TECHNICAL ANALYSIS STUDY - Software Motor Company
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
TECHNICAL ANALYSIS STUDY High Efficiency Fan Motors Presented to: The Sygma Network 13019 SE Jennifer St. STE 404 Clackamas, OR 97015 Provided by: Analysis Contractor: 1033 SE Main St. Suite 1 Portland, OR 97214 (971) 544-7211 Project Number: PE14314 Report Date: 6/4/2018
DISCLAIMER In no event will Energy Trust of Oregon, Inc. or Energy 350 be liable for (i) the failure of the customer to achieve the estimated energy savings or any other estimated benefits included herein, or (ii) for any damages to customer's site, including but not limited to any incidental or consequential damages of any kind, in connection with this report or the installation of any identified energy efficiency measures. The intent of this energy analysis study is to estimate energy savings associated with recommended energy efficiency upgrades. This report is not intended to serve as a detailed engineering design document, any description of proposed improvements that may be diagrammatic in nature are for the purpose of documenting the basis of cost and savings estimates for potential energy efficiency measures only. Detailed design efforts may be required by participant in order to implement potential measures reviewed as part of this energy analysis. While the recommendations in this report have been reviewed for technical accuracy and are believed to be reasonably accurate, all findings listed are estimates only, as actual savings and incentives may vary based on final installed measures and costs, actual operating hours, energy rates and usage. Energy Trust of Oregon -The Sygma Network- Page 2 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
CONTACTS & PREPARATION SITE CONTACTS The following plant personnel assisted with this report: Eric Brown, Facilities Manager The Sygma Network 13019 SE Jennifer St. STE 404 Clackamas, OR 97015 Phone: (503) 545-3031 E-mail: ebrown@sygmanetwork.com ENERGY TRUST CONTACTS The Program Delivery Contractor (PDC) is: Kelson Redding Energy 350 1033 SE Main St., Suite 1 Portland, OR 97214 Phone: (503) 442-0656 E-mail: kredding@energy350.com The Allied Technical Assistance Contractor (ATAC) that prepared this report is: Phillip McNamara, P.E., C.E.M. Energy 350 1033 SE Main St., Suite 1 Portland, OR 97214 Phone: (503) 819-8997 E-mail: phillip@energy350.com Energy Trust of Oregon -The Sygma Network- Page 3 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
TABLE OF CONTENTS 1.0 EXECUTIVE SUMMARY ................................................................................................. 5 1.1 Introduction ................................................................................................................................... 5 1.2 EEM Summary.............................................................................................................................. 5 1.3 Economic Summary ...................................................................................................................... 7 1.4 Potential Additional Benefits ...................................................................................................... 10 1.5 Recommendations ....................................................................................................................... 10 1.6 Implementation Summary ........................................................................................................... 10 2.0 DETAILED DESCRIPTION OF PROPOSED EQUIPMENT AND OPERATION ........ 11 2.1 EEM 1 – High Efficiency Fan Motors ........................................................................................ 11 2.1.1 EEM 1 – Source of Energy Savings.................................................................................... 11 2.1.2 EEM 1 – Specific Equipment Recommendations ............................................................... 11 2.1.3 EEM 1 – Setpoints and Algorithms Recommended to Achieve Energy Performance ....... 11 2.2 EEM 2 – Condenser Cleaning ..................................................................................................... 12 2.2.1 EEM 2 – Source of Energy Savings.................................................................................... 12 2.2.2 EEM 2 – Specific Equipment Recommendations ............................................................... 12 2.2.3 EEM 2 – Setpoints and Algorithms Recommended to Achieve Energy Performance ....... 12 3.0 EEM COSTS...................................................................................................................... 13 4.0 BASELINE AND ANALYSIS OVERVIEW ................................................................... 14 4.1 Baseline Description ................................................................................................................... 14 4.2 Overview of Technical Approach ............................................................................................... 15 4.2.1 Data Logging ...................................................................................................................... 15 4.2.2 Baseline Analysis ................................................................................................................ 17 4.2.3 EEM Analysis ..................................................................................................................... 27 4.3 Key Assumptions ........................................................................................................................ 31 4.3.1 Key Assumptions for Baseline Analysis ............................................................................. 31 4.3.2 Key Assumptions for EEM Analysis .................................................................................. 31 4.4 Summary of EEM Analysis ........................................................................................................ 32 5.0 COMMISSIONING REQUIREMENTS ........................................................................... 33 5.1 Purpose of Commissioning ......................................................................................................... 33 5.2 Logistical Requirements and Customer Assistance .................................................................... 33 5.3 List of Settings to be Observed/Confirmed/Recorded ................................................................ 33 5.4 Performance Verification Plan and/or O&M Persistence Plan ................................................... 34 6.0 APPENDIX ........................................................................................................................ 35 Energy Trust of Oregon -The Sygma Network- Page 4 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
