Laboratory HVAC Testing Research PlanCalifornia Public Utility Commission, Energy DivisionProposal Reference # 09PS5863BContract # 12PS5119Prepared by KEMA, Inc.5/6/2023

Copyright © 2014, KEMA, Inc.This document, and the information contained herein, is the exclusive, confidential and proprietary property of KEMA, Inc. and is protected under the trade secret and copyright laws of the United States and other international laws, treaties and conventions. No part of this work may be disclosed to any third party or used, reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or by any information storage or retrieval system, without first receiving the express written permission of KEMA, Inc. Except as otherwise noted, all trademarks appearing herein are proprietary to KEMA, Inc.

LEGAL NOTICEThis report was prepared under the auspices of the California Public Utilities Commission (CPUC). While sponsoring this work, the CPUC does not necessarily represent the views of the Commission or any of its employees except to the extent, if any, that it has formally been approved by the Commission at a public meeting. For information regarding any such action, communicate directly with the Commission at 505 Van Ness Avenue, San Francisco, California 94102. Neither the Commission nor the State of California, nor any officer, employee, or any of its contractors or subcontractors makes any warrant, express or implied, or assumes any legal liability whatsoever for the contents of this document.

Table of Contents

1. Background.......................................................................................................1-11.1 Quality Maintenance and Laboratory Work Synergies............................1-8

1.1.1 Laboratory Based HVAC Unit Testing.............................................1-101.1.2 Laboratory Based Field Instrument Testing...................................1-11

2. Objectives.........................................................................................................2-12.1 Key Research Questions..........................................................................2-12.2 Parameters Evaluated by Measure (Impact)...........................................2-2

3. Methods............................................................................................................3-13.1 Year 1 Laboratory Testing Priorities.......................................................3-1

4. Work Order Task Descriptions and Budget......................................................4-14.1 Year 1 ($500,000)....................................................................................4-1

4.1.1 Task 1: Develop HVAC Laboratory Testing Plan ($95,000)..............4-14.1.2 Task 2: HVAC Laboratory Testing—Continue 2010-12 Testing

($205,000).............................................................................4-24.1.3 Task 3: Support CPUC Consultants’ Collaboration of IOU HVAC

Laboratory Testing ($100,000).............................................4-34.1.4 Task 4: Lab Facility Pre-Payments ($300,000)..................................4-3

4.2 Year 2 ($300,000)....................................................................................4-4

List of Figures

Figure 1: Test Equipment Schematic 1-13

List of TablesTable 1: Tests for Manufacturer #1 7.5-ton R-22 non-TXV, 2-Circuit Unit (RTU3)3-1Table 2: Tests for Manufacturer #2 7.5-ton TXV, 2-Circuit Unit (RTU1)...............3-4Table 3: Tests for Manufacturer #2 7.5-ton TXV 2-Circuit Unit (RTU2)................3-4Table 4: Tests for Manufacturer #3 3-ton TXV 1- Circuit Unit (RTU4)..................3-5Table 5: Tests for Manufacturer #1 3-ton non-TXV 1-Circuit Unit (RTU5)Type....3-6Table 6: Tests for Manufacturer #1 3-ton R-410A non-TXV 1-Circuit Unit (RTU6). 3-

6Table 7: Tests for Manufacturer #4 3-ton R-410A TXV 1-Circuit Unit (RTU7)......3-7

KEMA, Inc. i 5/6/23

Table of Contents

Table 8: Year 1 Summary Budget...........................................................................4-1Table 9: Task 1 Budget...........................................................................................4-2Table 10: Task 2 Budget.........................................................................................4-2Table 12: Task 3 Budget.........................................................................................4-3Table 13: Task 4 Budget.........................................................................................4-4

KEMA, Inc. ii 5/6/23

1. BackgroundQuality Maintenance (QM) and Quality Installation (QI) laboratory and field testing activities span multiple program cycles. Over the expected long term duration of the activity, evaluation teams will test packaged commercial and residential split HVAC systems using R-22 and R-410a as well as the heating components of these systems. Laboratory and field testing form integral parts of the overall QM and QI impact evaluation as they provide insight into the impacts of service actions that are not possible in field-only or lab-only studies. Laboratory and field testing are also important for Database for Energy Efficiency Resources (DEER) Analyses and Investor Owned Utility (IOU) work papers1. Energy savings benefits provided by DEER for weather-dependent measures derive from eQUEST computer simulations of prototypical buildings whose cooling and heating requirements are based on calculated HVAC systems performance coincident with calculated space heating and cooling loads. The efficacy of this process is likely no better than the program’s ability to simulate real world HVAC system performance in workpapers. Current laboratory testing focuses on the cooling performance of numerous unitary systems covered under ANSI/AHRI Standard 340/360. Considering the impact of HVAC on grid peak electric loads, any improvement in actual system performance revealed by the laboratory test is important.

For buildings served by unitary cooling systems, the energy simulation of the cooling system in the DEER process uses a set of certified efficiency values and a number of performance maps that adjust rated values to non-rated conditions.2 Rated conditions include steady-state total and sensible cooling capacities and cooling efficiency and fan power values that occur at the AHRI “A” ratings point. The AHRI 210/240 and 340/360 “A” ratings point is defined as the system operating with an ambient temperature of 95ºF and return air conditions to the cooling coil of 80ºF dry bulb temperature and 67ºF wet bulb temperature. The DEER team derives performance maps from manufacturers’ expanded engineering tables and supply fan performance tables. These data are typically obtained from

1 DEER contains information on selected energy-efficient technologies and measures. DEER provides estimates of the energy-savings potential for these technologies in residential and nonresidential applications. The database contains information on typical measures – those commonly installed in the marketplace – and data on the costs and benefits of more energy-efficient measures. Energy-efficient measures provide the same services using less energy, but they usually cost slightly more. DEER updates have been developed by the California Public Utilities Commission (CPUC) with funding provided by California ratepayers. http://www.deeresources.com/2 Performance maps are bi-quadratic equations based on manufacturer performance data for makes and models of units sold in California.

KEMA, Inc. 1 5/6/23

heating and cooling system engineering literature and are likely based on computer simulations. The quality, completeness and usefulness of these data sets vary across manufacturers and system types. In almost all cases, some performance estimates at conditions not provided by the manufacturer are required to complete the performance maps. Laboratory tests help to clarify and develop those estimates not provided in manufacturers’ engineering literature and provide guidance on how rated conditions might differ from those that occur in the field. The evaluation team will test a significant sample of systems of different sizes from different manufacturers. The team hopes that generalized and scalable conclusions will be reached allowing wide application of the findings to other non-tested systems. This clarity is expected to provide improved estimates of cooling and heating system performance by modifying simulation algorithms in EQuest and, as a result, better estimates of DEER energy impact for all measures with cooling system impacts.

1.1 Past Laboratory Tests and DEER Impacts

The following examples explain how past and recent laboratory tests impact DEER assumptions.

