysis Adjustment Factor Operating Hours Commercial Industrial Variables that impact Adj. Factor • High-level analysis of Adjustment Factor Municipal Agricultural Adj. Factor Analysis • • Load Profile Motor Oversizing BEP Offset Static Head • Review the Adj. Factor Analysis incorporating information/insights from Variable Analysis Re-analysis of Adj. Factor

OPERATING HOURS RTF Operating Hours and Sources for the estimates Applications Commercial HVAC and DHW Industrial and municipal Agricultural irrigation Operating Hours 4000 5000 2400 Application Source Notes: Commercial HVAC and DHW DOE Rulemaking, Pump Unknown Subcommittee Industrial and Municipal Northwest motor database pumping applications, pump subcommittee Over 3000 pumps in the Motor Database Agricultural Irrigation Green Motor Rewind Program, BPA Ag Irrigation Program 280 pumps in sample 16

AGRICULTURAL IRRIGATION, Op. Hrs Sample Frame Disposition Data Source High-Level Statistical Analysis No. of Pumps Statistic All Pumps, Observed RTF Op. Hrs Estimate 139 280 Collected for VSD Upgrades 21 No. of Pumps in the sample Collected for Pump Tests 41 Average 2, 358 2, 386 Collected for VSD Standard Protocol Analysis 77 Max 6, 709 7, 745 Min 240 814 Standard Deviation 1, 120 1, 225 Coefficient of Variation 0. 48 0. 51 Standard Error 95 73 Load Control Method No. of Pumps Constant Load 52 80% Margin of Error 122 94 Variable Load 87 90% Margin of Error 157 121 80% Confidence interval 2236 to 2479 2293 to 2480 90% Confidence Interval 2201 to 2514 2265 to 2507 17

AGRICULTURAL IRRIGATION, Op. Hrs Motor HP No. of Pumps in the sample Average (hrs) 5 2 3, 910 10 3 697 15 5 2, 398 20 3 1, 525 25 12 2, 137 30 18 2, 485 40 13 2, 552 50 11 2, 380 60 9 2, 304 75 19 2, 418 100 14 1, 919 125 8 2, 215 150 15 2, 800 200 7 2, 603 Average Operating Hours by Motor HP, with 90% confidence intervals 5 000 4 500 Average Operating Hours 4 000 3 500 3 000 Average Operating Hours, by Horsepower 2 500 2 000 RTF Op. Hrs Estimate 1 500 1 000 500 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 Motor HP 18

COMMERCIAL HVAC and DHW, Op. Hrs High-Level Statistical Analysis Statistic All Pumps, Observed Cooling Tower Heating Pressure Boost No. of Pumps in the sample 101 55 27 7 12 Average 3, 987 3, 417 3, 509 4, 530 7, 358 Max 8, 760 7, 866 8, 280 8, 463 8, 760 Min 37 37 86 2, 490 620 Standard Deviation 2, 653 1, 971 2, 778 2, 007 3, 112 Coefficient of Variation 0. 67 0. 58 0. 79 0. 44 0. 42 Standard Error 264 266 535 758 898 80% Margin of Error 338 340 684 971 1, 150 90% Margin of Error 435 439 882 1, 251 1, 482 80% Confidence interval 3649 to 4325 3077 to 3757 2824 to 4193 3559 to 5501 6208 to 8508 90% Confidence Interval 3551 to 4422 2978 to 3855 2627 to 4391 3279 to 5781 5876 to 8840 The current RTF Measures have HVAC and DHW combined This indicates that the two applications should be separated Next step was to look at the Commercial Op. Hrs by Load Control 48 more Cooling Pumps expected 15 more Cooling Tower Pumps expected 33 more Heating Pumps expected 11 more Pressure Boost Pumps expected 19

