U.S. Department of Energy - Energy Efficiency and Renewable Energy
Building Technologies Office
Commercial Building Energy Alliance LED Site Lighting Performance Specification (text version)
Below is a text version of the Webinar titled "U.S. DOE Commercial Building Energy Alliance LED Site (Parking Lot) Lighting Performance Specification," originally presented on February 17, 2011. In addition to this text version of the audio, you can access the presentation slides and a recording of the Webinar (WMV 16 MB).
Terry Shoemaker: Welcome, ladies and gentlemen. I'm Terry Shoemaker with the Pacific Northwest National Laboratory, and I'd like to welcome you to today to our Webcast, U.S. Department of Energy LED Site Lighting (Parking Lot) Technology Specification Project brought to you by the U.S. Department of Energy's Commercial Building Initiative.
We are very happy today to have our speaker, Michael Myer of the Pacific Northwest National Laboratory and Mark Peternell, Regency Centers. Michael Myer has been with PNNL just under four years. Prior to joining PNNL, Michael worked as an architectural lighting designer in New York. Since joining PNNL, Michael has worked on a wide ranging number of lighting projects. Currently, Michael splits his time between three major programs: SSL Commercialization, Federal Appliance Standards and Commercial Building Energy Alliances. These programs provide Michael with a crosscutting point of view where he can borrow and share across the program. Within the SSL, which is Solid-State Lighting Commercialization Program, Michael has been involved in many SSL GATEWAY demonstrations. These are real world installations of LED products. Michael enjoys getting in the field and learning firsthand about some challenges. Michael also provides technical support to CALiPER and other SSL programs. In the Federal Alliance Program, Michael provides a technical analysis for lighting products and equipment for both test procedures and equipment standards. Finally, Michael's work with Energy Alliance is far reaching. Michael is the technical lead for the LED parking lot specification and the high performance parking structure specification. Beyond these projects, he provides lighting technical support for many other activities. The Energy Alliance work allows Michael to work with large, national companies where the largest potential to transform the market exists.
Michael A. Myer: Thanks, Terry. So today's presentation is on the commercial, sorry, commercial building energy alliance LED (site lighting) performance specification. It was developed about a year and a half ago and has actually been installed in a couple of sites, Regency being one of them, and Mark will provide his point of view and expertise on implementing the specification and some of the challenges of the site owner that they've had to deal with. The presentation today will include overview of the Energy Alliance's illuminance and why it was selected for this specification and some background information there, lighting power density, economics related to LED technology for parking lots, some reliability issues. Then we'll move into specifications and practice, and then I'll turn over the presentation to Mark who will present on Regency Centers.
It's always great to start off with energy use on an overall level, so this slide is a great breakdown of energy use by end uses for the United States in 2008. You can see by the pie chart that buildings take the largest share plurality of the energy use, and then the building is broken out into two separate graphics. As you go down the list, residentially its 10%, but the big number, commercially, is lighting represents about 20% of all commercial building energy use. Just for those who are wondering, across the country, overall end use lighting represents about 8% of all energy generated, so that's a big deal. And if you can actually go after a 20% savings in a single technology, it's something to do. And if you notice, if you look down at the commercial sector, it is the largest single end use. When you lump heating and cooling together, it is larger, but you don't always have to do that depending on your environment.
The specification is called the Commercial Building Energy Alliance Specification and what the Commercial Building Energy Alliances the CBEA is how we refer to it, is actually three smaller energy alliances. It started off first with the Retailer Energy Alliance. Then it went to the Commercial Real Estate Energy Alliance and finally the Hospital Energy Alliance. The Energy Alliances were a program started by the Department of Energy focusing on similar end users who have common challenges, and they're all moving towards energy efficiency for different needs and different reasons. Most often, as a company, they can reduce their operating expenses if they operate their sites more energy efficient wise, and that helps them. Some of them have company mandates. Some of them have state requirements; actually require them, depending on if they're building a new site. And let's say California, they're going to have to be energy efficient. And then others do it because it is part of their criteria as a good company and social issues, and then really some just come down to straight honest dollars, and that less money they're spending there is more money they're making somewhere else.