Part-load performance of larger (greater than 65,000 Btuh rated net cooling capacity) units cannot be determined from manufacturers’ data.3 Cycling loss coefficients for single- or multi-compressor RTUs are assumed to be proportional to cooling capacity if the ratio of cooling capacity to peak design cooling loads is constant. Correct assessment of part-load data requires cycling tests, as is done for smaller systems via the prescribed “C” and “D” tests in AHRI Standard 210/240. Instead, larger systems use the IEER test, a weighting of steady-state values for various test conditions. This rating is not based on measured cycling losses, but rather uses an assumed loss curve that may or may not represent actual system performance.4 Current part-load performance maps are based on typical cycling losses associated with smaller systems (since they are required to include the “C” and “D” tests as part of their seasonal efficiency rating, or SEER). A set of recent “C” and “D” laboratory tests5 on a larger 7.5-ton packaged air conditioner found 3 Manufacturers do not provide part-load performance data in terms of a cycling loss curves for units greater than or equal to 65,000 Btu/hr. Expanded performance data on SEER-rated units (with cooling capacities less than 65,000 Btu/hr) can be used to estimate cycling-loss coefficients and, thus, project part-load operation that includes cycling losses.4 AHRI Standard 340/360-2007 uses indoor conditions of 80F drybulb and 67F wetbulb and the following outdoor drybulb conditions to calculate IEER ratings: 95F (100%), 71F (75%), 68F (50%), and 65F (25%). 5 Test results will be made available under the Package Unit Laboratory Testing Preliminary Results section of the HVAC Impact Evaluation Report: WO32 HVAC.

KEMA, Inc. 2 5/6/23

that the relative cycling losses for these large systems are more than double their smaller counterparts. Features added to smaller system to control cycling losses that increase their SEER rating were not used on larger commercial systems (greater than 65,000 Btu/hr) where cycling losses do not impact efficiency ratings. Additional tests are needed to obtain a more representative estimate of typical large system cycling losses, the results of which will be used to inform simulation software.

Tests of economizer damper outdoor air leakage indicate fully closed leakage of 12 to 30% and fully open air flow of 30 to 75%. Previous simulations assumed 0% outdoor air leakage with closed dampers and 100% outdoor air flow with open dampers. Tests are being performed at indoor conditions of 75F drybulb and 62F wetbulb and outdoor conditions of 82F and 68F wetbulb, 95F and 75F wetbulb, and 115F and 80F wetbulb.

Recent laboratory testing included “out of the box” testing of a unit prior to the testing that replicates AHRI rating conditions. DNV GL understands that the AHRI testing process does not represent typical installed conditions (such as required economizers). The most notable characteristic of the AHRI testing is the test external pressure on the system’s supply air fan. Manufacturers perform a number of changes to their system during the AHRI test procedure6. These typically include changes to the supply fan pulleys to reduce fan power, sealing of the test unit cabinet to control cabinet leakage (increasing rated capacity), addition of insulation in the cabinet base, and, on occasion, modification of system charge to achieve single-phase flow across mass flow meters (if used) or manufacturer specifications regarding discharge pressure, suction pressure, suction temperature, liquid temperature, superheat, subcooling, or approach temperature. In past testing, the DNV GL team compared the “out of the box” testing to the processes required to match AHRI tests conditions; this has provided important insights. These comparisons provided consistently lower (7% to 10%) steady state efficiency at the “A” ratings point for the system in its as-delivered (“out of box”) condition7. DNV GL plans to complete additional testing to quantify the sources of these efficiency differences and apply those findings to DEER DX HVAC system models or other building simulation models. Changes to system models will be done recognizing the limits of the tests conducted.

DNV GL has performed laboratory tests on systems with refrigerant charge at various levels, from -40% to +40% of factory charge on five (5) new obsolete stock R 22 commercial packaged cooling systems (one 7.5-ton non-TXV, two 7.5-ton 6 HVAC Impact Evaluation Report: WO32 HVAC7 HVAC Impact Evaluation Report: WO32 HVAC

KEMA, Inc. 3 5/6/23

TXV, one 3-ton non-TXV, and one 3-ton non-TXV)8. These tests show that packaged system energy efficiency performance is much less affected by system over- or under-charge than assumed in DEER charge fault measures9. Additionally, system data and other research studies indicate that Air Conditioner Maintenance (ACM) fault detection diagnostic (FDD) methods as typically applied in the field often indicate “false alarm” over- or under-charge, misdetection, or misdiagnosis of faults. Yuill and Braun evaluated the CEC Refrigerant Charge Analysis (RCA) protocol and reported 41% correct diagnosis for non-TXV and 64% correct diagnosis for TXV equipped systems.10 Yuill and Braun evaluated five FDD protocols and reported false alarm rates of 37 to 85% based on experimental data.11 FDD protocols are further confounded by the presence of economizers or ventilation air dampers that lead to incorrect measurement of evaporator coil entering air conditions upon which manufacturer or generic ACM charge charts are based. Additional complications are caused by improper airflow (<>400 cfm/ton), coil blockage, non-condensables, refrigerant restrictions, or measurement instrument errors which the ACM FDD methods assume are not present. Laboratory and field tests of unit-specific manufacturer FDD protocols indicate fewer problems diagnosing refrigerant charge faults when no other faults are present due to wider tolerances and multi-step procedures (Mowris et al. 2013).12 Nevertheless, both types of protocols have limitations and neither can distinguish non-condensables and restrictions from refrigerant charge faults, condenser or evaporator heat transfer faults, low airflow, or expansion valve failure. These findings are significant and require additional testing. In particular, the evaluation team has not examined new system designs using micro-channel heat exchangers (MCHE). Manufacturers claim MCHE systems may be more sensitive to incorrect charge, non-condensables, restrictions, or coil blockage than

8 These systems were chosen based on what is most commonly seen in the market, availability of these discontinued but new off the shelf units, and on the fact that most maintenance activities are still performed on R-22 systems. 9 Test results will be made available in the HVAC Impact Evaluation Report: WO32 HVAC.10 Yuill, D, Braun, J. 2012. Evaluating Fault Detection and Diagnostics Protocols Applied to Air-Cooled Vapor Compression Air-Conditioners, International Refrigeration and Air Conditioning Conference. http://docs.lib.purdue.edu/iracc/1307.11 Bruan, J. Yuill, D. 2014. Evaluation of the effectiveness of currently utilized diagnostic protocols. Ray W. Herrick Laboratories Purdue University. Prepared for Portland Energy Conservation, Inc. “False alarm” is defined as diagnosis of a fault with the following: 1) fault intensity ratio (FIR) for capacity and COP are above 95% threshold, 2) refrigerant charge is less than 105% of “nominal,” and 3) superheat is between 1 and 36°F. “Nominal” is defined as charge yielding maximum COP at 95F outdoor and 80/67F indoor. The EM&V study does not adhere to this relative definition.12 Mowris, R., Eshom, R., Jones, E. 2013. Lessons Learned from Field Observations of Commercial Sector HVAC Technician Behavior and Laboratory Testing. IEPEC. http://www.iepec.org/conf-docs/conf-by-year/2013-Chicago/129.pdf#page=1

KEMA, Inc. 4 5/6/23

systems that use more conventional heat exchangers. MCHE systems use require 20- 40% less refrigerant and are about 10% more efficient than conventional tube and fin condensers.13