1.0 EXECUTIVE SUMMARY 1.1 INTRODUCTION The Sygma Network (Sygma) stores and distributes refrigerated and frozen foods to multiple restaurant chains. This Technical Analysis Study (TAS) focuses on the refrigeration systems serving Sygma’s Clackamas, OR site. The facility relies on multifarious air-cooled condensing units to condition one -20°F ice cream freezer, two large storage freezers at -10°F, two 30°F – 38°F coolers and a heavily utilized 40°F shipping and receiving dock. All conditioned areas (freezers, coolers, dock) are refrigerated 24 hours per day, 7 days per week for a total annual operation of 8,760 hours. The purpose of this TAS is twofold. One, high efficiency motor technology was tested in real world scenarios to determine performance. Software Motor Company (SMC) graciously donated the motors and labor to install and commission the high efficiency motors. Power data was metered for the original fan motors as well as the high efficiency fan motors. Since this is a no-cost measure, no incentives are available. The TAS also serves to quantify energy savings resulting from a condenser coil cleaning for all condensing units. This is considered an operations and maintenance (O&M) measure and is eligible for a bonus incentive offer; see Section 1.3 for details. 1.2 EEM SUMMARY EEM 1: High Efficiency Fan Motors The evaporator and condenser fan motors for freezer condensing unit circuit SC5-2 were retrofitted with high efficiency switched reluctance motors. Ex-ante and ex-post power metering was performed on the original fan motors and high efficiency motors, respectively. Energy savings are realized due to the higher efficiency of the motors. Although capable of variable speed operation, this TAS analyzed energy savings of the fan motors due to their efficiency operating at constant speed (full speed) using control mechanisms (cycling) identical to the baseline case. This was a no cost (no incentive) measure. Additional savings will result if motor speed is allowed to modulate. EEM 2: Condenser Cleaning Over time, debris has accumulated on the condenser coils of the air-cooled condensing units restricting air flow through the fins which negatively effects heat transfer effectiveness across the coils. This EEM recommends cleaning the condenser coils for all condensing units. Clean condenser coils will reduce the approach temperature resulting in compressor savings due to lower head pressure. Energy Trust of Oregon -The Sygma Network- Page 5 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
EEMS STUDIED BUT NOT RECOMMENDED EEM 3: Evaporator Cleaning This measure analyzed energy savings realized from cleaning three dock evaporator air units. The cost of the measure results in a high payback period that does not meet Energy Trust’s cost effectiveness criteria. Therefore, this measure is not recommended. EEM 4: Lower Minimum Condensing Pressure Reducing the minimum head pressure on the condensing units was originally considered. However, a combination of data logging and cut-in/cut-out pressure switch setting observations confirmed the condensing units were already operating with low minimum condensing pressure settings. Reducing the pressure further may preclude proper refrigerant feeding to thermal expansion and/or invoke refrigerant stacking in the condenser coils. Therefore, this measure is not recommended. Energy Trust of Oregon -The Sygma Network- Page 6 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
1.3 ECONOMIC SUMMARY Table 1: Estimated Savings and Cost Summary Electric Rate Schedule & Cost: PGE 85-S $0.0715 $/kWh Cost of Demand 5.63 $/kW/mo On-Peak Annual Electric Total Included in Demand Demand Electric Cost Annual Installed Pre-Incentive EEM Description Package? Reduction Savings ($) Savings Savings Savings Cost ($) Payback (kw/mo) (kWh/yr) ($) ($) 1 High Efficiency Fan Motors Yes 2.1 $139 26,336 $1,883 $1,883 $0 0.0 days 2 Condenser Cleaning Yes 2.0 $137 40,200 $2,874 $2,874 $4,424 1.5 years 3 Evaporator Cleaning No 0.4 $29 8,422 $602 $602 $3,219 5.3 years Totals 4.1 $276 66,536 $4,757 $4,757 $4,424 11.2 months Note: Pre-incentive payback = Installed cost/Total Annual Savings. Demand savings are not included in the payback calculation since the peak power of the system may not always coincide with the peak demand of the facility during each billing cycle. Energy Trust of Oregon -The Sygma Network- Page 7 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Table 2: Estimated Incentive Summary Eligible Project Cost Cap: 50% Capital Electric Savings Cap: $0.25/kWh (>1yr pre) | $0.02/kWh (
Energy Trust is providing an ongoing 90x90 O&M bonus incentive offering. The 90x90 O&M incentive is calculated based on $0.08/kWh saved up to 90% of the project cost. The standard O&M project incentive is calculated based on $0.08/kWh saved up to 50% of the project cost. The calculated incentive in Table 3 is based on the 90x90 O&M incentive offering. To qualify for the 90x90 O&M incentive, the customer needs to complete the recommended measures and provide final cost documentation to the PDC (Energy 350) within 90 days after Energy Trust signs an incentive agreement. In addition, the PDC must submit this final cost documentation and a verification report to Energy Trust soon after the measures have been implemented. If for some reason the project completion extends beyond the 90 days, then the incentive will be calculated based on the standard O&M incentive offering (Table 2). Table 3: Estimated Bonus Incentive Summary Incentive Cap, % of Project Cost 90% Energy Incentive Rate $0.08/kWh Total Customer Standard 90x90 Payback Incentive Cost after EEM Description Incentive Bonus with Total Including Total Offer Offer Incentives Bonus Incentive 2 Condenser Cleaning $2,212 $1,004 $3,216 $1,208 5 months Totals $2,212 $1,004 $3,216 $1,208 5 months Energy Trust of Oregon -The Sygma Network- Page 9 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
