6. Evaluation measurements and tests
To verify design and compliance with the minimum emergency light level requirements in Table 1, railroads shall ensure that a qualification test is conducted on at least one representative car / area for each emergency lighting system layout, in accordance with this section and Annex C.
For passenger cars placed in service before January 1, 2008, this test shall be conducted by December 31, 2008.
For passenger cars placed in service for the first time on or after January 1, 2008, this test shall be completed before the equipment is released for operation in revenue service.
6.1 Preparation for tests
The following general factors apply to all tests:
• The condition of the battery system shall be identified. The battery should be fully charged.
• All of the battery loads that may be applied under emergency conditions shall be identified. Circuit breaker switches shall be set so that those loads (door operators, PA system, controls, headlights or marker lights, etc.) that are normally present in revenue service are energized during the emergency lighting tests.
• The tests shall be conducted with battery power only; any feeds from HEP, auxiliary, or wayside sources must be disconnected.
• All extraneous light should be excluded to the extent practicable. Meaningful data can be collected only if ambient light can be eliminated almost completely from the areas being measured. Any approach is acceptable as long as ambient light is reduced below 0.01 fc (0.1 lux) in the areas being measured. Several methods can be used to eliminate ambient light for accurate data collection:
- Work at night with cars parked away from bright yard lights;
- Locate cars in a dark, windowless shop or carwash, if available;
- Mask windows and vestibules with roofing paper, flooring paper, or similar opaque materials; or
- Drape cars with opaque tarpaulins.
• When photoelectric cell type instruments are used, the car should be at a temperature above 60oF (16oC) and the instruments operated in accordance with the manufacturer’s operating instructions. This is not necessary with instruments that have temperature compensation built in.
It is recommended that information be recorded identifying date of manufacture and time in service of batteries.
Fluorescent lamps should have at least 100 hours in service before tests are made and be warmed up for at least 15 minutes before any readings are taken so that they will achieve full rated light output.
If the ambient light can’t be reduced to 0.01 fc, there are two alternative measurements that can be used to meet the requirements in Table 1:
1) Measure the ambient light at each location and subtract that value from the value measured with the emergency lighting operating; or
2) If the emergency lighting is at least twice the required levels in Table 1 plus the ambient light reading, consider that the required levels to be met.
6.3 Procedures for measuring illuminance of emergency lighting systems
Measurements of the emergency lighting system performance shall be taken as spatial averages in the immediate vicinity of an action point, as specified in this Section. No single reading used in the spatial average shall be below 0.1 fc (1 lux) (i.e., no dark spots are allowed). The action points considered within this standard are at the door exit release, at armrest level (25 inches (64 cm) above the floor), on the floor, and at specified stairway step locations.
Each section below specifies the minimum number of illuminance measurements required in a particular area of a car. Railroads and car builders are permitted to take more measurements and calculate averages of such measurements.
6.3.1 Required equipment
To ensure accurate illuminance measurements including measurements on vertical surfaces at which the angle of incident light is large, the light meter must be designed to take such measurements and possess:
• Basic accuracy: ± 3% of reading ±1 digit or better,
• Resolution: 0.01 fc or better,
• Cosine error: no more than 6%, measured at 50 degrees, and
• Color correction to CIE photopic curve.
Unless the floor measurement value is known to be at least 5 times the value in Table 1, a 6.5 foot (2 m) separation between the sensor head and the display must be used to ensure that the close proximity of the person taking the measurements does not affect the reading