The DNV GL team performed most of the recent laboratory tests on commercial packaged systems with economizers in place since economizers are code-required for systems with a rated capacity exceeding 54,000 Btuh. These tests provided unexpected and important results. Economizer damper leakage is much greater and maximum economizer outside air rates much lower than assumed in past DEER evaluations14. The Air Movement and Control Association (AMCA) classifies four levels of damper leakage at static pressure of 1 inch water column (IWC): Class 1A) 3 cfm/ft2, Class 1) 4 cfm/ft2, Class 2A) 10 cfm/ft2 (ASHRAE 90.1), and Class 3) 40 cfm/ft2.15 Laboratory tests of Class 2 dampers on economizers from three different manufacturers of ASHRAE 90.1 compliant economizers indicate leakage rates about 6.5 times higher than the AMCA Class 1A standard (i.e., 60 to 80 cfm/ft2). DNV GL consistently measured closed economizer damper leakage rates at 13 to 30% of operating airflow or greater. Additionally, DNV GL measured maximum outside air rates during economizer operation between 30% and 75% of operating supply air rates. This is significantly different from the 5% closed damper flow and 100% fully open economizer flow assumed in past DEER analyses. These findings are significant and, when included in DEER analyses, will impact energy measures both positively and negatively. Measures with a cooling impact (such as a lighting retrofit), would likely see the related cooling benefits increase as the economizer free cooling would decrease with the reduced maximum flow rate. Direct economizer repair or controls measures will certainly see a predicted reduction in benefit as the new values are implemented. At this point, test data is available for four different economizers on four packaged units from three manufacturers (two 3-ton and two 7.5-ton units)16. Preliminary laboratory tests conducted on roof top units (RTUs) with economizers indicate that approximately 13 Carrier, 2007, Commercial documentation on microchannel heat exchangers, www.carrier.com. Cremaschi, 2007, HPC, 2007, Heat Pump Center, Newsletter #3, 2007. Yanik, M. Jianlong, J. 2012. Application of MCHE in Commercial Air Conditioners. Danfoss. 2013. How to Cut Costs and Impacts of Your AC and Refrigeration Systems.14 Past DEER/EQuest economizer outdoor air (OA) leakage assumptions were 5% for closed dampers and 100% for fully open dampers. Laboratory tests of class 2 dampers on economizers from three different manufacturers of ASHRAE 90.1 compliant economizers indicate leakage rates of 13 to 30% or about 4.8 to 11.1 times higher than the AMCA Class 1A standard. Laboratory tests of Class 1A dampers indicate 17% leakage or 6.3 times greater than Class 1A. Higher OA leakage rates are likely due to perimeter frame leakage not included in standard ACMA tests.15 AMCA. 2010. AMCA 511-10 (Rev. 8/12) Certified Ratings Program–Product Rating Manual for Air Control Devices. Table 3. pp. 13. www.amca.org.16 Test results will be made available in the HVAC Impact Evaluation Report: WO32 HVAC.

KEMA, Inc. 5 5/6/23

42 to 55% of closed damper flow is from the perimeter or economizer/unit connection joint. This junction can be cost effectively sealed with UL-approved metal tape to improve space cooling and heating efficiency. Taping around the economizer frame improved application sensible cooling efficiency (EER*) by 7 to 16% when the damper is closed or open from 10 to 30%. Additional tests are critical to expand and confirm the range of savings opportunities from commercial QM measures related to economizer operation and installation.

The ASHRAE 90.1 mechanical subcommittee investigated economizer damper leakage described as follows:17

“The damper leakage for outside air dampers is only an issue on units when they are running in the unoccupied mode for heating or cooling. That means it is not an issue on a 24/7 operation and is only an issue in the buildings that have unoccupied heating and cooling. In the occupied mode the dampers are open for minimum ventilation air so leakage is a non-issue. In the unoccupied mode the leakage is only an issue when the fan is on for heating or cooling, but the fan is cycled in most applications so when the fan is off there is no leakage.”

This statement is correct if dampers meet the AMCA 511 standard, no other leakage exists except damper edge and jamb leakage, and minimum damper position meets ASHRAE 62.1 outdoor air requirements. As noted above, for units tested in the laboratory, economizer outdoor air leakage appears to be much higher than previously assumed when dampers are closed or partially open in the minimum position. Preliminary laboratory tests have also shown that dampers only provide approximately 60 to 65% of outdoor air when fully open. Industry publications have identified two economizer leakage areas:

1) “Jamb leakage” between damper blade ends and frame, and

2) “Edge leakage” between damper-blade edges.18

Preliminary laboratory tests conducted under WO32 discovered a third economizer leakage area:

3) “Perimeter and Gap Leakage” between economizer perimeter frame and HVAC cabinet and holes or gaps in the economizer or damper assembly.

17 D. Lord. 2010. Simplified Damper Leakage. ASHRAE 90.1 Mechanical Subcommittee Presentation.18 J. Knapp. 2007. Damper Leakage Rates–More Important than Ever. AMCA International Inmotion. pp. 19-21 (Fall 2007). www.amca.org.

KEMA, Inc. 6 5/6/23

Low-leakage dampers are supplied with blade and jamb seals. The type of seal supplied causes significant differences in leakage rates. There can be a 10-to-1 difference in a damper supplied with mechanically locked seals and flexible metal jamb seals versus a damper supplied with no seals at all. Economizers with no perimeter seals can increase leakage by 50% or more when dampers are closed for either Class 1 or Class 1A dampers.19 Preliminary laboratory tests indicate that damper leakage is much greater than previously assumed.

Field and laboratory tests indicate airflow is lower than manufacturer data indicates. Some manufacturers indicate that airflow below 400 CFM/ton can cause FDD false alarms, misdetection, or misdiagnosis. Some manufacturers recommend 400 CFM/ton when checking refrigerant charge and some recommend a range of 350 to 450 CFM/ton. Airflow is rarely, if ever, measured by technicians before checking refrigerant charge diagnostics. Airflow and external static pressure (ESP) also impact damper, cabinet, and perimeter leakage which impacts cooling and heating capacity, efficiency and run time. Laboratory tests will be conducted to evaluate these issues.

While noted as a laboratory research plan, this research plan will coordinate closely with other HVAC field work pilots (such as QM). Field data is necessary to develop reasonable operating conditions for laboratory tests. Additionally, some issues, like fouled condenser coils, can’t be readily reproduced in the laboratory. Evaluation of protocols used for charge correction requires accurate measurement of the “correctness” of the diagnosis and applicability of the protocol in the field. Measurement errors inherent to even the most careful field measurements can result in the requirement that a fault must have an efficiency impact that is much larger so that it rises above the uncertainty of the measurement. A further consideration is that instruments commonly used by technicians result in even greater uncertainty. The applicability of generic FDD protocols based on simple superheat or subcooling target values can introduce additional uncertainty especially if the generic protocol is inconsistent with unit-specific manufacturer protocols that involve multiple and different parameters.

DNV GL proposes the following laboratory and field tests for the 2013-14 cycle.

Complete any tests identified in the 10-12 plans or during the 10-12 cycle that are not included in the current scope during the 13-14 cycle20.

19 Ibid.20 Laboratory testing has inherent uncertainty, while both timing and budget are limited for each evaluation cycle. During the 10-12 cycle, the testing staff experienced unexpected complications and fundamental learning regarding lab emulation of field conditions. Additionally, a lack of real protocol to follow for many of the tests meant is was necessary

KEMA, Inc. 7 5/6/23

Conduct laboratory tests at AHRI-certified facilities per AHRI, ASHRAE and/or ANSI standards. The proposed laboratory test facilities include Intertek in Plano, TX and UL Laboratories in Plano, TX.

Evaluate other test facilities including PG&E Applied Technology Services in San Ramon, CA; SCE Technology Test Centers in Irwindale, CA, and UC Davis Western Cooling Efficiency Center in Davis, CA. DNV GL may plan or coordinate tests at other laboratories depending on availability and capability.