COMMERCIAL HVAC, Op. Hrs Pump Operating Hours, by Load Control Method The majority of the pumps operating at 8760 hrs/year are pressure boost 10000 9000 Average Operating Hours 8000 7000 Statistic 6000 n Constant Load Operating Hours 5000 4000 3000 2000 1000 0 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 Motor HP All Pumps, Observed Without Pressure Boost 105 93 12 Average 3, 808 3, 477 7, 358 Max 8, 760 Min - - 620 Standard Deviation 2, 719 2, 362 3, 112 Coefficient of Variation 0. 71 0. 68 0. 42 Standard Error 265 245 898 90% Margin of Error 438 404 1, 482 3370 to 4246 3073 to 3881 5876 to 8840 90% Confidence Interval 20

INDUSTRIAL AND MUNIC. , Op. Hrs High-Level Statistical Analysis Statistic All Pumps, Observed Industrial Municipal No. of Pumps in the sample 59 37 22 Average 5, 469 7, 280 2, 423 Max 8, 760 Min - 154 - Standard Deviation 3, 207 2, 003 2, 455 Coefficient of Variation 0. 59 0. 28 1. 01 Standard Error 417 329 523 80% Margin of Error 534 421 670 90% Margin of Error 689 543 864 80% Confidence interval 4934 to 6003 6858 to 7701 1753 to 3093 90% Confidence Interval 4780 to 6157 6736 to 7823 1559 to 3286 The current RTF Measures have Industrial and Municipal pumps combined. We are seeing a big difference in Op. Hrs between the 2 sectors Next step was to look at the Industrial Op. Hrs by sub-application *22 more Industrial Pumps expected *14 more Municipal Pumps expected 21

ADJFACTOR, High-Level Analysis Pump is off High Level Adj. Factor Statistic Number of pumps Average Adjustment Factor Min Adj Factor Max Adj Factor Standard Deviation Standard Error 90% Margin of Error 90% Confidence Interval Value 165 0. 899 2. 800 0. 626 0. 049 0. 080 0. 899 ± 0. 08 A variable load pump (low PEI) that operates at 100% the entire year Observed Adj. Factor by Operating Hours Observed Adj. Factor by Motor HP 3, 00 2, 00 1, 50 1, 00 0, 50 0, 00 0 2000 4000 6000 Operating Hours 8000 10000 Obs. Adj. Factor 2, 00 Observed Adj. Factor by PEI 2, 50 Obs. Adj. Factor 3, 00 2, 00 1, 50 1, 00 0, 50 0, 00 0 50 100 150 Motor HP 200 250 300 0, 20 0, 40 0, 60 PEI 0, 80 1, 00 1, 20 22 1, 40

ADJFACTOR, High-Level Analysis Load Control Adj. Factor Statistic No. of Pumps in the sample Constant Load Variable Load 64 101 0. 900 0. 898 Min - - Max 1. 649 2. 800 Standard Deviation 0. 337 0. 756 Standard Error 0. 042 0. 075 90% Margin of Error 0. 069 0. 124 0. 9 ± 0. 069 0. 898 ± 0. 124 Average 90% Confidence Interval 23

MOTOR OVERSIZING RTF Operating Hours and Sources for the estimates Motor Oversizing, by Motor HP 275% Statistics No. of Pumps in the sample Motor Oversizing 250% Trim 225% 220 me 200% 24% 175% Max 131% 150% Min -59% Standard Deviation 28% Coefficient of Variation 1. 18 %Motor Oversizing Average Imp elle pell er r wa s tr imm ed 125% d Im from 12" to 9" 100% 75% 50% 25% Standard Error 2% 0% 90% Margin of Error 3% -25% -50% 90% Confidence Interval 21% to 27% -75% 0 20 40 60 80 100 120 140 Motor HP 160 180 200 220 240 260 24

MOTOR OVERSIZING RTF Operating Hours and Sources for the estimates 275% 250% Imp 9" eller w a 225% s tr 200% 175% 50% 150% %Motor Oversizing The smaller pumps have a larger % difference because of the relative impact mis-sizing a pump has on the % oversizing Average Motor Oversizing, grouped Motor Oversizing, by into Motor HP HP bins 125% Imp elle R 2 = 0, 8677 imm ed from 12" to r wa s tr imm ed from 12" to 9" 100% Investigating the impact varying motor oversizing has on Energy Consumption Model 75% 50% 25% 0% -25% -50% -75% 0% 0 - 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150160 170180 190200 210220 230240 250260 Motor HP 25