1.4 POTENTIAL ADDITIONAL BENEFITS The recommended efficiency measures have benefits beyond saving energy. Cleaning the condensers, effectively reducing the head pressure, will reduce equipment wear since the units will operate at a reduced duty and compression ratio. 1.5 RECOMMENDATIONS We recommend the installation of EEMs 1 and 2. These measures provide a simple payback of 5.6 months with Energy Trust incentives, as shown in Tables 1 and 2. Additional bonus incentives are available and outlined in Table 3, which further reduce the payback to 5 months. 1.6 IMPLEMENTATION SUMMARY Review this report and make an implementation decision Your staff has assisted with the development of this report. Because equipment and operational changes are recommended, your organization needs to be comfortable with the data, the analysis and the proposed EEMs for the project to be a success. Please independently evaluate the information contained in this report as you normally would for other projects of this scope. Contact vendors to firm up bids, do your normal diligence and make a decision. Sign an Energy Trust incentive application (Form 420C) prior to signing any Purchase Orders Contact your PDC with your decision, and request and sign an incentive application prior to signing purchase orders or making other financial commitments to proceed with the project. Implement the project Finalize the design in a manner consistent with equipment, set-points, and algorithms described in Section 2 of this report. Any significant differences should be discussed with your PDC and ATAC to confirm that they do not have a negative impact on energy efficiency performance. Sign purchase orders and contracts with contractors. Complete the installation. Commission the project Commission the project according to guidelines in section 5 of this report. Project closeout Send your PDC written notification of project installation completion, commissioning submittals, and documentation of costs by energy efficiency measure. Your PDC will make a site visit to inspect the equipment and prepare a verification report. Your incentive will be paid after Energy Trust approves the verification report. Energy Trust of Oregon -The Sygma Network- Page 10 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
2.0 DETAILED DESCRIPTION OF PROPOSED EQUIPMENT AND OPERATION 2.1 EEM 1 – HIGH EFFICIENCY FAN MOTORS 2.1.1 EEM 1 – Source of Energy Savings Energy savings are realized from the high efficiency of the switched reluctance motors. Therefore, less energy is lost when comparing input to useful work. While this study only analyzed the energy savings resulting from an increase in motor efficiency, the software driven motors are also capable of variable speed operation. For additional energy savings it is recommended to vary motor speed. 2.1.2 EEM 1 – Specific Equipment Recommendations Evaporator and condenser fan motors were upgraded to switched reluctance motors provided by SMC o Evaporator test unit ID: SC5-2 (see Section 4 for details) All 3 fan motors were replaced with switched reluctance motors; however, only 2 were commissioned during the testing period o Condenser test circuit: SC5-2 (see Section 4 for details) Both condenser fan motors were replaced with switched reluctance motors 2.1.3 EEM 1 – Setpoints and Algorithms Recommended to Achieve Energy Performance The high efficiency motors by SMC rely on the same control mechanisms as the baseline case o Evaporator fan motors cycle based on zone temperature via the Beacon II controller o Evaporator fans de-energize during defrost cycles including a short delay post defrost for a coil cool or drip dry cycle o Condenser fans cycle to maintain a targeted head pressure based on cut-in and cut-out pressure switches For additional savings: o Evaporator Fans Modulate fan speed to maintain zone temperatures Minimum speed: 50% Maximum speed: 95% Cycle fans once minimum fan speed is reached Implement a fan delay to operate fans for approximately 5 minutes once the liquid line solenoid has shut before cycling off. This will ensure any residual liquid has vaporized. Electronic expansion valves (EEVs) may be required to implement variable speed evaporator fan control; consult your preferred refrigeration contractor o Condenser Fans Modulate fan speed to maintain head pressure Minimum speed: 10% Maximum speed: 100% Cycle fans once minimum speed is reached Minimum head pressure: 165 psig (for R404A) Energy Trust of Oregon -The Sygma Network- Page 11 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
2.2 EEM 2 – CONDENSER CLEANING 2.2.1 EEM 2 – Source of Energy Savings Over time, debris builds up on the condenser coils restricting air flow across the tubes/fins which negatively impacts heat transfer. This EEM recommends a thorough cleaning of the condensers on all condensing units. By cleaning the condensers, the approach temperature will be reduced, which will reduce compressor input power due to lower compressor lift. 2.2.2 EEM 2 – Specific Equipment Recommendations This upgrade does not require any new equipment 2.2.3 EEM 2 – Setpoints and Algorithms Recommended to Achieve Energy Performance No setpoints nor algorithms are necessary to achieve energy savings Energy Trust of Oregon -The Sygma Network- Page 12 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