Perform field tests to identify and evaluate QM and QI measures that cannot be fully evaluated or tested in the laboratory. For example, field-tests are used to evaluate refrigerant diagnostic protocols by recovering and accurately weighing-out refrigerant, evacuating to 500 microns for 20 minutes, and weighing-in factory charge (i.e., weigh-out/in method). The weigh-out/in method provides a baseline for evaluating FDD protocols and is recommended by some manufacturers.21 Field tests indicate that it is difficult to correctly diagnose RCA faults from false alarms, misdetections, or misdiagnoses even when no other faults are present. Diagnosing RCA faults is difficult under typical field conditions when multiple faults are present such as low airflow, blocked/dirty condenser/evaporator, restrictions, non-condensables, or expansion valve failure.

Laboratory and field tests from the 2010-12 cycle demonstrated that the weigh-out/in method might be a fundamental requirement to achieve an ACCA 180 “performance baseline.” In this cycle, DNV GL will conduct field and laboratory tests to evaluate repeatability, accuracy, timing, and applicability of the weigh-out/in method. DNV GL may also evaluate on-board FDD using the weigh-out/in method. Field tests in the 2010-12 cycle demonstrated the validity of laboratory tests of low airflow, uninsulated TXV sensing bulbs, restrictions, non-condensables, and condenser coil blockage. DNV GL plans on additional field and laboratory tests of these faults in the 2013-14 cycle.

Perform laboratory tests with multiple faults similar to field conditions to evaluate FDD protocols with respect to false alarms, misdetections, or misdiagnoses of faults. Multiple faults will include low airflow (not corresponding to manufacturer factory settings regarding external static pressure and cfm), high static pressure, fan speed or horsepower, condenser

to develop these testing protocols extemporaneously. These elements resulted in tests taking longer than originally expected and therefore not all planned 10-12 tests were completed under that cycle.21 Lennox Industries, Inc. 2006. Lennox Service Literature Unit Information. http://tech.lennoxintl.com/C03e7o14l/xddfVVShbC/9901i.pdf http://tech.lennoxintl.com/C03e7o14l/xddfVVShbC/1008c_CORP1008L2_003.pdf

KEMA, Inc. 8 5/6/23

and evaporator coil blockage, incorrect charge, economizer or mixed air dampers open, and improperly insulated TXV sensing bulbs. DNV GL will use this information to improve the field and laboratory study.

Perform low, medium, and high airflow tests and low and high static tests for belt-driven systems including low rpm; loose fan belts (1 inch deflection per 9 lbs. force versus 0.4 inches deflection for 9 lbs. force); misalignment by ⅛, ¼ and ½ inches; cracked belt (1/8 inch deep, several per inch); and dirty air filters. Air filter pressure drop in light commercial buildings ranges from 0.14 IWC (34 Pa) for MERV 2 to 0.22 IWC (55 Pa) for MERV 11.22 Dirty filters with medium to high loading can increase air filter pressure drop by 0.12 to 0.16 IWC.23

Perform low, medium, and high airflow tests for direct-drive systems, including low or medium “speed” (horsepower) blower-motor settings and dirty air filters. Tests of variable frequency drive (VFD) blower fans are not currently included in the research plan.

1.2 Quality Maintenance and Laboratory Work Synergies

DNV GL will perform field- and laboratory-based measurements to assess savings potential from HVAC installation, retrofit, tune-up and maintenance measures. DNV GL will develop a list of measures that are currently being implemented as well as identify candidates for future implementation. These must be realistic and economic to implement in an IOU managed program. The HVAC Project Coordination Group (PCG) and Western HVAC Performance Alliance (WHPA) will review and make recommendations concerning the measure list, thereby providing stakeholder input. DNV GL expects that for some measures, multiple implementation methods are possible; DNV GL will assure the testing covers all current methods.

DNV GL will use laboratory and field testing to establish baselines and savings potential for various measures. The following is a list of variables DNV GL will test in the laboratory and in the field:

Service instrument evaluation (type, placement, accuracy)

22 Stephens, B., Siegel, J.A., and Novoselac, A. 2010a. Energy Implications of Filtration in Residential and Light-Commercial Buildings. ASHRAE Trans. 116. Pt. 1:346-357. http://www.caee.utexas.edu/prof/novoselac/Publications/Novoselac_ASHRAE_Transactions_2010.pdf23 Walker, I.S., Dickerhoff, D. J., Faulkner, D., Turner, W. 2013. System Effects of High Efficiency Filters in Homes. LBNL-6144E. http://eetd.lbl.gov/sites/all/files/lbnl-6144e.pdf.

KEMA, Inc. 9 5/6/23

Refrigerant charge (FDD, manufacturer/CEC/hybrid protocols) Restrictions, charge faults (FDD, liquid line driers, expansion devices, etc.) Non-condensables, charge faults, blocked condenser (FDD) Provide additional cycling tests of EER-rated units (greater the 65,000 Btu/h

net capacity) to evaluate cycling loss assumptions in the IEER rating calculation

Measuring airflow (methods, accuracy, diagnostics) Determine optimal airflow for diagnostics and efficiency (fan speed, ESP,

CFM/ton) Condenser coil diagnostics, blockage, efficacy of different cleaning methods,

and airflow Measure liquid and discharge pressure at different outdoor temperature

conditions of 55F, 70F, 82F, 95F, and 115F to evaluate how to estimate liquid pressure (and condenser saturation temperature) when only discharge pressure service valves are available

Economizer perimeter leakage Economizer damper position to identify the relationship between outdoor

airflow and damper position in terms of percentage of outdoor air versus geometric position of the damper as determined by control adjustment and low- versus high-speed fan operation24

Standard, Class 1 and Class 1A economizer dampers Economizer sensors (type, placement, etc.) TXV sensing bulb placement/insulation Notched V-belts (alignment, tension, etc.) Charge recovery/evacuation/factory recharge (methods, repeatability) Begin evaluating on-board FDD regarding accuracy, false alarms,

misdetection, or misdiagnoses based on time and budget considerations Economizer control strategies (mixed air temperature, outdoor air

temperature, outdoor minus return air temperature difference)

The DNV GL team will use field testing to establish the variation of typical “as found” conditions. The team will explore the usability of existing data from current programs as a means to increase potential precision of the as-found conditions determined from the evaluation sample. DNV GL will field- and laboratory-test both manufacturer-specific and generic refrigerant charge diagnostic protocols for accuracy (i.e., false alarms, misdetections, and misdiagnoses).

24 Some economizer manufacturers have a default of 3.2V (20% open) for low speed fan (heating mode) and 2.8V (10% open) for high speed (cooling mode). Minimum and maximum ventilation settings will impact percentage outdoor air and operational time and energy required for the system to meet thermostat settings.

KEMA, Inc. 10 5/6/23

Based on the expected variations in effectiveness of different measures by different implementers, the evaluation will provide a basis for the ex-ante values for various measures implemented. The expected value statistical method used to estimate average energy efficiency measure unit savings in workpapers is not included in the EM&V research plans. Building energy simulations using inputs based on laboratory test data may be used to estimate DEER ex ante energy and peak demand savings.

1.2.1 Laboratory Based HVAC Unit Testing

DNV GL will:

Test energy efficiency ratings under ANSI/AHRI 210/240 and ANSI/AHRI 340/360.