LOAD PROFILE RTF Operating Hours and Sources for the estimates 80% Constant Load - Load Profiles, All Applications 45% 40% 60% 50% RTF Load Profile 40% Observe d Load Profile 30% 20% 10% Percent of Operating Hours 70% Percent of Operating Hours Variable Load - Load Profiles, All Applications 35% 30% RTF Load Profile 25% 20% Observe d Laod Profile 15% 10% 5% 0% 0% 25% 50% 75% 100% Percent of BEP Flow 110% Above 110% 25% 50% 75% 100% 110% Above 110% Percent of BEP Flow 26

DATA ANALYSIS, Next Steps • • • Finalize Interim Analysis Results, Commercial and Industrial Pumps April 20 th Interim Report, End of April Incorporate the rest of pump data into database, mid-May Circulator Analysis/Update C&I Analysis, late-May Final draft report and analysis likely available in June Next: CASE STUDIES 27

CASE STUDIES Five targeted case studies to illustrate potential of XMP program from both a savings and implementation perspective Savings Opportunities from Smart Pumps & HI PSAP Certification Savings from Pump Efficiency Monitoring and Optimization Pump and Water Heater Energy Savings from Circulator Controls Benefits of Pump Monitoring in Preventative Maintenance • University of Washington • Replacing existing CL hydronic heating pumps with new “smart pumps” • Also demonstrating additional “system” savings via Pipe. Flo® modeling (from PSAPcertified professional) • Partnership with SCL, Pump mfgrs, Engineered Software, UW, NEEA • Georgia Pacific Plant in Toledo • Monitoring 5 “smaller” (75 -200 hp) clean water pumps with new KCF pump efficiency monitoring system to optimize pump operation and collect data for research study • Partnership with KCF, BPA ESIP, NEEA, GP • Bellwether & Plymouth Low-Income Multifamily Housing • Replacing existing uncontrolled, constant speed pumps on DHW and HH systems with new efficient ECM circulators with efficient control options • Partnership with Emerald Cities, Bellwether & Plymouth Housing, NEEA, Energy 350, Pump mfgrs • Identified failing pump prior to failure and was able to schedule planned replacement • Reduced downtime, increased reliability and energy savings • Partnership with Kidder Matthews We will be reaching out to participating utilities Benefits from Pump Booster Retrofits • Opportunities for variable speed booster skids in large commercial buildings • Partnership with Grundfos 28

OTHER XMP RESEARCH EFFORTS $/HP • Fielding parallel market research activity to verify C&I and circulator market characteristics and current practice baseline • Developing next stage of program development research • Pursuing targeted distributor outreach to pilot mid-stream incentives and program approach • Goal of 4 distributors enrolled in 2019 Interested in folks who may like to participate in a follow-on advisory group for the XMP program research 29

HIGH LEVEL NEXT STEPS Data Collection & Analysis • Finalize data collection & analysis Reporting • Compile and finalize report and database (Q 2/3 2019) • Share with RTF and use to update Pump & Circulator Planning measures at the RTF, likely through Pumps subcommittee Other XMP Activities • Finalize case studies • Complete market research • Continue distributor outreach Let us know if you’d like to participate in RTF Pumps subcommittee! FINAL TWG Meeting • Week of July 8 th or 22 nd • Goal: send draft report and database for review in advance of meeting. • Use meeting to discuss high-level results and takeaways and answer any questions 30

Questions? Contact Us! Sarah Widder swidder@cadeogroup. com Nathan Baker nbaker@cadeogroup. com Warren Fish wfish@neea. org 31

