Evaluation Measurement and Verification of the California Public Utilities Commission

APPENDIX TO: Evaluation Measurement and Verification of the California Public Utilities Commission HVAC High Impact Measures and Specialized Commercial Contract Group Programs

2006-2008 Program Year

Final Review Draft v1

Submitted:December 11, 2009

Volume 2: Appendices

Prepared for the

California Public Utilities Commission

C/O Jeorge Tagnipes

505 Van Ness Ave

San Francisco, CA94102

A.RCA Standard Contextual Data

B.Investigated RCA Analysis Methods

C.RCA Diagnostic Calculations

D.RCA Details of Metered Units

E.Rooftop or Split System Air Conditioner Replacement Program Methods

F.Rooftop or Split System Air Conditioner Replacement Site and Meter Issues

G.Rooftop or Split System Vendor Analysis

H.AC Replacement Vendor Survey Instrument

I.Reference and Background for Duct Sealing Methods Applied

J.MAP Closure Discussions

K.SCE 2537 TURBOCOR Field Data Collection Forms

L.Upstream HVAC SDG&E 3029: PTAC Savings Algorithm

M.Upstream HVAC SDG&E 3029: High Efficiency Motors

N.Non-HIM Programs Closed and Not Subject to Site Inspection

O.Guidelines for Estimating Net-To-Gross Ratios Using the Self-Report Approaches

P.Simple Residential/Small Commercial Free-Ridership Algorithm

Q.Free-Ridership (kWh weighted) Stability Indicators

A. RCA Standard Contextual Data

Res RCA DEER Parameter Data Collection

C&I RCA DEER Parameter Data Collection

Res AC Replacement DEER Parameter Data Collection

C&I AC Replacement DEER Parameter Data Collection

Duct Sealing DEER Parameter Data Collection

B. Investigated RCA Analysis Methods

Introduction

Ultimately, the airside estimates provided the most repeatable method for deployment on the planned samples.The evaluators developed a matrix of the implementation plan and how measurements relate to the various analysis options.The emphasis of developing and presenting the matrix to the CPUC Energy Division was on finalizing the evaluation plan while leaving open the discussions of detailed analysis methodologies.To support this effort, the evaluators simulated a range of heat loss assumptions for the simplest approach which showed reasonable uncertainties even with the assumption being the major unknown.Additional work was done with an uncertainty committee to look into techniques for simulating uncertainty with MonteCarlotechniques for uncertain measurement that then fed into a model with its own level of uncertainty.Also the evaluators and working groups investigated the availability and uncertainties of modeling, maps and their adjustment factors.The matrix is presented with the key at the top in Table B1, with the green highlighted cells absolutely required of all methods and the orange highlights representing the more difficult to measure items.The yellow highlighted cells indicate that we are still exploring those uncertainties.

There was agreement that the Simple COP using the Assumed Heat Loss method was feasible while exploring the other methods using pilot data.The regression option remained feasible while compressor maps and simulations had low feasibility due to unknown adjustment factors and unknown heat exchanger details.We collected enough data from the pilots to attempt all methods, but the full scale effort could only include a subset of the measurements required for less feasible analysis options.

Table B1: Bi-Quadratic DOE-2

The following describes three methods for estimating AC system efficiency considered but not pursued for the final data analysis: simple COP, compressor map, and refrigerant cycle simulation.

Simple COP (Assumed Compressor Heat Loss)

This approach used a simplified energy balance across the compressor.The overall equation simplifies to the following:

Equation B1: Mass Flow Rate Equation

Where is the mass flow rate, f is the heat loss from the compressors (and can be assumed as 2-10% of compressor power input,), and is the enthalpy difference of the refrigerant before and after it goes through the compressor.Enthalpy values can be obtained based on the temperature and pressures at the suction and liquid lines.An overall mass flow rate can be determined from these parameters.