Test at in-situ conditions with typical return air temperature conditions, external standard pressure, airflow, and economizer damper position to evaluate differences in performance for standard and high efficiency packaged RTUs. In-situ test conditions will be as follows:

o Outdoor temperature (DB/WB): 82/62, 95/75, 115/8025

o Indoor temperature (DB/WB): 70/57, 75/62, 80/67,o Economizer outdoor temperature tests (DB/WB): 70/60, 65/57, 60/54,

55/5126

o Airflow (CFM/ton): 250, 300, 350, 400, 425 cfm/ton27

o External static pressure (IWC): 0.15 to 2.028

DNV GL will complete specific economizer related tests identified in the previous testing cycle:

25 Outdoor wetbulb temperatures are defined in the tests to measure the impact of economizer outdoor air leakage on total cooling capacity.26 Economizer outdoor temperature test conditions are selected to measure system efficiency (EER) and cooling capacity without compressor operation and with 1st-stage and 2nd-stage operation (for multi-compressor systems). The tests are performed to evaluate change over settings and performance based on outdoor air provided by unit-specific economizers. Climate zone performance needs to be evaluated with building energy simulation software using realistic outdoor air fractions based on the laboratory tests.27 Airflow target is in the test matrix. Variations in airflow will occur due to limitations of blower-drive system, motor, and external static pressure of laboratory setup.28 Specific external static pressure for each test will vary depending on speed (rpm), airflow (cfm), and horsepower of the blower-drive system being tested. Test conditions will be based on field data from WO32 and field data available in the “Small HVAC Problems and Potential Savings Reports,” October 2003, California Energy Commission 500-03-082-A-25. http://www.energy.ca.gov/2003publications/CEC-500-2003-082/CEC-500-2003-082-A-25.PDF.

KEMA, Inc. 11 5/6/23

Continue economizer dampers tests to determine their minimum and maximum outside air rates at typical damper positions.29

Test temperature sensors and controls in economizers from different manufacturers to evaluate sensor placement, design, and configuration. Some sensors are manufactured inside plastic boxes and located on the side, top or bottom of the economizer or fan housing. Some are wand-style that can be placed in the airstream. The team will test different types of sensors and controls for accuracy and compatibility issues.

Test analog and digital motor actuator compatibility and controls in the field and in the laboratory. Compatibility is an issue for IOU programs that involve retrofitting automated digital economizer controls on existing economizers with analog sensors and motors.

Test airflow (fan speed, horsepower) and external static pressure impacts on economizer dampers and perimeter leakage, efficiency, and capacity.

Test airflow and economizer outdoor air leakage impacts on FDD false alarms, misdetections, and misdiagnoses.

1.2.2 Laboratory Based Field Instrument Testing

In the laboratory, DNV GL will:

Evaluate impact of instrument placement, design and configuration. Soak temperature sensors in the outdoor chamber to reach equilibrium

before tests begin and determine how to attach (and insulate if possible) sensors and specific locations to place sensors on each tube during tests.

Place manifold pressure sensors with refrigerant hoses inside an oven at 130 °F to simulate typical field-service conditions. Additionally, the team will test pressure sensors at specific low, medium, and high pressures for R-22 and R-410A refrigerants in a repeatable manner using a test bench.

Measure airflow using standard airflow traverse methods, flow-capture hoods, or pitot-tube arrays and compared to the Code Tester measurements.30 The laboratory previously tested four (4) pitot-tube arrays with digital pressure gauges. Measured accuracy was +9 to +11% of

29 Leakage Classifications per AMCA Publication 511-10 (Rev. 8/12) Certified Ratings Program - Product Rating Manual for Air Control Devices, Air Movement and Control Association International, Inc.30 Banks, E. Sills, C. Graves, C. 2002. Airflow Traverse Comparisons Using the Equal-Area Method, Log-Tchebycheff Method, and the Log-Linear Method, and Including Traverse Location Qualification. http://www.orau.org/ptp/PTP%20Library/library/Subject/stack%20sampling/airflow_traverse.pdf. The “code tester” is the airflow measuring apparatus described in Section 5.3 Test Chambers (Code Testers), ANSI/ASHRAE 41.2-1987 (RA92).

KEMA, Inc. 12 5/6/23

laboratory “code-tester” airflow measurements.31 The DNV GL team will perform tests with a digital balometer flow capture hood with “Equal-area” method and integrated Log-Tchebycheff method duct traverse (range 25 to 2500 cfm with +/-3% accuracy). The team will perform tests on ten (10) anemometers from four manufacturers. The traverse for the 10" wide x 16" deep return duct will obtain nine (9) equally spaced measurements described below (draft). The wand must be perpendicular to flow (to do this, the team will turn the wand until maximum measurement is obtained). The team will measure and store each value. Test procedure is as follows:

o 1) Insert anemometer straight into hole at 4", 8" and 12." o 2) Insert anemometer at 17.4 degrees and 8.4" into hole to obtain left

and right midpoint measurements. o 3) Insert anemometer at 32 degrees and 4.7" into hole to obtain lower

left and lower right measurements. o 4) Insert anemometer at 11.8 degrees and 12.3" into hole to obtain

upper left and upper right measurement. o This should provide an average measurement close to the “equal-

area” method. Check fan belt tension and alignment tools with laser guided equipment.

DNV GL will evaluate different HVAC manufacturer methods for checking belt tension and alignment. DNV GL will use test results to evaluate how belt tension and alignment impact airflow, cooling capacity, efficiency and FDD.

Compare static pressure to laboratory-grade pressure array measurements to evaluate manufacturer performance data regarding external static pressure versus airflow and fan speed.

Compare field power measurement instruments to laboratory-grade power measurements.

The test equipment schematic for a single-compressor packaged unit is shown in Figure 1. Refrigerant-side pressure/temperature measurements are installed before the expansion device, evaporator outlet, compressor suction, compressor discharge, and condenser outlet Setup requires precision louvered dampers installed on supply and return ducts to control inlet static pressure (ISP) and external static pressure (ESP) similar to in-situ conditions. Controlling inlet and total static pressure provides realistic test conditions to measure performance 31 The “code tester” is the airflow measuring apparatus described in Section 5.3 Test Chambers (Code Testers), ANSI/ASHRAE 41.2-1987 (RA92). Pitot-tube array measurements had a 200 cfm offset which could be corrected with additional testing. Additional tests of the Pitot-tube array on other RTUs need to be performed to determine if measurements are always high.

KEMA, Inc. 13 5/6/23

when varying airflow, fan speed, and economizer outdoor-air damper positions from closed to fully open.

Figure 1: Test Equipment Schematic

KEMA, Inc. 14 5/6/23

2. Objectives

2.1 Key Research Questions

The following is a partial list of research questions.

1. Can laboratory and field measurement of commercial packaged HVAC maintenance actions provide information to improve field FDD procedures and to help inform the evaluation of load impacts and improve subsequent QM program designs?

2. What is the accuracy of technician field instrument measurement tools in terms of measuring refrigerant temperature and pressure, airflow temperature and pressure, airflow, and economizer operation? How is accuracy influenced by application and test conditions? How is accuracy influenced over time?

3. How do generic FDD systems (i.e., CEC RCA, etc.) compare to unit-specific manufacturer FDD in terms of false alarms, misdetections, and misdiagnoses? How does FDD compare to recovery, evacuation, and weigh-in of factory charge?

4. Technicians have been observed introducing non-condensables into systems due to not using EPA 608 low-loss fittings and not purging lines of non-condensables prior to attaching hoses to refrigerant systems. How can laboratory tests quantify the impact of introducing non-condensables when conducting maintenance services? How many occurrences of contamination does it take to impact efficiency?

5. What is the range of performance and fault impacts of packaged cooling and heating systems commonly used in commercial applications32? Faults include outdoor damper and perimeter leakage, high static pressure, low airflow, fan speed, condenser/evaporator coil blockage, expansion device (TXV and non-TXV), refrigerant over/under charge, restrictions, non-condensables, economizers and indoor temperature settings.

32 It is not expected that this round of testing will be able to definitively answer this research question; this is one element of the multi-year HVAC lab research effort and represents an important aspect of our longer term goal.