Thanks for your time! 46

WELCOME AND INTRODUCTIONS Name Organization Brent Ross Armstrong Fluid Technology Gabor Lechner Armstrong Fluid Technology Andy Paul Avista Corp Todd Amundson Bonneville Power Administration Erin Hope Bonneville Power Administration Mike Bailey Energy Trust of Oregon Chris Ireland Grundfos Devin Carle Hurley Engineering Edgar Suarez Hydraulic Institute Randy Thorn Working Group participants selected to represent a range of perspectives across the region and the industry Supported by Internal Team Name Organization Idaho Power Warren Fish Northwest Energy Efficiency Alliance Nicholas Ricciardi Lockheed Martin Geoff Wickes Northwest Energy Efficiency Alliance Melanie Danuser Northwest Energy Efficiency Council Kevin Smit Northwest Power and Conservation Council Kathryn Bae Northwest Energy Efficiency Alliance Patrick Moore Pacific Gas & Electric Steve Phoutrides Northwest Energy Efficiency Alliance Nancy Goddard Pacifi. Corp Christian Douglass Ptarmigan Research/ RTF Maria Alexandra Ramirez Northwest Energy Efficiency Alliance Jim White PUD No. 1 of Chelan County Alisyn Maggiora Northwest Energy Efficiency Alliance Chao Chen Puget Sound Energy John Owen Seattle City Light Sarah Widder Cadeo Group Lucie Huang Seattle City Light Nate Baker Cadeo Group Jim Conlan Snohomish County PUD Nick O’Neil Energy 350 Rick Rosenkilde Snohomish County PUD Allison Grinczel Snohomish County PUD Mark Chaffee Taco Comfort Solutions Matthew Walker Tacoma Public Utilities Shari Kelley Xcel Energy 33

XMP TECHNICAL WORK GROUP Meeting #5 Goal of the Technical Work Group: • Guide the pumps energy savings research to meet the XMP Initiative targets • Provide insight, expertise, and data to support successfully completing research objectives Objectives for this Meeting • Review the outcome of the data collection • Review the analysis performed on collected data Expectations • Keep comments constructive, any concerns/issues must also have a (feasible) suggestion that would resolve the issue • Keep comments and suggestions focused on meeting key research outcomes and XMP objectives 34

OVERVIEW For each sample frame we looked at both the number of pumps collected against targets and the distribution of pumps to ensure complete and representative data Each Sample Frame Summary Slide Reviews: • Stoplight Ranking • Outreach Complete for Adjusted Sample Frame • May not complete sample frame target, final outreach currently underway • Notes/Next Steps 35

COMMERCIAL COOLING TOWERS Variable Speed Source Operating Hours Constant speed Variable speed baseline Constant speed baseline Total Existing Data 34 2 32 0 34/32 Primary Data 8 3 5 0 8/3 Total 42 5/10 38/15 0/5 42/35 Revised Total 42 5/5 38/20 42/35 7 Number of Data Contributors Number of Data Sites 15 All In Scope? 8 Extraregional sites Horsepower Distribution 1 -3 HP 3 -10 HP 10 – 25 HP 25 – 200 HP Out of Range Unknown Primary Data 1 - - 6 - 1 Existing Data - 13 4 20 - 0 Total 1 13 4 23 - 1 Status/Notes: Geographic Distribution - Idaho Seattle Portland Extra-regional Sample frame is complete with decreased CS target (10 ->5) and removing dual VL baselines. 36

COMMERCIAL COOLING Variable Speed Operating Hours Constant speed Variable speed baseline Constant speed baseline Total Existing Data 84 5 74 5 84/27 Primary Data 19 4 15 0 19/3 Total 103 9/10 89/15 5/5 103/30 Source 17 Number of Data Sites 36 All In Scope? 18 Extra. Regional Sites Horsepower Distribution 1 -3 HP 3 -10 HP 10 – 25 HP 25 – 200 HP Out of Range Unknown Primary Data 2 1 5 9 - 2 Existing Data 2 21 5 43 7 13 Total 4 22 10 52 7 15 Status/Notes: Number of Data Contributors Geographic Distribution - Idaho Seattle Richland Tacoma King County Sample Frame Complete 37