3.0 EEM COSTS The tables below summarize project costs. SMC donated materials and labor necessary to install and commission the high efficiency motors for EEM 1. A copy of the vendor’s proposal for the coil cleanings can be found in the Appendix. The cost estimate for EEM 3 is also shown below; however, this measure does not pass Energy Trust’s cost effectiveness criteria, and therefore, is not recommended. Table 4: Estimated costs for EEM 1 EEM 1: High Efficiency Fan Motors Item Description Vendor Qty Unit Total 1 Switched Reluctance Motors SMC 5 $0 $0 2 Installation and Commissioning SMC/PermaCold 1 $0 $0 Total Cost $0 Table 5: Estimated costs for EEM 2 EEM 2: Condenser Cleaning Item Description Vendor Qty Unit Total 1 Condenser Coil Cleaning Permacold 1 $4,424 $4,424 Total Cost $4,424 Table 6: Estimated costs for EEM 3 EEM 3: Evaporator Cleaning Item Description Vendor Qty Unit Total 1 Evaporator Coil Cleaning Permacold 1 $3,219 $3,219 Total Cost $3,219 Energy Trust of Oregon -The Sygma Network- Page 13 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
4.0 BASELINE AND ANALYSIS OVERVIEW 4.1 BASELINE DESCRIPTION The facility relies on multiple, dedicated packaged refrigeration units to condition the following spaces: (1) -20°F ice cream freezer (2) -10°F storage freezers (1) 30°F cooler (1) 38°F cooler (1) 40°F shipping and receiving dock Each refrigeration system consists of a packaged, air-cooled condensing unit piped to remote evaporator air units in the conditioned spaces. Compressors unload either by cycling or unloading cylinders, depending on size and type. Condenser fans are constant speed and cycle to maintain a targeted head pressure via cut-in/cut-out pressure switches. Evaporator fans are also constant speed. Most evaporators rely on electric resistance heating for defrost cycles (with the exception of the dock evaporators). A Heatcraft Beacon II controller for each system cycles evaporator fans with respect to zone temperature setpoints as well as initiates and terminates defrost cycles based on suction pressure and temperature. All refrigerated spaces are maintained at temperature 24 hours per day, 7 days per week for a total annual operation of 8,760 hours. Table 7 and 8 summarize the refrigeration equipment at the site. Table 7: Condensing Units C/U C/U Circuit C/U ID C/U Model Location Refrigerant Make ID SC1-3 Freezer R404A SC1 Bohn JDDS 6000L6 SC1-4 Freezer R404A SC5-1 Freezer R404A SC5 Bohn JDDS 6000L6 SC5-2 Freezer R404A NFU-1A Freezer R404A NFU-1 Bohn JDDS 6000L6 NFU-1B Freezer R404A SC1-1 Freezer R404A SC1 Bohn JDDS 4400L6 SC1-2 Freezer R404A NFU-3 Bohn BDVS 1500L6 NFU-3 Freezer R404A SC5-3 Cooler R22 SC5 Bohn JDDS 3000H2 SC5-4 Cooler R22 SC9-2 Bohn BDS 1000H2 SC9-2 Cooler R22 SC9-3 Dock R22 SC9 Bohn JDDS 6000H2 SC9-4 Dock R22 NDU-1 Bohn BDS 1500H2 NDU-1 Dock R22 NCU-3 Bohn BDVS 2501H2 NCU-3 Cooler R22 Note: SC5-2 circuit test subject for high efficiency fan motors Energy Trust of Oregon -The Sygma Network- Page 14 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Table 8: Evaporators C/U Circuit Evap Evap C/U ID Evap Model Location Defrost ID Make Qty SC1-3 Bohn BHL 1220 1 Freezer Electric SC1 SC1-4 Bohn BHL 1220 1 Freezer Electric SC5-1 Bohn BHL 1220 1 Freezer Electric SC5 SC5-2 Bohn BHL 1220 1 Freezer Electric NFU-1A Bohn BHL 1220 1 Freezer Electric NFU-1 NFU-1B Bohn BHL 1220 1 Freezer Electric SC1-1 Bohn BHL 480 1 Freezer Electric SC1 SC1-2 Bohn BHL 840 1 Freezer Electric NFU-3 NFU-3 Bohn BHL710 1 Freezer Electric SC5-3 Bohn BHE 1200 1 Cooler Electric SC5 SC5-4 Bohn BHE 1200 1 Cooler Electric SC9-2 SC9-2 Bohn BHE 810 1 Cooler Electric SC9-3 Bohn BHA 1100 2 Dock Air SC9 SC9-4 Bohn BHA 1100 2 Dock Air NDU-1 NDU-1 Bohn BHA 1100 1 Dock Air NCU-3 NCU-3 Bohn BHE 1650A 1 Cooler Electric Note: SC5-2 circuit test subject for high efficiency fan motors 4.2 OVERVIEW OF TECHNICAL APPROACH The technical approach considers a combination of logged data, equipment specifications, operational schedules, site observations, and discussions with plant personnel and the refrigeration vendor, PermaCold Engineering. Most of the condensing units are dual circuited meaning two independent refrigeration circuits are present in units SC1-1&2, SC1-3&4, SC5-1&2, SC5-3&4, NFU-1, and SC9. Because of this arrangement, redundancy is inherent making these units great candidates for the motor test. Freezer unit SC5 circuit 2 (or SC5-2) was chosen as the test subject. 4.2.1 Data Logging In order to help us better understand the operation of the facility, Energy 350 deployed data loggers for relevant system equipment. All data logging was done in 1 minute intervals from 3/26/2018 to 4/27/2018. In addition, SMC installed line powered loggers on the evaporator and condenser fan motors on 4/9/2018 during the high efficiency motor installation. Important dates during the data logging period: 4/5/2018: Site specific outdoor weather logger malfunctioned, NOAA data used in lieu 4/9/2018: SMC replaced both condenser fan motors on circuit SC5 with switched reluctance motors 4/10/2018: SMC replaced all 3 evaporator fan motors on circuit SC5-2 with switched reluctance motors 4/10/2018: After repeated efforts SMC was unable to successfully operate all 3 evaporator fan motors in unison. Instead, two of the 3 fan motors were allowed to operate Energy Trust of Oregon -The Sygma Network- Page 15 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