KEMA, Inc. 1 5/6/23

6. Are the impacts of refrigerant faults from older, field recovered units the same as those from new units used in previous tests?

7. Tests conducted to answer these questions will be used to better quantify the impacts of coil cleaning which cannot be accurately replicated in the laboratory, and to determine if unit age and exposure to elements impacts energy efficiency. How long do condenser coil cleaning savings last and what does the degradation curve look like? Is the curve steep or flat? Does it vary depending on building, business type and/or unit? What are the impacts of frequency and method of cleaning? Do field and laboratory measurements indicate that dirty condenser coils impact RCA diagnostics (false alarms and misdiagnoses)? If so, does recovery and weighing-in charge eliminate this problem? Coil cleaning will be tested in the laboratory at high to low side pressure differences mapping condenser coil blocking at 30, 50, and 80%.

8. Commercial Quality Maintenance? If necessary, how easy is it for technicians to measure and adjust airflow to properly diagnose RCA?

9. How does unintended economizer damper leakage (jamb, edge, perimeter/gap, improper minimum position, stuck open, etc.) impact efficiency, outdoor airflow, and economizer savings (i.e., outdoor airflow between minimum and fully open position)?

10.How does economizer setup impact efficiency regarding manually setting minimum damper position based on rules-of-thumb (1-finger, 2-finger, etc.) versus advanced digital economizer control setting minimum damper position based on percent open (i.e., 10%, 15%, 20%, etc.)?

11.How does economizer temperature sensor location and default settings impact economizer operation and energy savings?

12.How does total static pressure, outdoor air damper position, fan speed, blower pulley diameter, and horsepower impact cooling capacity and energy efficiency?

2.2 Parameters Evaluated by Measure (Impact)

Commercial Quality Maintenance Gross Savings are an Efficiency Savings Performance Incentive (ESPI) Measure for 2013-14.

KEMA, Inc. 2 5/6/23

KEMA, Inc. 3 5/6/23

3. MethodsAny testing identified in the 10-12 plans or during the 10-12 cycle that is not included in the current scope will be addressed during the 13-14 cycle. Laboratory testing will be done at a variety of conditions to establish performance encountered in the field. These conditions include economizer presence, return and outdoor air conditions appropriate to California climate conditions, and introducing faults deemed to be typical. The HVAC PCG and the WHPA will review and make recommendations concerning the list thereby providing stakeholder input on what is “typical.”

Testing will be done in two phases. Year 1 will conduct laboratory work planned but not completed in the 10-12 cycle and Year 2 will reserve funds for new tests based on new field findings. The close coordination of laboratory and field testing makes planning details of laboratory testing in advance essentially impossible. General laboratory plans will be developed as part of the initial planning activities, but the planned testing will almost certainly need to respond to interim laboratory and field results.

3.1 Year 1 Laboratory Testing Priorities

The initial work order will contain a few primary tasks including the following:

Develop laboratory testing plan Oversee laboratory tests according to testing plan QC and report initial laboratory test data Analyze and report laboratory test data Support CPUC Consultants’ Collaboration of IOU HVAC laboratory testing Conduct testing in general accordance with the tables below

Table 1: Tests for Manufacturer #1 7.5-ton R-22 non-TXV, 2-Circuit (2 compressor) Unit (RTU3)

Test Type Status Shifts Tests Budget AHRI Verification No Economizer Vertical Finished Finished Finished Finished Economizer Damper (closed, 1-finger, 2-finger, 3-finger, 100% open) Vertical Finished Finished Finished Finished Economizer 55 to 70F OAT, None-1-2 comps Vertical Finished Finished Finished Finished Airflow 100%, 83.3%, 67.7% of 400 cfm/ton Vertical Finished Finished Finished Finished

KEMA, Inc. 1 5/6/23

Test Type Status Shifts Tests Budget Refrigerant Charge -40 to +40% (in 20% intervals) of factory charge Vertical Finished Finished Finished Finished Laboratory Test Setup Vertical (new tests) Vertical Pending 2.00 NA $6,204

Airflow 100%, 83.3%, 67.7% of 400 cfm/ton (supply/return dampers control ISP & ESP) Vertical Pending 4.00 36 $9,288

Refrigerant Charge -40 to +40% (in 20% intervals) of factory charge (supply/return dampers control ISP & ESP) Vertical Pending 5.00 54 $11,610 Mixed Measures: 80%, 100%, 120%, 140%, 160% of factory charge with 67%, 83%, 100% airflow, with dampers closed and 1-finger open Vertical Finished Finished Finished Finished Measurement Instruments (remainder to be tested on horizontal setup)) Vertical Pending Pending Pending Pending

AHRI Verification Horiz. Finished Finished Finished Finished Economizer Damper closed, 1F, 2F, 3F, open Horiz. Finished Finished Finished Finished

Economizer 55 to 70F OAT, No-1-2-compressors Horiz. Finished Finished Finished Finished

Airflow 100%, 83.3%, 67.7% of 400 cfm/ton Horiz. Finished Finished Finished Finished

Refrigerant Charge 40 to +40% (in 20% intervals) of factory charge Horiz. Finished Finished Finished Finished

Fan Speed (1-6 turns, 820-950 rpm) Horiz. Finished Finished Finished Finished Fan Speed (1-6 turns, 820-950 rpm) (supply/return dampers control ISP & ESP) Horiz. Pending

See Airflow

See Airflow

See Airflow

Restrictions ** Install Needle Valve Horiz. Finished Finished Finished Finished Non-Condensables Horiz. Finished Finished Finished Finished Laboratory Test Setup Horizontal Horiz. Pending 2.00 NA $6,204 Airflow 100%, 83.3%, 67.7% fan speed 515-1032 rpm, 3-6 turns 5-10” dia. pulleys (supply/return dampers control ISP & ESP) Horiz. Pending 4.00 36 $9,288

Economizer Damper Leakage Tests with and w/o perimeter tape (C, 1, 2, 3, O) at 55F OAT and no compressors (control ISP & ESP) Horiz. Pending 1.50 10 $3,483

Economizer Damper with and without perimeter tape (C, 1, 2, 3, O) (supply/return dampers control ISP & ESP) Horiz. Pending 1.50 10 $3,483

Economizer Efficiency Tests 55 to 70F OAT, Damper C, 1, 2, 3, and open with none and open with 1-2-compressors (supply/return dampers control ISP &

Horiz. Pending 2.50 28 $5,805

KEMA, Inc. 2 5/6/23

Test Type Status Shifts Tests BudgetESP)

Refrigerant Charge -40 to +40% (in 20% intervals) of factory charge at 82F, 95F and 115F OAT and 250, 300, 350, 400 cfm/ton (control ISP & ESP) Horiz. Pending 5.00 54 $11,610

Condenser Coil Blockage (50%, 80%) (supply/return dampers control ISP & ESP) Horiz. Pending 0.50 5 $1,161

Evaporator Coil Blockage (50%, 80%) (supply/return dampers control ISP & ESP) Horiz. Pending 0.50 5 $1,161

Restrictions (supply/return dampers control ISP & ESP) Multiple Fault Tests Horiz. Pending 1.50 8 $3,483

Non Condensables (supply/return dampers control ISP & ESP) Multiple Fault Tests Horiz. Pending 1.50 11 $3,483

Multiple Fault Tests low airflow, 2F-open damper, untaped economizer perimeter/gaps, 50% blocked condenser, -10% refrigerant charge (control ISP & ESP) Horiz. Partial 2.00 11 $4,644

