LED Traffic Signal Monitoring, Maintenance, and Replacement Issues (2008)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

9The development of successful practices has many complexi- ties owing to the differences in how agencies operate and main- tain their traffic signals. Further, while experience is building, LED traffic signal modules are an evolving product. The fol- lowing contains insights developed from several agencies. ROAD COMMISSION OF OAKLAND COUNTY, MICHIGAN, CASE STUDY The Road Commission of Oakland County (RCOC), Michigan, began installing LEDs in 2000. The total number of intersec- tions maintained is 1,300, with 680 on LEDs. The LEDs are projected to last between 5 and 10 years (LEDs last longer in cooler climates). RCOC in 2006 received lab tests back on sample LEDs installed in 2000. These tests showed that 10 years is a reasonable life span for LEDs. Therefore, RCOC expects that their LEDs will last from 8 to 10 years. Given the above results and the existing age of RCOC LEDs, the following schedule for LED replacing has been undertaken. Replacement of First Wave of LEDs • Year 2008—Replace 36 locations installed in 2000 and 2001 at a cost of $150,000. • Year 2009 to 2016—Replace about 81 per year at a cost of $350,000 per year. On-Going Replacement Schedule • Years 2017 and beyond—Start replacing 10% per year, estimated to be about 95 and ramp up to 200 by year 2026 when all signal heads will use LED modules. This replacement schedule will provide for incremental upgrades from incandescent to LEDs as well as for the addition of new signal installations. Table 2 shows the estimated funding that is recom- mended to meet this schedule. This schedule assumes that in years 2006 and 2007 the $200,000 in the budget will not be used to install new LEDs, but will instead be rolled over to provide future funding for replacement. Also, in years 2008 to 2010 (assuming a continued $200,000 budgeted), the remaining monies not used to replace LEDs will be rolled over. In years 2012 to 2016, $250,000 will be needed in the RCOC budget. This plan will allow three things to occur: 1. A mostly uniform and known budget amount will be established. 2. The budget amount will not be increased over what has been established over the last few years. 3. In several years, a large number of LEDs were installed. This schedule allows for a smoothing of the replace- ments so no year will see a big budget increase. Essentially, by looking a few years out and developing a plan that allows more consistent funding, the RCOC has developed a strategy that puts them on what is essentially a 10-year replacement cycle. However, it should be noted this is not based on wishful thinking, but on evaluation of what they expect will be the performance of their LEDs based on manufacturers’ specifications and known degradation curves. ASPECTS OF SUCCESSFUL PRACTICES Synthesis study efforts revealed that there are several aspects to successful LED traffic signal module practices including purchasing, monitoring, maintaining, and replacing LED modules. Each of these issues is interrelated and cannot be considered in isolation. Understanding successful practices is an evolving science and users of LEDs should continue to monitor industry practices. Purchase specifications should also consider requesting degradation information from man- ufacturers under varying operating conditions. There is clearly a need for an LED replacement schedule that routinely replaces a portion of the LED signal modules every year so that agencies are not placed in a position of infrequent, but expensive, replacement programs that tend to postpone replacement of LEDs that have passed their useful life. This problem is likely to be most challenging in agencies that replaced their incandescent lamps all at one time. Purchasing, as a minimum, LEDs compliant with ITE specifications will address a number of issues associated with early LED implementations. When making large purchases, it may be desirable for a sample of the product to be sent out for conformance testing. It would also be desirable to test in CHAPTER THREE SUCCESSFUL PRACTICES

the shop the LEDs with the specific brands of load switches and MMUs and/or CMs being used by the agency. Monitoring the maintenance experience with LEDs and keeping records will assist the agency in identifying trends and developing a proactive approach to replacement. While the life of LEDs is long compared to incandescent lamps, LEDs are more susceptible to variations in the environment (heat and cold) than are incandescent lamps. This is another reason for developing a maintenance history of LEDs including the color and type (circular or arrow) of the display, make and model, date of manufacture, date of installation, and the type of opera- tion. Yellow indications have shorter duty cycles than reds and greens. Likewise, some reds and greens may have long duty cycles due to the traffic patterns. A rural traffic signal may have long green times on the arterial and long red times on the cross street. Failure to understand these varying conditions could lead to erroneous conclusions with regard to LED module life. Maintenance intervals are naturally going to increase because of the longer life of LEDs. This reduced maintenance interval is likely to lead to a lack of cleaning of the LEDs. This lack of cleaning will lead to a maintenance depreciation of the light output in addition to the normal degradation owing to usage and age. To determine if intermediate cleaning is nec- essary, a systematic monitoring program, as previously men- tioned, is essential. Testing of light output with and without 10 cleaning will assist in determining if lens cleaning is a cost- effective strategy to increase service life. This would require a program where samples of modules (cleaned and not cleaned) were sent for laboratory testing. The replacement strategy for LEDs is the most complex and currently least understood aspect of LEDs. While the early generations of LEDs are reaching their useful life, they do not represent current technology and are at best only an indi- cator of what may be experienced with newer technology and designs. Agencies will have to be proactive if they are to avoid being driven to costly reactive maintenance. The simplest form of replacement strategy would be to begin by not doing a systemwide replacement program. The most conservative and potentially most costly approach is to have a maintenance program based on the LED warranty period. Assuming a 5-year warranty, 20% of the traffic signal modules would be replaced each year. This would allow an initial establishment of a worse-case budget. At the end of 5 years, an experience base of failures at 1 through 5 years would be established. At the end of 5 years, the budgeting process would be established that would allow for the com- plete replacement of 5-year-old LEDs. However, at this point, laboratory testing could be undertaken to determine the light output (before and after cleaning) for a representative sample of LEDs in different operating environments. Based on the results of testing, a revised maintenance sched- ule could be developed. It would be speculative to suggest what adjustment should be made, but consideration should be given to the fact that LEDs are different from incandescent lamps. It might be desirable to experiment with different strategies that reflect the duty cycle, operating environment, and relative importance of the indication. Reds might be replaced where duty cycles are high and yellows and greens cleaned if testing showed the approach to be cost-effective. Therefore, until an adequate history is developed, careful experimentation may be a cost-effective strategy to not pre- maturely replace LEDs that have additional useful service life. However, just waiting until the LEDs can no longer be seen is not a recommended strategy. Year LED Repl. Cost RCOC Cost Local Cost RCOC Budget Roll-Over Monies 2006 $200,000 $200,000 2007 $200,000 $400,000 2008 $150,000 $135,000 $15,000 $200,000 $450,000 2009 $350,000 $325,000 $25,000 $200,000 $300,000 2010 $350,000 $300,000 $50,000 $200,000 $150,000 2011 $350,000 $275,000 $75,000 $200,000 $0 2012 $350,000 $250,000 $100,000 $250,000 2013 $350,000 $250,000 $100,000 $250,000 2014 $350,000 $250,000 $100,000 $250,000 2015 $350,000 $250,000 $100,000 $250,000 2016 $350,000 $250,000 $100,000 $250,000 TABLE 2 RCOC REPLACEMENT STRATEGY