5/7/2018: SMC provided Energy 350 with real power data for high efficiency evaporator and condenser fan motor performance Table 9 details the deployed data loggers. Table 9: Data Logger Summary E 350 Data Channel CT/Channel Description Launch Date End Date Logger ID Interval H81 --- Temp/RH Freezer Temp/RH 3/26/2018 1 minute 4/27/2018 H14 --- Temp/RH Cooler Temp/RH 3/26/2018 1 minute 4/27/2018 1 0-500 psig Freezer C/U B2 Suction Pressure 3/26/2018 1 minute 4/27/2018 Brick (H201) 1 0-500 psig Freezer C/U B2 Discharge Pressure 3/26/2018 1 minute 4/27/2018 D4 --- (3) RoCoils Freezer C/U B1 & B2 Real Power 3/26/2018 1 minute 4/27/2018 D8 --- (3) RoCoils Cooler C/U B3 & B4 Real Power 3/26/2018 1 minute 4/10/2018 3 50 Amp Freezer C/U B2, Cond Fan B2-A 3/26/2018 1 minute 4/2/2018 H65 4 50 Amp Freezer C/U B2, Cond Fan B2-B 3/26/2018 1 minute 4/2/2018 3 100 Amp Cooler C/U B4, Cond Fan B4-A 3/26/2018 1 minute 4/2/2018 H62 4 100 Amp Cooler C/U B4, Cond Fan B4-B 3/26/2018 1 minute 4/2/2018 H102 3 50 Amp Cooler Evap SC5-4 Fans & Heater 3/26/2018 1 minute 4/27/2018 H17 --- Temp/RH Outdoor Temp/RH 3/26/2018 1 minute 4/27/2018 H94 3 100 Amp Freezer Evap SC5-2 Fans & Heater 3/26/2018 1 minute 4/10/2018 H154 --- Motor on/off Freezer C/U B1 Compressor 3/26/2018 1 minute 4/27/2018 H222 --- Motor on/off Cooler C/U B3 Compressor 3/26/2018 1 minute 4/10/218 H91 3 100 Amp Freezer C/U B2 Compressor 3/26/2018 1 minute 4/2/2018 H90 3 100 Amp Cooler C/U B4 Compressor 3/26/2018 1 minute 4/2/2018 1 100 Amp Freezer C/U B2 Compressor 4/2/2018 1 minute H69 2 50 Amp Freezer C/U B2, Cond Fan B2-A 4/2/2018 1 minute 4/27/2018 3 50 Amp Freezer C/U B2, Cond Fan B2-B 4/2/2018 1 minute 1 100 Amp Cooler C/U B4 Compressor 4/2/2018 1 minute H177 2 100 Amp Cooler C/U B4, Cond Fan B4-A 4/2/2018 1 minute 4/10/2018 3 100 Amp Cooler C/U B4, Cond Fan B4-B 4/2/2018 1 minute H270 1 100 Amp Freezer Evap SC5-2 Fans & Heater, L1 4/10/2018 1 minute 4/27/2018 H243 1 100 Amp Freezer Evap SC5-2 Fans & Heater, L2 4/10/2018 1 minute 4/27/2018 Energy Trust of Oregon -The Sygma Network- Page 16 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
4.2.2 Baseline Analysis The following analysis methodology was developed to quantify energy savings for all measures analyzed in this report. Cooling loads for cold storage warehouses are largely influenced by transmission losses between the conditioned, inside temperature and the ambient outside air temperature. Although other factors that influence load are present such as infiltration, internal loads and product pulldown, it is the conduction through the envelop of the cold storage that primarily drives the load. Load is calculated in tons of refrigeration or TR. As such, we begin the data analysis looking at ambient weather trends. Figure 1 compares the outdoor dry bulb temperature (DBT) measured directly at the site with the closest weather station at PDX International Airport. On average, there was a 3.4% difference between the datasets. This is important to note since the onsite data logger malfunctioned on 4/5/2018. Since the difference in temperature between the datasets is minor and since DBT is important for modeling purposes it was necessary to have concurrent weather data. Thus, the analysis uses DBT from NOAA PDX. Figure 1: DBT Comparison 80 70 60 Dry Bulb Temperature (F) 50 40 30 20 Site Sp. 10 NOAA PDX 0 Energy Trust of Oregon -The Sygma Network- Page 17 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Figures 2 and 3 illustrate the suction and discharge pressures, respectively, for circuit SC5-2. The break in the data from 4/10/2018 to 4/19/2018 is due to commissioning efforts by PermaCold Engineering and SMC, primarily for evaporator fan motor troubleshooting. This break is evident in all logged data pertaining to SC5-2. Figure 2: Suction Pressure for Circuit SC5-2 50 40 30 Suction Pressure (psig) 20 10 0 ‐10 ‐20 Figure 3: Discharge Pressure for Circuit SC5-2 300 250 Discharge Pressure (psig) 200 150 100 50 0 ‐50 Pressure data was converted to saturated temperatures based on pressure temperature tables. R404A is a zeotropic halo fluorocarbon blend meaning the composition changes during the boiling and condensing phases. It is important to note the bubble and dew points on the following figure when converting pressure to temperature. Energy Trust of Oregon -The Sygma Network- Page 18 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Figure 4: Bubble Point and Dew Point Illustration for Zeotropic Refrigerants The following temperature conversion curves were used for suction and discharge pressure, respectively. Figure 5: Saturated Suction Temperature vs. Suction Pressure for R404A R404A 0 0 5 10 15 20 25 30 Dew Point Temperature (°F) ‐10 ‐20 ‐30 y = ‐0.0233x2 + 2.242x ‐ 49.617 ‐40 R² = 0.9996 ‐50 ‐60 Pressure (psig) Energy Trust of Oregon -The Sygma Network- Page 19 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Figure 6: Saturated Condensing Temperature vs. Discharge Pressure for R404A R404A 120 Bubble Point Temperature (°F) 100 80 60 y = ‐0.0006x2 + 0.5899x ‐ 3.9033 R² = 1 40 20 0 120 140 160 180 200 220 240 Pressure (psig) The raw discharge pressure data was converted to SCT which was pivoted with respect to DBT. The following regression was revealed for SC5-2. Figure 7: Saturated Condensing Temperature vs. DBT for SC5-2 120 R404A Sat. Condensing Temp (F) 100 80 y = 0.8282x + 42.067 R² = 0.879 60 40 20 Min Float 0 30 35 40 45 50 55 60 65 70 Ambient Dry Bulb Temperature (F) Head pressure (or SCT) is controlled to a minimum of 165 psig (76.7°F SCT) for SC5-2. This is necessary to prevent refrigerant stacking in the condenser and properly feed the metering device at the evaporator. Energy Trust of Oregon -The Sygma Network- Page 20 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Compressors for each condensing unit were mapped using performance data from Bohn, the condensing unit manufacturer, and Copeland, the compressor manufacturer. A multivariate regression analysis was performed to solve for compressor efficiency (kW/TR) considering saturated suction temperature (SST), a proxy for suction pressure, saturated condensing temperature (SCT), a proxy for discharge pressure, and input power (kW) as variables. Regressions statistics and coefficients for SC5-2 is shown in Table 10. Notice the adjusted R Square value is close to 1 indicating a strong correlation between the variables and compressor efficiency. Table 10: Multivariate Regression Statistics and Coefficients for SC5-2 Used to Solve for Compressor Efficiency Regression Statistics Multiple R 0.997491389 R Square 0.994989072 Adjusted R Square 0.994255766 Standard Error 0.030577502 Coefficients Intercept 1.180375494 SST ‐0.017746417 SST^2 0.000210807 SCT ‐0.01988556 SCT^2 0.000241766 kW 0.005090117 kW^2 ‐0.000150601 Saturation temperatures (converted from raw pressure data) and concurrent input power (kW) data from SC5-2 was used to calculate compressor efficiency and ultimately TR. Average TR was then pivoted with respect to ambient DBT to arrive at the following cooling load relationship shown in Figure 8. This is a fairly atypical load relationship for a freezer unit. Notice the load increases with DBT until approximately 47°F. At this point the compressor delivers its maximum capacity which diminishes as DBT increases, or the lift across the compressor increases. As mentioned previously, this is a dual circuit unit with each circuit serving one evaporator in the same freezer. A slight offset in zone temperature setpoints means circuit SC5-2 is the lead circuit and SC5-1 is the lag. Energy Trust of Oregon -The Sygma Network- Page 21 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Figure 8: TR vs. DBT for Circuit SC5-2 14 12 SC5‐2 Comp B2 Load (TR) 10 8 y = 0.2384x + 0.7543 R² = 0.7898 6 4 Trim Load 2 Base Load 0 30 35 40 45 50 55 60 65 70 DBT (F) Figure 9 presents raw data from the compressor serving freezer circuit SC5-2. Figure 9: SC5-2 Compressor Motor Current (Amps) vs. Time 45 SC5‐2 Comp Motor Current (Amps) 40 35 30 25 20 15 10 5 0 Energy Trust of Oregon -The Sygma Network- Page 22 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Real power data from the condensing unit was used to develop a power factor curve with respect to load. This relationship was applied to the motor current for the compressor serving SC5-2. Figure 10: Power Factor vs. Percent Full Load Amps (FLA) 0.80 0.70 0.60 y = ‐1.64x2 + 3.5454x ‐ 1.1828 Power Factor 0.50 R² = 0.9958 0.40 0.30 0.20 0.10 0.00 70% 75% 80% 85% 90% 95% 100% % Full Load Current (Amps) Figure 11 presents the raw condenser fan motor power for the baseline case. It is important to note only one of two condenser fans cycled on during the baseline data logging period; thus, the following figure is representative of such. This was due to the relatively low ambient DBT during the logging period. Spot measurements for voltage and power factor resulted in an average power of 2.05 kW per condenser fan motor when on. Figure 11: Baseline Condenser Fan Motor Power for Test Circuit SC5-2 6 SC5‐2 Cond Fan Motor Power (kW) 5 4 3 2 1 0 Energy Trust of Oregon -The Sygma Network- Page 23 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Figure 12 presents the raw evaporator fan motor and defrost heater power for the baseline case. It is important to note all three evaporator fans cycle in unison. Average fan motor current is 11.18 Amps. This evaporator is equipped with an electric resistance heater for defrosts. Average defrost current is 25.49 Amps. Logged data revealed the following average values regarding defrost cycles: Defrost cycles initiate every 219.1 minutes Each defrost period is 29.5 minutes A fan delay of 4 minutes exists post defrost cycles to allow the unit to drip dry and prevent the evaporator from blowing residual water droplets down the freezer aisle Spot measurements for voltage and power factor resulted in an average fan power of 6.68 kW when in cooling mode. Figure 12: Baseline Evaporator Fan Motor and Defrost Heater Power for Test Circuit SC5-2 25 Evap Motor & Defrost Power (kW) 20 15 10 5 0 The following equation was used to calculate input power (kW) from metered motor current (Amps) and spot measurements for voltage (V) and power factor for a 3 phase circuit: √3 1,000 Energy Trust of Oregon -The Sygma Network- Page 24 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
A similar multivariate regression analysis was developed for the 8,760 energy model to solve for compressor efficiency (kW/TR) considering SST and SCT as variables. Regressions statistics and coefficients for SC5-2 is shown as an example in Table 11. Similar regression analyses were performed for each compressor model. Table 11: Multivariate Regression Statistics and Coefficients for SC5-2 Used to Solve for Compressor Efficiency for Annualized 8,760 Energy Model Regression Statistics Multiple R 0.997414233 R Square 0.994835153 Adjusted R Square 0.992957027 Standard Error 0.034602431 Coefficients Intercept 1.21155942 SST -0.018113435 SST^ 0.000202512 SCT -0.019721997 SCT^2 0.000240584 TMY3 weather data for PDX International Airport (provided by NREL) was used to develop an annualized 8,760 energy model. Cycle rates for the logged cooler unit (SC5-3&4) and the logged freezer unit (SC5-1&2) were applied to all other cooler and freezer unit power profiles. The input power for each non-logged unit was calculated using multivariate regression equations and cycle rates. Table 12 summarizes the annual baseline energy use for each major component of condensing unit SC5 and Table 13 presents the annual baseline data for the compressors of all condensing units. Table 12: Baseline Annual Energy Consumption at Component Level for SC5 SC5‐1,2 Freezer System Component kWh B1 Compressor 66,246 B2 Compressor 167,108 B1 Condenser Fans 15,525 B2 Condenser Fans 18,998 SC5‐1 Evaporator Fans 42,620 SC5‐1 Defrost 18,750 SC5‐2 Evaporator Fans 44,205 SC5‐2 Defrost 23,438 Total 396,889 Energy Trust of Oregon -The Sygma Network- Page 25 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Table 13: Baseline Annual Energy Consumption for Compressors of all Condensing Units C/U ID Circuit ID Location Refrig. Comp kWh SC1‐3 Freezer R404A 66,246 SC1 SC1‐4 Freezer R404A 167,108 SC5‐1 Freezer R404A 66,246 SC5 SC5‐2 Freezer R404A 167,108 NFU‐1A Freezer R404A 66,246 NFU‐1 NFU‐1B Freezer R404A 167,108 SC1‐1 Freezer R404A 43,726 SC1 SC1‐2 Freezer R404A 110,300 NFU‐3 NFU‐3 Freezer R404A 93,363 SC5‐3 Cooler R22 8,049 SC5 SC5‐4 Cooler R22 20,306 SC9‐2 SC9‐2 Cooler R22 18,466 SC9‐3 Dock R22 70,142 SC9 SC9‐4 Dock R22 176,938 NDU‐1 NDU‐1 Dock R22 98,299 NCU‐3 NCU‐3 Cooler R22 27,661 Total 1,367,310 Energy Trust of Oregon -The Sygma Network- Page 26 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