Measurement Instruments Horiz. Pending TBD TBD TBDTotal 35.5 268 $80,907

KEMA, Inc. 3 5/6/23

Table 2: Tests for Manufacturer #2 7.5-ton TXV, 2-Circuit (2 compressor) Unit (RTU1)

Test Type Status Shifts Tests Budget AHRI Verification Horiz. Finished Finished Finished Finished AHRI Verification A, B, C, and D Horiz. Pending *** *** *** Laboratory Test Setup Horiz. Pending 2.00 NA $6,204 Airflow 106%, 100%, 87.5%, 75%, 62.5% of 400 cfm/ton (supply/return dampers control ISP & ESP) Horiz. Pending 4.00 36 $9,288

Airflow 106%, 100%, 87.5%, 75%, 62.5% of 400 cfm/ton (supply/return dampers control ISP & ESP) with PSC motor (EC Motor tests will be considered) Horiz. Pending 4.00 36 $9,288

Economizer Damper Leakage Tests with and without perimeter tape (C, 1, 2, 3, O) at 55F OAT and no compressors (supply/return dampers control ISP & ESP) Horiz. Pending 1.50 10 $3,483

Economizer Damper with and without perimeter tape (C, 1, 2, 3, O) (supply/return dampers control ISP & ESP) Horiz. Pending 1.50 10 $3,483

Economizer Efficiency Tests 55 to 70F OAT, Damper C, 1, 2, 3, and open with none and open with 1-2-compressors (supply/return dampers control ISP & ESP) Horiz. Pending 2.50 28 $5,805

Refrigerant Charge -40 to +40% (in 20% intervals) of factory charge at 82F, 95F and 115F OAT and 250, 300, 350, 400 cfm/ton (control ISP & ESP) Horiz. Pending 5.00 54 $11,610

Condenser Coil Blockage (30%, 50%, 80%) (supply/return dampers control ISP & ESP) Horiz. Pending 0.50 5 $1,161

Evaporator Coil Blockage (base, 30%, 50%, 80%, charge adjustment) (supply, return dampers to control ISP/ESP) Horiz. Pending 0.50 5 $1,161

Restrictions (supply/return dampers control ISP & ESP) Multiple Fault Tests Horiz. Pending 1.50 8 $3,483

Non-Condensables (supply/return dampers control ISP & ESP) Multiple Fault Tests Horiz. Pending 1.50 11 $3,483

Multiple Fault Tests (supply/return dampers control ISP & ESP) Horiz. Partial 2.00 11 $4,644

Measurement Instruments (easier cabinet access for instrument placement) Horiz. Pending TBD TBD TBD

Total 22.50 178$53,80

5

KEMA, Inc. 4 5/6/23

Table 3: Tests for Manufacturer #2 7.5-ton TXV 2-Circuit (2 compressor) Unit (RTU2)

Test Type Status Shifts Tests BudgetAHRI Verification 3HP fan (3-6 turns, 925-750 rpm)

Horiz. Finished Finished Finished Finished

KEMA, Inc. 5 5/6/23

Table 4: Tests for Manufacturer #3 3-ton TXV 1- Circuit (1 compressor) Unit (RTU433)

Test Type Status Shifts Tests BudgetAHRI Verification A, B, C, and D Horiz. Pending Pending Pending Pending

Laboratory Test Setup Horiz. Pending 1.00 NA $3,102 Airflow 106%, 100%, 87.5%, 75%, 62.5% of 400 cfm/ton (supply/return dampers control ISP & ESP)

Horiz. Pending 4.00 36 $9,288

Economizer Damper Leakage Tests with and without perimeter tape (C, 1, 2, 3, O) at 55F OAT and no compressors (supply/return dampers control ISP & ESP) Economizer manufacturer #2

Horiz. Pending 1.50 10 $3,483

Airflow 106%, 100%, 87.5%, 75%, 62.5%of 400 cfm/ton (supply/return dampers control ISP & ESP) with PSC motor (EC Motor tests will be considered)

Horiz. Pending 4.00 36 $9,288

Economizer Damper with and without perimeter tape (C, 1, 2, 3, O) (supply/return dampers control ISP & ESP) Economizer manufacturer #2

Horiz. Pending 1.50 10 $3,483

Economizer Efficiency Tests 55 to 70F OAT, Damper C, 1, 2, 3, and open with none (supply/return dampers control ISP & ESP)Economizer manufacturer #1 and #2

Horiz. Pending 2.00 24 $4,644

Refrigerant Charge -40 to +40% (in 20% intervals) of factory charge at 82F, 95F and 115F OAT and 250 to 450 cfm/ton (supply/return dampers control ISP & ESP)

Horiz. Pending 5.00 54 $11,610

Condenser Coil Blockage (Base 30%, 50%, 80% of inlet area, charge adjustment) (supply/return dampers control ISP & ESP)

Horiz. Pending 0.50 5 $1,161

Evaporator Coil Blockage (50%, 80%) (supply/return dampers control ISP & ESP)

Horiz. Pending 0.50 5 $1,161

Restrictions and Multiple Fault Tests(supply/return dampers control ISP & ESP)

Horiz. Pending 1.50 8 $3,483

Non-Condensables and Multiple Fault Tests(supply/return dampers to control ISP/ESP)

Horiz. Pending 1.50 11 $3,483

33 The plan is to test RTU4 in the horizontal configuration only. If OEM manual provides notations on the difference between Horizontal and Vertical configurations, these will be discussed in the reporting.

KEMA, Inc. 6 5/6/23

Test Type Status Shifts Tests Budget Multiple Fault Tests (supply/return dampers control ISP & ESP)

Horiz. Partial 2.00 11 $4,644

Measurement Instruments Horiz. Pending TBD TBD TBDTotal 21.00 174 $49,542

KEMA, Inc. 7 5/6/23

Table 5: Tests for Manufacturer #1 3-ton non-TXV 1-Circuit (1 compressor) Unit (RTU5)

Test Type Status Shifts Tests Budget AHRI Verification A, B, C, and D Horiz. Pending Pending Pending Pending Laboratory Test Setup Horiz. Pending 1.00 NA $3,102 Airflow 106%, 100%, 87.5%, 75%, 62.5% of 400 CFM/ton fan speed 530-1100 rpm, 3-6 turns 5-7” dia. pulleys (supply/return dampers control ISP & ESP) Horiz. Pending 4.00 36 $9,288

Economizer Damper Leakage Tests with and without perimeter tape (C, 1, 2, 3, O) at 55F OAT and no compressors (supply/return dampers control ISP & ESP) Horiz. Pending 1.50 10 $3,483

Airflow 106%, 100%, 87.5%, 75%, 62.5% of 400 CFM/tonfan speed 530-1100 rpm, 3-6 turns 5-7” dia. pulleys (supply/return dampers control ISP & ESP) with PSC motor (EC Motor tests will be considered) Horiz. Pending 4.00 36 $9,288

Economizer Damper with and without perimeter tape (C, 1, 2, 3, O) (supply/return dampers control ISP & ESP) Horiz. Pending 1.50 10 $3,483

Economizer Efficiency Tests 55 to 70F OAT, Damper C, 1, 2, 3, and open with none (supply/return dampers control ISP & ESP) Horiz. Pending 2.00 24 $4,644

Refrigerant Charge -40 to +40% (in 20% intervals) of factory charge at 82F, 95F and 115F OAT and 250 to 450 cfm/ton (supply/return dampers control ISP & ESP) Horiz. Pending 5.00 54 $11,610