COMMERCIAL HEATING Variable Speed Source Operating Hours Constant speed Variable speed baseline Constant speed baseline Total Existing Data 27 12 15 0 27/37 Primary Data 13 13 0 0 13/3 Total 40 25/10 15/10 0/10 40/30 25/10 Revised Total 15/20 40/30 Number of Data Contributors 15 Number of Data Sites 15 All In Scope? YES Horsepower Distribution 1 -3 HP 3 -10 HP 10 – 25 HP 25 – 200 HP Out of Range Unknown Primary Data 4 - - 8 Existing Data 4 5 9 1 - 8 Total 8 5 10 1 - 16 Status/Notes: Sample frame is complete after removing dual VL baselines. Geographic Distribution - Portland Seattle Bellingham Richland Tacoma Snohomish County 38

COMMERCIAL DHW RECIRC Variable Speed Source Operating Hours Constant speed Variable speed baseline Constant speed baseline Total Existing Data 0 0 0/7 Primary Data 2 2 0 0 2/8 Total 2 2/5 0/5 2/15 Revised Total Combine with DHW Circ Number of Data Contributors 1 Number of Data Sites 1 All In Scope? YES Horsepower Distribution 1 -3 HP 3 -10 HP 10 – 25 HP 25 – 200 HP Out of Range Unknown Existing Data - - - Primary Data 2 - - - Total 2 - - - Geographic Distribution - Seattle Status/Notes: Through Conversations with Distributors/Operators this application is most often served by Circulators (vary rarely have a pump above 3 HP) 39

COMMERCIAL PRESSURE BOOST Variable Speed Source Operating Hours Constant speed Variable speed baseline Constant speed baseline Total Existing Data 7 1 7 0 8/12 Primary Data 16 10 4 1 15/8 Total 23 11/10 11 1 23/20 11/10 Revised Total 12/10 23/20 Number of Data Contributors 12 Number of Data Sites 14 All In Scope? YES Horsepower Distribution 1 -3 HP 3 -10 HP 10 – 25 HP 25 – 200 HP Out of Range Unknown Existing Data - 7 - 1 - - Primary Data 3 6 4 - - 1 Total 3 14 4 1 - 3 Geographic Distribution - Seattle Portland Status/Notes: Sample frame is complete after removing dual VL baselines. 40

AGRICULTURAL IRRIGATION Energy Use Source Operating Hours Variable Speed Constant Speed Existing Data 139 30 Primary Data NA NA 139/30 30 Total Variable speed baseline Constant speed baseline 12 NA 42 NA 12 Total NA Number of Data Contributors 3 Number of Data Sites 8 42/35 California Data All In Scope? Incorporated Horsepower Distribution Geographic Distribution All Pumps in Large (25 -200 HP) HP Bin Status/Notes: This application is being filled using Variable-Specific Pump Data. Not all pumps submitted have information for both operating hours and energy usage - Chelan County California Region-wide (RTF Data) 41

INDUSTRIAL, GENERAL Variable Speed Source Operating Hours Constant Speed Variable speed baseline Constant speed baseline Total Existing Data 14 6 8 0 14 Primary Data 9 9 0 0 9 Total 23 13 1 0 23 Horsepower Distribution 1 -3 HP 3 -10 HP Existing Data 10 – 25 HP Number of Data Contributors 12 Number of Data Sites 12 All In Scope? 25 – 200 HP Out of Range 6 Unknown 8 Primary Data 1 1 5 1 Total 1 1 11 1 - 1 9 Status/Notes: We had already collected Industrial data that did not fall into Condenser or BFW before we separated the applications. Included the Industrial, General application to support the need for separate industrial applications YES Geographic Distribution - Salem Mukilteo Weed Grandview Astoria Bend Hood River Portland 42

INDUSTRIAL, CONDESNER Variable Speed Operating Hours Constant speed Variable speed baseline Constant speed baseline Total Existing Data 14 6 8 0 14/10 Primary Data 16 16 0 0 16/5 Total 30 22 8 0 30/15 Source 1 -3 HP 3 -10 HP 10 – 25 HP 25 – 200 HP Out of Range Unknown Existing Data - - 2 10 - 2 Primary Data 8 - 5 - 3 Total 8 2 15 - 5 Status/Notes: 6 Number of Data Sites 6 All In Scope? Horsepower Distribution 5 Number of Data Contributors YES Geographic Distribution - Newberg Clackamas Portland Albany Eugene Sample Frame Complete 43