4.2.3 EEM Analysis 4.2.3.1 EEM 1 – High Efficiency Fan Motors SMC supplied real power data for the high efficiency condenser fan motors. The average power data was used in the 8,760 energy analysis when the condenser fan(s) were operational to control head pressure. Figure 13 presents the raw data for condenser fan power. Notice during this data period the second condenser fan motor cycles on. This is due to the slightly higher ambient DBT, and thus higher head pressure, during this time. The average power for each condenser fan motor is 1.54 kW, a difference of 0.51 kW per motor. Figure 13: EEM 1 Condenser Fan Motor Power for Test Circuit SC5-2 6 SC5‐2 Cond Fan Motor Power (kW) 5 4 3 2 1 0 Metered data for the high efficiency, switched reluctance motors was also provided by SMC. As previously mentioned, SMC and the refrigeration vendor were unsuccessful in commissioning all three evaporator fan motors for the test. Instead, two of the three high efficiency motors on the test evaporator were operated. During this period, the compressor pulled a lower suction pressure on average; this is illustrated in Figure 2 from 4/19/2018 onward. This is due to the unit requiring a higher temperature difference (TD) across the coil to compensate for the reduced air flow (two fans vs. three fans). The power data was proportioned to three fans for the 8,760 energy analysis, assuming the third fan will eventually be commissioned. Figure 14 presents the raw high efficiency evaporator fan data as well as the proportioned fan data if all three fan motors were operational. The average input power for all three evaporator fan motors in cooling mode is 4.59 kW, a difference of 2.09 kW from the baseline case. Energy Trust of Oregon -The Sygma Network- Page 27 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Figure 14: EEM 1 Evaporator Fan Motor and Defrost Heater Power for Test Circuit SC5-2 25 Evap Motor & Defrost Power (kW) 20 15 2 of 3 fan motors 10 3 fans 5 0 The reduction in evaporator fan power also saves compressor energy. This is due to the reduction in motor heat dissipation as a result of the higher efficiency at the motors which are located in the conditioned environment. The reduction in fan power was converted to refrigeration load (TR) and multiplied by the operating compressor efficiency (kW/TR) for each hour in the model. The high efficiency evaporator and condenser fan motors are still installed on circuit SC5-2. It is recommended the third evaporator fan motor be replaced or recommissioned as well as variable speed fan control. It is also recommended the high efficiency condenser fan motors be operated variable speed. This will result in additional motor savings and compressor savings. Once all three evaporator fan motors are operational it is likely the site will also realize additional energy savings as a result of fewer defrost cycles. The Beacon II controller initiates defrosts with respect to demand via pressure and temperature monitoring. Reduced evaporator fan energy reduces the internal load which will reduce time the liquid solenoid valve is feeding liquid refrigerant to the coil. Though, this was difficult to model with the current state of affairs with two evaporator fan motors operational and the compressor pulling a lower than average suction pressure to compensate. Energy Trust of Oregon -The Sygma Network- Page 28 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Table 14 summarizes the annual energy for freezer condensing unit SC5. Again, circuit SC5-2 was tested with high efficiency evaporator and condensers fan motors. Table 14: EEM 1 Annual Energy Consumption at Component Level for SC5 SC5‐1,2 Freezer System Component kWh B1 Compressor 66,246 B2 Compressor 159,329 B1 Condenser Fans 15,525 B2 Condenser Fans 14,271 SC5‐1 Evaporator Fans 42,620 SC5‐1 Defrost 18,750 SC5‐2 Evaporator Fans 30,374 SC5‐2 Defrost 23,438 Total 370,553 Energy Savings 26,336 % Circuit Savings 10.4% % Evap Fan Savings 31.3% % Cond Fan Savings 24.9% % Comp Savings Mtr Ht 4.7% % Defr & Comp Savings 0.0% Energy Trust of Oregon -The Sygma Network- Page 29 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
4.2.3.2 EEM 2 – Condenser Cleaning Removing debris from the condenser coil will improve heat exchange effectiveness. The 8,760 energy model assumes a 5°F reduction in approach temperature between ambient DBT and SCT. This assumption is based on previous project experience. The minimum SCT, or head pressure, was not altered in the model as this is important to allow for enough head to move condensed refrigerant to the evaporators and properly feed expansion valves. Energy use was calculated for each hour in the model for each condensing unit. Annual energy savings of 40,200 kWh are realized, or 2.9% from the baseline scenario. Table 15: EEM 2 Annual Energy Consumption Energy Trust of Oregon -The Sygma Network- Page 30 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
4.3 KEY ASSUMPTIONS This section describes important assumptions made in the baseline and EEM analyses 4.3.1 Key Assumptions for Baseline Analysis Table 16: Key Assumptions for Baseline Analysis Baseline Assumption Value Source Analysis Typical Meteorological Weather data TMY3 weather data Weather N/A compiled by NREL. Considered best practice (Portland, OR) for weather sensitive energy analysis Perform as indicated by the Equipment specifications: Bohn condensing Compressors manufacturer's N/A units, Copeland compressors specifications Cycle rates from logged sample freezer and All freezer and cooler Cycle Rates N/A cooler condensing units are representative of condensing units not logged the condensing unit population 4.3.2 Key Assumptions for EEM Analysis Table 17: Key Assumptions for EEM Analysis EEM Assumption Value Source Analysis Typical Meteorological Weather data compiled by TMY3 weather data Weather N/A NREL. Considered best practice for weather (Portland, OR) sensitive energy analysis Perform as indicated Equipment specifications: Bohn condensing units, Compressors by the manufacturer's N/A Copeland compressors specifications All freezer and cooler Cycle rates from logged sample freezer and cooler Cycle Rates condensing units not N/A condensing units are representative of the condensing logged unit population Each condensing unit will realize a 5F reduction in SCT with a clean condenser. Minimum head pressure EEM 2 SCT Reduction 5°F settings to remain. This is an estimate made based on project experience. Energy Trust of Oregon -The Sygma Network- Page 31 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