Condenser Coil Blockage (50%, 80%) (supply/return dampers control ISP & ESP) Horiz. Pending 0.50 5 $1,161

Evaporator Coil Blockage (50%, 80%) (supply/return dampers control ISP & ESP) Horiz. Pending 0.50 5 $1,161

Restrictions and Multiple Fault Tests (supply/return dampers control ISP & ESP) Horiz. Pending 1.50 8 $3,483

Non-Condensables and Multiple Fault Tests(supply/return dampers control ISP & ESP) Horiz. Pending 1.50 11 $3,483

Multiple Fault Tests (supply/return Horiz. Partial 2.00 11 $4,644

KEMA, Inc. 8 5/6/23

Test Type Status Shifts Tests Budget

dampers control ISP & ESP) Measurement Instruments Horiz. Pending TBD TBD TBD

Total 21.00 174$49,54

2

Table 6: Tests for Manufacturer #1 3-ton R-410A non-TXV 1-Circuit (1 compressor) Unit (RTU6)

Test Type Status Shifts Tests BudgetAHRI Verification Horiz. Pending Pending Pending PendingSame as above with supply/return dampers to control ISP/ESP Horiz. Pending Pending Pending Pending

Table 7: Tests for Manufacturer #4 3-ton R-410A TXV 1-Circuit (1 compressor) Test of BPM MotorUnit (RTU7)

Test Type Status Shifts Tests BudgetAHRI Verification Horiz. Pending Pending Pending PendingSame as above with supply/return dampers to control ISP/ESP Horiz. Pending Pending Pending Pending

KEMA, Inc. 9 5/6/23

4. Work Order Task Descriptions and Budget

4.1 Year 1 ($700,000)

The budget for year one is split into four tasks across two firms.

Table 8: Year 1 Summary BudgetTask Task Name DNV GL RMA Expense Total

1 Develop Laboratory Testing Plan $45,000 $50,000 $0 $95,000

2HVAC Laboratory Testing—Continue 2010-12 Testing

$5,000 $160,000 $40,000 $205,000

3

Support CPUC Consultants’ Collaboration of IOU HVAC Laboratory Testing

$30,000 $70,000 $0 $100,000

4 Lab Testing Facility Pre-payments $300,000 $300,000

Year 1 Total $700,000

4.1.1 Task 1: Develop HVAC Laboratory Testing Plan ($95,000)

Under the direct supervision of the Prime Contractor Project Manager, staff will support the Prime Contractor Project Management Team with evaluation planning, general start-up activities and administration including, but not limited to:

Attending project kick-off meeting with ED staff and other ED contractors. Developing detailed HVAC evaluation and laboratory test plans.

KEMA, Inc. 1 5/6/23

Performing tasks involved in the development and management of the research plan and future work orders, including review and comment by the IOUs.

The Task 1 Project Start-up Budget will be divided as shown in Table .

Table 9: Task 1 BudgetRecipient Budget Amount

DNV GL $45,000Robert Mowris & Associates $50,000Total $95,000

4.1.2 Task 2: Continue 2010-12 Laboratory HVAC Testing ($205,000)

Under the direct supervision of the Prime Contractor Project Manager, staff will conduct HVAC Laboratory Testing to assess savings potential from HVAC retrofit, tune-up and maintenance measures:

Develop a list of measures currently being implemented as well as candidate measures for future implementation that are realistic and economic to implement in an IOU managed program.

Use laboratory testing of HVAC equipment and diagnostic measurement tools and instruments to establish energy savings potential for various HVAC maintenance and installation measures when implemented.

Apply field testing results to establish the variation of typical “as found” conditions.

Provide a basis for ex ante values for various qualities of work and measures implemented.

Analyze and report laboratory test data. Give a webinar that is recorded to present and discuss the results.

Development laboratory testing plan and conduct testing in general accordance with the tables in section 3.1.

The Task 2 Project Start-up Budget will be divided as shown in Table 8.

Table 8: Task 2 BudgetRecipient Budget Amount

DNV GL $5,000Robert Mowris & Associates, Inc. $160,000Equipment and Instrumentation $40,000Total $205,000

KEMA, Inc. 2 5/6/23

4.1.3 Task 3: Report Laboratory Test Results and Support Collaboration with IOU HVAC Laboratory Testing ($100,000)

DNV GL will present interim and final results in a report for the CPUC and other stake holders. The team will also summarize past laboratory testing and results into a single document.

At the direction of CPUC staff and their advisors, DNV GL staff will observe HVAC laboratory testing at IOU test facilities.

Work cooperatively with IOU personnel who are conducting HVAC laboratory testing to assess savings potential from HVAC installation, retrofit, tune-up, and maintenance measures.

Work cooperatively with IOU personnel to develop a list of potential measures currently being implemented as well as candidates for future implementation that are realistic and economic to implement in an IOU managed program.

Evaluate the basis for ex ante values for various qualities of work and measures implemented.

Provide suggestions and other recommendations, as needed, during collaborative sessions.

Analyze and report on IOU laboratory test data.

The Task 3 Project Start-up Budget will be divided as follows in Table 9.

Table 9: Task 3 BudgetRecipient Budget Amount

DNV GL $30,000Robert Mowris & Associates $70,000Total $100,000

4.1.4 Task 4: Lab Facility Pre-Payments ($300,000)

This task will cover costs associated with further lab testing. DNV GL will continue using Intertek Testing Services NA, Inc. as a subcontractor for HVAC lab testing. Intertek requires an overall upfront payment of $265,095 to reserve lab testing space. An additional up-front cost will be the purchase and shipment to the testing site of suitable HVAC equipment. Under the direct supervision of the Prime Contractor Project Manager, Robert Mowris and Associates (RMA) will locate and

KEMA, Inc. 3 5/6/23

purchase the equipment required. After testing, RMA will maintain ownership and be responsible for equipment disposition. Intertek will store the tested equipment for some time (at no cost) should questions arise from the testing data analysis that require supplemental testing.

The Task 4 Project Start-up Budget will be divided as follows in Table 10.

KEMA, Inc. 4 5/6/23

Table 10: Task 4 BudgetRecipient Budget Amount

DNV GL $300,000Robert Mowris & Associates $0Total $300,000

4.2 Year 2 ($300,000)

Residential HVAC systems and heating tests are expected in year 2 along with currently undefined testing based on new field findings or new measures.

KEMA, Inc. 5 5/6/23

SAFER, SMARTER, GREENER

THIS IS DNV GLDriven by our purpose of safeguarding life, property and the environment, DNV GL enables organizations to advance the safety and sustainability of their business. We provide classification and technical assurance along with software and independent expert advisory services to the maritime, oil & gas and energy industries. We also provide certification services to customers across a wide range of industries.

Combining leading technical and operational expertise, risk methodology and in-depth industry knowledge, we empower our customers’ decisions and actions with trust and confidence. As a company, we continuously invest in research and collaborative innovation to provide customers and society with operational and technological foresight. With our origins stretching back to 1864, our reach today is global. Operating in more than 100 countries, our 16,000 professionals are dedicated to helping customers make the world safer, smarter and greener.

In the Energy industryDNV GL delivers world-renowned testing and advisory services to the energy value chain including renewables and energy efficiency. Our expertise spans onshore and offshore wind power, solar, conventional generation, transmission and distribution, smart grids, and sustainable energy use, as well as energy markets and regulations. Our 3,000 energy experts support clients around the globe in delivering a safe, reliable, efficient, and sustainable energy supply.

For more information on DNV GL, visit www.dnvgl.com.