Don't be fooled by readings of HID light sources from cheap light meters.
Although light meters can accurately read footcandle (fc) values under incandescent light sources, did you know some are quite inaccurate under some High Intensity Discharge (HID) sources? To get around this problem and ensure reliable, accurate measurements, you should use meters having the following features: cosine correction; filters or adjustments to match the human eye (CIE) response curve; multiple ranges; low battery indicator; and remote sensor with a "sample-and-hold" feature. If you need absolute accuracy under a given light source, then look into laboratory calibration of your meter for the particular light source you plan to measure. But first, let's take a look at how the human eye comes into play.
Visual response and the human eye. Photometry is the measurement of light radiation in terms of normal human visual response. The idea is to measure light in the way we perceive it, by eliminating those frequencies of light to which our eyes are blind (like ultraviolet or infrared), and replicating the way our eyes perceive the intensities of visible light colors.
In 1921, the Commission Internationale de L'Eclairage (CIE) established a standard observer response curve known as the photopic luminous efficiency function. Per this standard, detectors in the eye or in a light-measuring instrument respond differently to different regions of the spectrum. Note that the intended meaning of an fc value directly relates to this standard observer response curve. An ideal light meter responds to light in the same manner and amplitude.
Basically, the human eye responds to low-level light (below one fc in scotopic colorless night vision) differently (much less) than to higher lights levels (in photopic color day vision). This is an easy simplification, since most fc requirements are in the photopic range.
Problem areas. Consider the typical response curve of a simple, inexpensive light meter constructed with an uncorrected selenium sensor. The fc value for a pure violet light at 400 nanometers (nm) is almost infinitely greater than a human eye can discern. So this light meter would accurately record fc values of "black light" (what nightclubs use), but the human eye would only barely perceive this light. Manufacturers resolve this problem by adding light filters and band amplifiers to their light meters. This enables the fc readout value to match the human eye and CIE Luminous Efficiency Curve.
One problem light source is the deluxe metal-halide (M-H) lamp. An uncorrected meter will indicate a higher fc value of deluxe M-H light than a CIE-corrected meter. This is because the deluxe M-H lamp emits violet, ultraviolet, and infrared colors of light the CIE color-corrected meter can't measure (and to which the human eye isn't sensitive). In other words, for a given lumen level under deluxe M-H light sources, a cheap uncorrected selenium cell light meter that's calibrated against an incandescent lamp tends to read a higher fc value than its more costly CIE-curve corrected counterpart.
The High Pressure Sodium (HPS) lamp is another problematic light source. Since a 4100K, cool white (CW), fluorescent lamp has a relative light color output that closely mimics the CIE luminous efficiency curve, a typical light meter will indicate more fc per lumen for this type of lamp than that from a deluxe HPS lamp. This is because the HPS lamp develops much of its light at wavelengths more red than values to which the meter can respond. So, using a CIE color-corrected meter for a given fc reading indicates you'll need more watts of deluxe HPS than for 4100K fluorescent lamps. This tends to be offset by the high lumen/watt characteristic of HPS lamps compared to that of fluorescent lamps. For this reason, the high lumen/watt characteristic of HPS lamps frequently makes them preferable over fluorescent lamps, especially where you need a high intensity light. Most people choose HID lamps because of the huge reduction in luminaire quantities.
Similarly, since the 4100K CW fluorescent lamp's relative color output closely mimics the CIE luminous efficiency curve, a typical meter will read more fc per lumen than that measured under the 4200K deluxe CW fluorescent lamp.
Besides incandescent sources, you can measure other sources, with equal accuracy; with corrected and uncorrected meters. For example, because light from a clear mercury vapor (CMV) lamp falls within the response curves of uncorrected selenium and color-corrected meters, both types of meters provide identical fc readings under this type of lamp. But, the commonalities of these sources are few.
Although not as serious as those problems related to light color (spectrum), consider the following problem areas with light meters:
• Battery output can vary, causing reading values to vary. You can correct this by checking the meter's batteries often; using a meter with a "low battery" warning light; or remeasuring the first point measured in a session and verifying you get the same value at the end of the session.
• Light entering the meter at an angle can refract away, causing reading errors of up to 25%. You can eliminate this problem by using a light meter that has a "cosine correction" feature.
• Meter calibration can vary between measurement scales. Have the light meter calibrated often to counteract this effect.
• Ambient temperature can affect the performance of photoconductor sensors. You can permanently damage these sensors by placing the meter in temperatures above 120 DegrF.
• As with the human eye, sensor sensitivity changes as it is exposed to light. For more accurate readings, remove the sensor cover between five and 15 min, prior to making measurements. Also, don't place the meter in a direct light that is many times the intensity of the light you're measuring.
• Light meters (particularly analog meters) have scale errors within the extreme ends of their scales. Several fc scales on one meter eliminates this problem. In fact, you can enhance accuracy by selecting the scale that reads near its center.
• Some of the light meters have a sensor as a part of the meter readout assembly, so the operator blocks some of the light when reading the meter. You can eliminate this by connecting the sensor to the meter through a long flexible cord and having a sample-and-hold button on the meter.
Keep these possible errors in mind when making lighting measurements and when working on a job involving contractual fulfillment or requiring high technical accuracy. Using a light meter that is color-corrected and calibrated to measure the colors and anticipated levels of light is best. Without this proper tool to determine useful lighting values, you can be literally "left in the dark"
Paschal is a Senior Engineering Specialist, Bechtel Corp., Houston.