Some commercially available refrigeration system analyzers use this simple energy balance across the compressor to determine efficiency[1].This is achieved by assuming the heat loss from the compressor as a percentage of the electrical input power.The heat loss assumption is typically in the range of 2-10% and may be chosen as 7% for hermetic and semi-hermitic air cooled systems.By taking temperature and pressure measurements of the refrigerant entering and leaving the compressor along with simultaneous power measurements, one can estimate the system mass flow rate.The approach is most effective for systems with easily accessible compressors which is not typical for residential split systems and may require removing the condenser fan, installing the temporary sensor and re-installing the fan.This raises additional liability and warranty issues and for some older split systems the fan grill may be riveted to the case preventing access altogether.Infrared measurements may be feasible for some configurations but with lower accuracy than other measurements being simultaneously recorded and additional issues of poor view angles even when distance and line of sight are good.Newer larger systems, especially SEER13 and R-410a systems may have access to the compressor more typical of packaged systems.The heat loss assumption will be investigated further for scroll compressors and two stage compressors.

Compressor Map

A second method to determine the mass flow rate is by using compressor maps as published by their manufacturers.This data is specified by ANSI/ARI standard 540.The maps have data on the compressors that describes the equations for mass flow rate and are of the form:

Equation B2: Map Coefficients Equation

M(TS,TD) = C1 + C2TS + C3TD + C4TS2 + C5TDTS + C6TD2 + C7TS3 + C8TDTS2 + C9TSTD2+ C10TD3

Where C1 - C10 are the map coefficients, TS & TD are the compressor suction & discharge saturation temperatures (°F), and M is the mass flow rate as a function of its TS & TD.

Given the refrigerant operating pressures, we can determine from compressor maps the efficiency, capacity, power draw, and even amps to within five percent of manufacturer specifications which is the specified accuracy of the maps established by ANSI/ARI Standard 540.The maps also gave us refrigerant mass flow rate from the operating pressures, by using the flow rates and the operating temperatures we estimated capacity and compared the map values.By adding a spot power measurement, we calculated an operating COP and also compared to manufacturers rated COP for those operating conditions according to the performance map.The ease of compressor identification and availability of documentation was adequate for units manufactured after 1992 and poor for older units.Units after 1992 also have manufacturer literature on superheat adjustment factors for the maps at different superheat conditions.When pressures, temperatures, and power input were monitored over varying ambient conditions, ARI Standard 540 was also used to generate new performance maps.There are superheat adjustment factors at least for some for some compressors at different levels, although for TXV systems we did not come across any adjustment factors for different subcooling conditions.

Simulation Modeling

A third analysis option involved inputting the system through a simulation model such as NIST Cycle D or Oak Ridge National Lab Mark V programs.These models have built in compressor maps though user specified mapping is an available option.The user inputs system characteristics as measured or identified in the field collection phase.The program was able to simulate a vapor compression cycle and calculate a mass flow rate based on these user inputted specifications.

The background development for the Mark V model was also included in the collection of reference material.The primary challenge in using these models was the lack of known or observable information on heat exchanger designs.It was feasible to develop prototype models in one of these software packages and calibrate them to field measured data.These models allowed for the use of default or user specified compressor maps.The models also presented errors themselves beyond the compressor maps they were based on.Based on comparisons to field data, the models themselves may have uncertainties on the order of +/- 3-10%.

C.RCA Diagnostic Calculations

The program, Title 24, and industry standard procedures are consistent and are used by the verification team.Target superheat or subcooling values are obtained from manufacturer’s data or calculated from the 2005 Residential ACM Approval Manual and compared to actual values.The procedure for calculating actual subcooling and superheat is as follows:

1. For Non-TXV systems determine the evaporator saturation temperature (Tevap, sat) from ASHRAE saturation tables for the measured suction pressure. For TXV systems use the measured discharge pressure to determine condenser saturation temperature (Tcond, sat).
2. Calculate the Actual Superheat or Subcooling for Non-TXV and TXV systems, respectively. This is calculated as follows:

Actual Superheat = Tsuction - Tevap, sat

Actual Subcooling = Tcond, sat - Tdischarge

1. Determine the Target Superheat or Subcooling for Non-TXV and TXV systems, respectively.
2. Calculate the difference between actual and target values as follows:

Actual Superheat – Target Superheat

Actual Subcooling – Target Subcooling

1. Non-TXV systems:If the absolute value of the difference is less than or equal to 5 the system is considered to be adequately charged.
2. If the difference is greater than 5, the system is likely undercharged.
3. If the difference is less than -5, the system is likely overcharged.
4. TXV systems:If the absolute value of the difference is less than or equal to 3 the system is considered to be adequately charged.
5. If the difference is greater than 3, the system is likely overcharged.
6. If the difference is less than -3, the system is likely undercharged.

Additional post-field analysis includes using the superheat and subcooling calculations along with pressure and airflow data to determine additional outcomes indicating other potential system issues such as insufficient evaporator or condenser flow, improper TXV operation, etc.

The program requires airflow verification similar to Title 24.The actual temperature split between supply and return dry bulb is calculated as shown in the steps below and compared against the target split as outlined in the 2005 Residential ACM Approval Manual.The method essentially verifies that flow is greater than 350cfm/ton for a large percentage of units based on empirical data.

1. Calculate the Actual Temperature Split as follows:

Actual Temperature Split = Treturn, db – Tsupply, db

1. Determine the Target Temperature Split using the appropriate tables from the 2005 Title 24 Residential ACM.
2. Calculate the difference between the actual and target values as follows:

Actual Temperature Split – Target Temperature Split

1. If the absolute value of the difference is less than or equal to 3 than the system has adequate airflow.
2. If the difference is greater than 3, the airflow is too low.
3. If the difference is less than -3, it is unlikely that the airflow is too high.Most likely the capacity is low on the system.

In addition the verification effort will utilize direct flow measurements using an orifice plate flow grid and digital manometer.The temperature split method verifies that most systems have flow greater than 350cfm/ton, but we require flow measurements for evaluation purposes as well as later comparison to DEER values.

D.RCA Details of Metered Units

Table D1: Service Performed on Units with No Refrigerant Charge Adjustment

The team also examined the resulting changes in performance as a result of any service noted in the table above. These only occurred for units in SCE 2507 territory. The remaining units in SDGE 3043 experienced a failure event, as noted in Table D1above. The results of additional tune up service actions are shown in Table D2, and show a range of improved and degraded performance as a result of service.

Table D2: HVAC Changes Resulting from Non-Refrigerant Charge Service

Detailed RCA Field Findings

Residential Pre- Post RCA

Multifamily

Arrangements were made to monitor 89 units serving apartments in moderate and hot climates. Of those units, 30 had sufficient data to meet all the criteria for inclusion in the analysis. The addition and removal of charge was recorded along with calculated pre and post cooling outputs and efficiencies below.

Table D3: Multifamily Pre-Post Results

Table D4: Multifamily Pre-Post Results Continued

Single Family

The single family homes that were monitored generally had two air conditioning units and were coordinated through the programs in normal operation. The change in system performance and degradation factors were calculated and shown in Table D5.

Table D5: Single Family Pre-Post Results

Manufacture / Mobile Homes

The third-party comprehensive manufactured and mobile homes programs were also sampled for the residential RCA high impact measure pre and post monitoring study, in addition to the mass market programs presented above. The results are shown in Table D6 below.

Table D6: Mobile Homes Pre-Post Results

Commercial Pre- Post

Similar to the residential programs, the pre and post metering sample was evaluated on the basis of improvements in EER, capacity, and sensible capacity, as well as demand reduction. The changes in total and sensible capacity and energy input ratio were developed as fractions to represent the pre condition where the factors for the post maintenance case would all be equal to one. Finally the change in efficiency was expressed as the pre and post energy efficiency rating (EER). The data has been divided into two separate tables, the first for dual compressor units (Table D7) and then for single compressor units (Table D8).

Table D7: Dual Compressor C&I RCA Pre-Post Results

Table D8: Single Compressor C&I RCA Pre-Post Results