4.4 SUMMARY OF EEM ANALYSIS Table 18: Modeling Summary Included Demand kWh Demand EEM Description in Total kWh kW Savings kW Package? Savings --- Freezer SC5-1,2 Baseline --- 396,889 --- 66.8 --- 1 High Efficiency Fan Motors Yes 370,553 26,336 64.7 2.1 --- All C/U Compressor Baseline --- 1,367,310 --- 238.7 --- 2 Condenser Cleaning Yes 1,327,110 40,200 236.7 2.0 3 Evaporator Cleaning No 1,318,688 8,422 235.6 1.1 Totals 66,536 4.1 Energy Trust of Oregon -The Sygma Network- Page 32 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
5.0 COMMISSIONING REQUIREMENTS 5.1 PURPOSE OF COMMISSIONING The purpose of commissioning is to ensure that the EEMs are properly installed, working as intended, and delivering energy savings. Some simple EEMs, such as motor replacements, do not need to be commissioned. Although Energy Trust of Oregon does not have a requirement for commissioning, doing so for some measures makes very good business sense. 5.2 LOGISTICAL REQUIREMENTS AND CUSTOMER ASSISTANCE Commissioning should be conducted during typical plant operation. Ideally, the facility would have most or all equipment in use. Commissioning is a cooperative effort between your staff and the contractor. Of course, it is your equipment and you will have the final decision regarding how it is operated. Generally, the contractor will spend a day on site for an initial commissioning visit (with periodic assistance from your staff). Some projects require an iterative process of changing set-points/algorithms and observing performance to achieve optimum performance. Your staff will be involved in these steps as well. 5.3 LIST OF SETTINGS TO BE OBSERVED/CONFIRMED/RECORDED This section is meant for use by facility operators to ensure that settings have been implemented to achieve energy savings. Note that these settings may be modified during the commissioning process. EEM 1: High Efficiency Fan Motors o If the high efficiency motors provided by SMC are still operating at constant speed and rely on the same control mechanisms as the baseline case, then: Evaporator fan motors cycle based on zone temperature via the Beacon II controller Evaporator fans de-energize during defrost cycles including a short delay post defrost for a coil cool or drip dry cycle Condenser fans cycle to maintain a targeted head pressure based on cut-in and cut-out pressure switches EEM 2: Condenser Cleaning o No setpoints are necessary to realize energy savings Energy Trust of Oregon -The Sygma Network- Page 33 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
5.4 PERFORMANCE VERIFICATION PLAN AND/OR O&M PERSISTENCE PLAN Table 19 describes the procedure recommended for the PDC to verify that the system achieves the estimated energy savings. Note that these settings could be modified during the commissioning process and savings should be re-calculated if significant changes were made. Table 19: Verification Plan Type of Item Verification Item Notes Information # Visual motor verification of the high efficiency EEM 1: High Eff motors for evaporator and condenser fans on 1 Physical Fan Motors circuit SC5-2 were confirmed by Energy 350 on Inspection 4/10/2018 during motor installation. EEM 2: Condenser Inspect a sample of condensing unit condenser 2 Cleaning coils to ensure they are free of debris. High performance motor data provided by SMC EEM 1: High Eff 3 via WattNode loggers. No additional verification Fan Motors Data is necessary. Logging Log condensing pressure (or SCT) and ambient EEM 2: Condenser 4 DBT at 1 minute intervals for a period of 1-2 Cleaning weeks. Energy Trust of Oregon -The Sygma Network- Page 34 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
6.0 APPENDIX APPENDIX A – Costs APPENDIX B – Baseline and EEM Analyses APPENDIX C – SMC Motor Literature Energy Trust of Oregon -The Sygma Network- Page 35 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
APPENDIX A – COSTS Energy Trust of Oregon -The Sygma Network- Page 36 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
EEM 2, Item 1 Energy Trust of Oregon -The Sygma Network- Page 37 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
EEM 3, Item 1 Energy Trust of Oregon -The Sygma Network- Page 38 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
APPENDIX B – BASELINE AND EEM ANALYSES Energy Trust of Oregon -The Sygma Network- Page 39 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Energy Trust of Oregon -The Sygma Network- Page 40 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Energy Trust of Oregon -The Sygma Network- Page 41 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
APPENDIX C – SMC MOTOR LITERATURE Energy Trust of Oregon -The Sygma Network- Page 42 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Energy Trust of Oregon -The Sygma Network- Page 43 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Energy Trust of Oregon -The Sygma Network- Page 44 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Energy Trust of Oregon -The Sygma Network- Page 45 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Energy Trust of Oregon -The Sygma Network- Page 46 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Energy Trust of Oregon -The Sygma Network- Page 47 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Energy Trust of Oregon -The Sygma Network- Page 48 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Energy Trust of Oregon -The Sygma Network- Page 49 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
Energy Trust of Oregon -The Sygma Network- Page 50 of 50 Production Efficiency Program Technical Analysis Study (TAS) High Efficiency Fan Motors
You can also read