Prepared for: 1997 Energy Evaluation Conference - Chicago
Paper dated: November 5,1997
Prepared by Proctor Engineering Group
Contributor: Michael Blasnik
In addition to measure retention, technical degradation and market effects (net persistence) are two other key aspects of savings persistence which have been identified. This paper presents the methodology and results of a study of relative technical degradation performed for the California DSM Measurement Advisory Committee (CADMAC).
The study focused on 13 key measures identified by major California electric and gas utilities as providing the bulk of expected DSM program benefits. The measures ranged from residential air conditioners to commercial lighting, drives, motors, and cooking appliances. The overall approach involved a combination of an exhaustive search for existing information on the technical degradation of efficient and standard measures and an engineering analysis of specific design differences between the two measures. Conclusions were drawn for measures where sufficient evidence was found that relative degradation was either zero or negative (increasing savings over time). For measures where considerable uncertainty remained, research plans were developed to collect the needed data for assessing relative degradation.
The study found that, although little hard data on technical degradation is available, strong and defensible conclusions concerning technical degradation could be developed for nearly all measures. For most measures, the engineering analysis was able to determine that efficient measures suffer from comparable or less technical degradation than the standard measures which they replace. For two of the thirteen measures, significant relative degradation was deemed possible and conclusions could not be drawn. The results of the study are being used by CADMAC to refine estimates of measure persistence and overall net DSM program benefits.
The study also provided recommendations for retention study design concerning technical degradation issues which were deemed premature failure/removal. The study could also be useful to others in assessing some of the technical issues related to promoting efficiency measures.
How will DSM program savings be affected over time by changes in the technical performance of efficient measures compared to the standard measures they replace?
More specifically, the project sought to quantify the annual changes in energy savings which can be expected over the effective lives of specific measures due to any differences in the technical degradation rates of the efficient measures compared to the baseline measures.
The focus of the project was on longitudinal changes in the energy usage associated with the measures. The analysis timeframe was from the period covered by the first year impact evaluation (defining the base level of performance) through the end of the measure's useful lifetime as determined in the California evaluation protocols or by another CADMAC study. Changes in energy usage due to operating conditions, product design or human interaction were included within the scope of the project. Types of degradation which do not affect energy usage, but only level of service were examined where applicable but were not the focus of this project and have not been subject to the same level of research and analysis.
The research question was interpreted to exclude premature measure failure or removal in order to avoid overlap with retention studies.
| Table 1. Study Measures | |
| Efficiency Measure | Baseline Technology |
| Residential Central A/C - high efficiency. | Std. SEER A/C |
| Commercial A/C - Package DX | Std. eff. unit |
| Oversized evaporative cooled condenser | Air cooled condenser |
| Refrigerator - high eff. | Std. eff. refrigerator |
| Electronic ballast | Eff. magnetic ballast |
| T8 with elec. ballast | T12 w/eff. mag. ballast |
| Reflector & delamp | Standard Fixture |
| Metal halide lighting 250-40OW | Mercury Vapor 400-1000w |
| Occupancy sensor | On/off switch |
| Motor - high efficiency | Std. eff. motor |
| Adjustable speed drive for HVAC Fan | Variable inlet vanes or damper |
| Infra-red gas fryer | Std. atmospheric fryer |
| Residential ceiling insulation | Std. insulation levels |
Several of the measures encompass a wide variety of specific products with differing characteristics. For example, higher efficiency air conditioner designs may involve changes to evaporators, condensers, compressors, refrigerant metering devices, and fans. In order to focus the research on the most applicable design characteristics, considerable effort was expended to identify the particular products that were most representative of the California market. In some instances, program databases had to be analyzed to calculate market shares for specific products and then dealers and distributors were interviewed to identify comparable baseline products.
The goal of the engineering analysis was to identify, understand, and quantify the underlying mechanisms of technical degradation for each measure. Once the physical causes for changes in performance of a measure are understood, then existing information can be fully and appropriately utilized in assessing any technical degradation. The analysis plan involved employing a combination of engineering and statistical techniques to estimate degradation rates and/or identify key uncertainties.
