Effects of High Temperature Operation
CFLs installed in "enclosed" fixtures or insulated ceiling rated airtight (ICAT) recessed fixtures are subjected to sustained elevated temperatures for which they are not designed. Elevated temperature operation has three profound negative impacts on CFL performance: reduced light output, reduced lamp life and color shift. The specifications developed for this project address light output and lamp life. Color shift is an individual manufacturer issue.
Reduced Light Output
CFL light output is a function of the vapor pressure inside the lamp. Fluorescent lamps contain a larger amount of liquid mercury than will become vaporized at any one time. Excess liquid mercury condenses at the coolest locations ("cold spot(s)") on the lamp and vapor pressure depends upon the temperature of the cold spot(s). Lamp design, ambient temperature, drafts, etc., all affect the cold spot. A lamp that is too hot (or too cold) will see a light output reduction on the order of 10% to 20% due to too much or too little mercury being available for the discharge. A general rule of thumb for light loss is 1% for every 1.1°C (2°F) increase in ambient temperature above 38°C (100°F). To combat the problem, some manufacturers have resorted to using mercury amalgams (an alloy of mercury and other metals). The amalgam stabilizes and controls vapor pressure inside the lamp by absorbing or releasing the available mercury. Lamps with amalgam provide more than 90% of their rated light output over a wider range of ambient temperatures. The one negative impact of amalgam technology is that lamps take longer to come up to full brightness, although they still meet the ENERGYSTAR criteria of a maximum three-minute run-up time.
Reduced Lamp Life
Typical CFL products are designed to operate at ambient temperatures of approximately 30 to 40°C (86 to 104°F). This not only maximizes the light output but provides the highest efficacy. When temperatures exceed the optimal range, the electrical properties of the lamp change, which in turn causes the ballast to operate outside its design parameters, allowing more than the rated current flow through the lamp. Long-term operation at higher-than-rated current shortens the life of the lamp.
Another cause of high ambient temperature operation (and reduced lamp life) unique to integral (or screw-based) CFLs is the fact that the ballast components are in close proximity to the heat-generating lamp cathodes. The problem is further exacerbated in ICAT or enclosed applications. Ballast components are exposed to temperatures that approach and/or exceed the temperature ratings of the individual components. The ballast is only as good as its "weakest link" and a single component failure can be catastrophic. CFL manufacturers identified the electrolytic capacitor as the component most susceptible to heat and premature failure. This component, as used in integral CFLs, is typically rated for a maximum operating temperature of 85 or 105°C operation.
Reflector CFL (R-CFL) manufacturers typically warranty their products for an ambient temperature of 50°C and at least 6,000 hours of rated life2 (an ENERGYSTAR requirement). Figure 1 shows the results of 10 current model R-CFLs tested by Pacific Northwest National Laboratory (PNNL) in a simulated ICAT environment. It can clearly be seen that most operated above the manufacturers' maximum operating temperature guidelines.
Figure 1. Ambient Temperature vs. Watts
Color Shift
The color of light provided by fluorescent lamps is dependent upon the type of phosphor used and the mercury discharge. Each of these reacts differently to temperature. High ambient operation tends to shift lamp color toward the blue/green end of the visible spectrum; this shift is noticeable to the naked eye (IESNA, Ninth Edition).
High Failure Rates
While product return data are not widely available, one retailer promoting energy-efficient lighting indicates that the overall returns for R-CFLs are higher than for any other CFL product type, with premature product failure the most prevalent reason for customer returns. Energy Federation Incorporated (EFI) indicates that the "returns to sales" ratio for R-CFLs is over four times higher than for bare glass CFLs (Steele 2002). Size and fit issues account for some reflector returns as well. Return rates are higher for R-40 reflectors than for R-30 reflectors (about 5% vs. 3%), which would most likely be attributable to a size/fit problem. Thermal related stress is probably the single most common cause of compact fluorescent lighting product failures for both lamps and fixtures.
In addition to the limited retailer return data, there is significant anecdotal evidence to indicate performance problems. The Program for the Evaluation and Analysis of Residential Lighting (PEARL) is a watchdog program "created in response to complaints received by utility program managers about the performance of certain ENERGYSTAR lighting products being promoted within their service territories and the lack of a self-policing mechanism within the lighting industry that would ensure the reliability of these products and their compliance with ENERGYSTAR specifications". PEARL, through its member organizations (utilities, EPA, DOE, market transformation groups, etc.), purchases products available in the marketplace and conducts independent testing to verify compliance with ENERGYSTAR requirements. Test results are only provided to PEARL members and individual manufacturers whose products were tested. Test results are not public information; however, there is consensus within the market transformation community that the R-CFL product category has had difficulty in a number of areas including, but not limited to, lumen maintenance, life, and efficacy.
(2) The point at which 50% of a large sample of lamps are still in operation.