You know the drill, a supplier comes in to your facility and begins to make assertions that you can save 50%, no wait, 150% on your lighting bills by switching to this new energy saving lighting. In recent years the hot debate for lighting has been the use of T-5 lamps in place of traditional metal halide, high pressure sodium, or mercury vapor (you're not still using mercury vapor are you?) lighting. There may actually be something to this upcoming trend, in this article we will work out some of the issues and try to clarify some of the murky issues surrounding the debate. When working through the issues there are several important factors plant personnel will need to understand:

• Light Depreciation
• Efficacy
• Initial and Operating Costs
• Fixture Light Dispersion
• Controls
• Light Quality

Light Depreciation
When a light bulb is first installed, a certain amount of light is emitted by the bulb (assuming it is turned on), this light is most often measured in lumens. There is room for argument here about good light versus bad light, but lumens is a measure of all light. Over the life of a bulb the total lumen output of a bulb will decrease. For HID lamps such as high pressure sodium or metal halide, this decrease can be as much as 40% of the original output, or "initial lumens". This decrease is considered the light depreciation level. For fluorescent lamps this level is most often about 10% of the initial lumens. This is where a majority of the confusion comes from surrounding light bulbs, whether you measure using the initial output, or the "mean" output, which is the output at 40% of the bulbs life. A manufacturer will always use the one that sells their product better for their calculations.

Efficacy
Similar to efficiency, higher efficacy is better. Any energy geek (like myself) will refuse to call the light output (lumens) divided by the power input (watts) efficiency because efficiency is a unitless measure of performance, and this has units. Table 1 shows the efficacy for various lamp types. Based on efficacy alone, one would typically choose the T-8 fluorescent, but that is only a part of the picture.

It is important to note that the output of the T5 lamp is optimized for different operating temperatures (95°F) than T8 and T12 lamps (77ºF). It is hard to know what this will mean in a fixture in your facility, but if operating conditions are not at the optimal temperature for any lamp type including HID, the output will be lower. It is likely that lamp manufacturers are trying to design to the more realistic temperatures seen inside a fixture. This only adds to the beautiful mystery that lighting can provide.

Initial and operating costs
The initial cost of fluorescent fixtures is very comparable between T5, T8 and T5HO at somewhere between $11 and $15 per thousand mean lumens delivered, while Metal Halide fixtures run at around $9 per thousand mean lumens. Light fixture costs will vary greatly because of the range of quality of materials and design, this range will likely dictate how effectively the light is delivered to the necessary places. The difference in initial cost is quickly dwarfed in an annual operating cost of around $6.50 per thousand mean lumens for metal halide versus $4.50 per thousand mean lumens for the fluorescent systems. As far as lamp life goes, each one is typically rated for 20,000 hours and the cost per bulb will range greatly. The moral of the story is that energy cost will often trump equipment cost. The other moral is that fluorescent fixtures are about 30% less expensive to maintain on a light output basis.

Fixture Light Dispersion
Light dispersion is where most of the controversy around HID versus T5 versus T8 lies. In high bay spaces a brighter light source must be used to get light down to the operating level. When I say brighter, I mean more lumens, but one benefit of high bay areas is that more lumens can come from a smaller point and be dispersed using lenses and mirrors. In the lighting industry, a lot of attention is paid to glare. Glare is caused when too much light is coming from one source and making it unpleasant to see in areas near the source. In high bay areas, glare is not as big of a problem as it would be if the ceiling was at 8 feet and the lights were located in the work area. With the commercialization of T5 lights in the mid 90s, the problem of glare became apparent for lower commercial spaces, and the lights were relegated to being hidden in recessed ceilings and above fish display cases and promoted as a "low profile" product. But as manufacturers became more creative and learned to spread the light better, the industrial use started to take off and the use of high output (HO) T5 lamps began. Manufacturers have quite successfully designed fixtures that create less glare than HID fixtures, and can reach the same or greater levels of light output per fixture. The final result for the consumer is a fixture that can shed as much light as a metal halide fixture with better light dispersion.

Controls
Fluorescent fixtures have always outperformed HID fixtures in terms of controls. The ability to turn fluorescent fixtures on instantaneously provides the opportunity for the use of occupancy sensors that turn the lights on only when needed. The response rate of fluorescent lamps allow for dimming ballasts to be used to take light output from 1% to 100% of the range of the light, similar products for HID lamps are many times more expensive and can only reduce the light output to 60%. These dimming technologies allow for active systems that work with daylighting to continuously determine necessary light levels and respond accordingly.

Light Quality
There are any number of measures of light quality with the most recognized being the Color Rendering Index (CRI). CRI is a percentage scale that works to describe the surreal discomfort that is felt under different light types with respect to sunlight (100%). Incandescent lights provide a CRI of about 95% while a metal halide HID fixture scores a 65 to 70. Some typical CRIs are given in Table 3.

The difference in CRI for fluorescent lights is in the type of phosphor used. Traditionally manufacturers have used RE70, which provides CRIs in the 70s. A Newer type, RE80 provides CRIs in the 80s. These various phosphor types are also responsible for the increased lumen maintenance of newer lamp types. A lot of dialog is ongoing regarding the concept of photopic lumens, the ones the industry uses to report output, and scotopic lumens, the ones that the human eye responds to. My best advice for this is to wait for the experts to work it out before paying too much attention to it. In general the claim is that fluorescent lights are much better than many other lamp types in terms of scotopic output.

Summary
The battle over the lighting market is currently raging, but it does appear that fluorescent fixtures are winning the battle for now.

• In general fluorescent lamps provide better energy efficiency
• The light output of an HID fixture will quickly degrade to about 60% of the rated output while fluorescent will only degrade to 90 or 95%
• The control performance of fluorescent lights far exceeds that of HID by eliminating the 15 minute restrike time, allowing for occupancy sensors and dimming capability
• Fluorescent lights outperform HID in terms of light quality
• When using Fluorescents above about 20 feet, T5 fluorescent should be used and preferably T5 High Output
• Fluorescents below 20 feet should be T8 or possibly T5
• Proper fixture selection is essential to having good light quality light efficiency

There is a large abundance of lighting literature out there, and now we have gone and added to it. For qualified technical information it is usually best to locate an unbiased third party such as the Lighting Research Center in New York at www.lrc.rpi.edu or your Progress Energy representative.