INDUSTRIAL, BOILER FEEDWATER Source Operating Hours Constant speed Variable Speed Variable speed Constant speed baseline Total Existing Data 4 4 0 0 4 Primary Data 9 9 0 0 9 Total 13 13/5 0/5 13/15 Number of Data Contributors 9 Number of Data Sites 9 All In Scope? Horsepower Distribution 1 -3 HP 3 -10 HP 10 – 25 HP 25 – 200 HP Out of Range Unknown Existing Data - - - 3 Primary Data 3 2 2 - 2 Total 3 2 2 - 5 - Status/Notes: Almost complete, but no variable speed data (potentially due to limited applicability). YES Geographic Distribution - Eugene Portland Astoria Hood River Newberg 44

MUNICIPAL WATER TREATMENT Variable Speed Source Operating Hours Constant Speed Variable speed baseline Constant speed baseline Total Existing Data 36 11 25 0 36/30 Primary Data 0 0 0/3 Total 36 11 25 0 36/35 36/30 11/10 Revised Total 25/25 36/35 Horsepower Distribution 1 -3 HP 3 -10 HP 10 – 25 HP 25 – 200 HP Out of Range Unknown Existing Data - - 4 30 - 2 Primary Data - - - Total - - 4 30 - 2 Status/Notes: Sample frame is complete after removing dual VL baselines. Number of Data Contributors 10 Number of Data Sites 15 All In Scope? YES Geographic Distribution - Portland Salem Everett Astoria Corvallis Lincoln City Pendleton King Co. 45

HYDRONIC HEATING Application Single Family, Existing Single Family, Primary Data Total_ Application Multi-Family, Existing Multi-Family, Primary Data Total_ Single Speed 0 20 15 Single Speed 0 0 0 Pressure Adaptive Pressure Temperature Total 0 0/20 8 8 6 6 4 4 20/3 20/20 Number of Data Contributors 12 Pressure Adaptive Pressure Temperature Total 0 0 0 0 0/20 0/3 0/20 Number of Data Sites 12 All In Scope? Application Commercial, Existing Commercial, Primary Data Total_ Revised Total Single Speed 0 5 5 Pressure Adaptive Pressure Temperature Total 0 0 0 0 0 Combine with Efficient C&I Commercial Heating 0/20 5/3 5/20 Status/Notes: SF & Com sample frames complete after combining Commercial Hydronic Heating Circulators with Efficient C&I Commercial Heating Focus resources on filling the applications in the multi-family sector (outreach currently underway) Yes Geographic Distribution - Bend Clacksman Helena Portland Seattle 46

DOMESTIC HOT WATER RECIRC. 24/7 Operation Single Family, Existing 1 Single Family, Primary Data 5 Total_ 6/5 Adaptive Pressure 1 2 3/3 24/7 Operation Multi-Family, Existing 6 Multi-Family, Primary Data 18 Total_ 24/5 Adaptive Pressure 0 8 8/5 24/7 Operation Commercial, Existing 2 Commercial, Primary Data 19 Total_ 21/5 Adaptive Pressure 0 1 9/5 Application Learning On Demand Total 1 4 5 1 1 2/2 4/20 12/3 16/15 Learning On-Demand Total 4 2 2 4 0 4 14/20 28/3 42/20 Learning On-Demand Total 2 0 2 1 6 7 5/20 26/20 31/20 Status/Notes: MF and Commercial sample frames are complete if Adaptive Pressure cases are used in both sectors. SF complete after reducing sample targets for Adaptive Pressure (limited savings) and On Demand (savings characterized well in existing literature). Number of Data Contributors 23 Number of Data Sites 33 All In Scope? Extra-Regional Data Geographic Distribution - Salem Portland Spokane Seattle Hood River Clacksman Milwaukie Takoma Tillamook 47