We also anticipated that engineering analysis alone may be able to determine or put bounds on degradation rates for some measures. While engineering methods have often been found to be biased in developing point estimates of program impacts due to inaccurate inputs/assumptions and oversimplified algorithms, their application in this project was quite different in that we were generally seeking only the sign (i.e., direction) of the effect, not an estimate of its magnitude.
If the result of the analysis indicated substantial uncertainty for a measure, then the engineering framework would be utilized to help develop optimal research and sampling plans for estimating relative performance changes. By identifying the key performance factors and sources of uncertainty, the analysis would allow these plans to focus on just one or two factors which are amenable to quick laboratory tests or simple spot measurements over time instead of expensive long-term monitoring.
The engineering analysis found that relative degradation is very unlikely for ten of the thirteen measures. Indeed, some measures (residential air conditioners and refrigerators) are likely to degrade less than their standard efficiency counterparts, resulting in increasing savings over time, or "negative" degradation. In one case, HID lighting, a small and quantifiable degradation was found. In three cases (occupancy sensors, optical reflectors, and adjustable speed drives), the potential degradation mechanisms were considered related to measure retention and further investigation would be best performed via retention studies. In two cases (commercial package air conditioners and oversized evaporative cooled condensers), the engineering analysis found that potentially significant relative technical degradation could occur and therefore research plans were developed to collect additional information.
While few measures were found to suffer from relative degradation, many measures are likely to experience absolute degradation (i.e. decreases in efficiency over time). In particular: air conditioners, refrigerators, fryers, and insulation may all suffer from absolute technical degradation. However, this degradation tends to lead to stable or increasing savings over time relative to the standard measure.
Measure specific analysis results are briefly summarized below.
Figure I shows the relationship between condenser face area and normalized efficiency based on air conditioner simulations performed using the Oak Ridge National Laboratory PUREZ model. The condenser area and efficiency are both normalized (i.e., expressed as percentages relative to a baseline system). The figure shows that a 60% increase in the effective heat exchange area of the baseline unit improves efficiency by about 11 %. A 120% increase in area only improves efficiency by about 5% more. The nature of this relationship has important implications for assessing fouling impacts because fouling may be viewed as a decrease in effective heat exchange area (although the actual effects may be reduced air flow, lower surface heat transfer rates, and reduced area).
One could estimate the impacts of a given amount of fouling on the efficiency of systems with differing condenser sizes by expressing the fouling rate in terms of a percentage change in effective area. For example, if fouling reduced effective area by a third each on systems with standard and doubled condenser areas, then the effective normalized areas after fouling would be .67 and 1.33, respectively. The efficiencies of these systems would change from 1.00 to 0. 87 for the standard system and 1. 14 to 1.07 for the efficient system. Therefore, percent savings would increase from 12.3% to 18.7% and kWh savings would increase to 21.5% of the baseline system's initial usage. Energy savings increase from equal fouling percentages because the system with the oversized condenser is less sensitive to changes in effective area than the system with the standard sized condenser -- the efficiency curve is flatter with a larger condenser.
Therefore, relative degradation will only occur if the relative rate of fouling for the efficient unit's heat exchanger is significantly greater than the standard unit's rate.
The use of scroll compressors in some of the more efficient units was also examined and again the efficient unit was found to be less likely to experience degradation than the baseline unit.
We did conclude that the small percentage of efficient units which utilize tighter fin spacing on the evaporator may experience some relative degradation from increased fouling. In addition, some efficient units have TXV instead of orifice refrigerant metering. TXVs may experience positive or negative degradation depending on the circumstances. Overall, we concluded that the potential magnitude of relative degradation due to fin spacing and TXVs on a modest percentage of units is likely to have a much smaller impact on overall savings than the long-ten-n superior performance expected from the other design changes. We also examined other performance factors such as refrigerant charge problems, duct leakage, and evaporator air flow and no relative degradation for the efficient unit was indicated.
Based on detailed component and system level analyses of design differences, we concluded that the high efficiency residential central air conditioners rebated in 1994 are unlikely to experience relative performance degradation compared to baseline units. In fact, they are more likely to exhibit superior long-term performance.