The Energy Alliances break down as follow. This is market share of Energy Alliance members in 2010. So the Retailer Energy Alliance is 54 members and about 2.8 billion square feet. The Commercial Real Estate Energy Alliance is about 61 members and another 4.5 billion square feet. And the parts of the circle that are colored in represent members in the overall market sector, so CBEA members represent about 20% of the general retail industry at large. And then the Hospital Energy Alliance is about 49 members, and they represent about a half a billion square feet in the United States.
And then I just wanted to break down this - today's Webinar by the occupations of the people who signed up for it. So across from the sign-in sheets, roughly 30% of the people attending this phone call claim manufacturing as their occupation. There's a small contingent of state and federal agencies. EEPS, being Energy Efficient Program Sponsors, and then others are people such as designers, engineers, manufacturer representatives and even some utilities. It's good to know who we're speaking to. At some points, I will try to tailor parts of it to different entities to make it more appealing.
The parking lot specification is just one program going on by Energy Alliances. There are three other ones. The upper left, of course, in this picture shows the LED parking lot specification and practice. The middle one is a parking structure or parking garage specification, and that technology encourages the use of LED, fluorescent or induction sources. The idea there is that more sources that could be used with lighting controls, whether it be day lighting or an occupancy sensor, is very advantageous saving energy in a parking structure. The other lighting specification that exists is the LED refrigerated case display - - case lighting specification, as shown in the bottom right hand image. And the reason why LEDs were selected there was that typically fluorescent is used, but fluorescents don't like the cold, and actually as temperature decreases, output also decreases. In contrast with LEDS, as temperature decreases, output actually can increase, and so they like the cold much more than the fluorescents do, and so it's actually a better fit from a lighting technology point of view. There are other things that the specification looks at, but from a physics point of view, LEDs make more sense than fluorescent.
And then the fourth one that's not lighting here is the unitary rooftop air conditioner, which you're seeing in the bottom, left hand side, and there's more information coming on that on other energy alliances and other programs. I'm not an HVAC person by any means, so I'm just going to mention that and move on.
So I always like to start off by what is a specification. The site lighting specification is a performance specification, and different people have different opinions of what a performance specification is versus a technology specification. So it is a performance specification meaning that it's adopted by the end user or the site, so in this case it could be adopted by a Walgreens or a Walmart or Regency Centers or even a government entity or a municipal parking lot. That's who adopts it, and what they're requiring is that the product must deliver X for Y energy unit. So in this case, the specification says that the end site wants so much light in footcandles and within so much range being uniformity that sets the maximum and minimum, and we'll talk about it more in a second, for energy units. In this case, it's watts per square foot, and so that's how this performance specification works.
There are technology specifications which can be done by a request for product or mass procurements, and they're often incentivized by utilities or energy efficiency programs. These are different in that they typically, in lighting terms, they say product X delivers Y lumens for Z watts. A great example of this is Energy Star or the Consortium for Energy Efficiency Premium T8s. There are a number of programs out there where, for instance, Energy Star. It's on the TV, and it just says it must draw this much power in the off state and these type of factors. It doesn't tell you how many channels it'd have to have or anything like that, and that's where a performance specification comes in. It really tries to set what the lighting system is doing, not just how much power the lighting system's drawing. It's really looking at where the light is going and trying to encourage that more than anything else.
The next slide is a discussion about illuminance. So in a parking lot, illuminance is the amount of light falling on a surface. That's the definition for illuminance for all cases. But in a parking lot, many sights use what's known as an average illuminance, yet the IES, which is the Illuminating Engineering Society of North America, make recommendations. In their document, RP-20, which is the recommended practice for parking facilities, recommends that parking lots should be lighted to a minimum, not an average, horizontal illuminance. And so in this case, they're saying a minimum of 0.2 footcandles. And so that's a really relatively low minimum illuminance, yet many sites actually require much higher numbers. It's not uncommon to see 3.0 footcandles or 5.0 footcandles, 10.0 footcandles. I know some companies actually say as much as allowed by code. That's how much they want.
I want to take a step back for everybody into high school or junior high mathematics, but in average, it's a sum of a lot of values. So in this case, the average over on the bottom right, where it says it's an average of six in one case and the other one has an average of 2.5, but the average of six is being affected by the maximum in the red box. So if you have a lot of light in one spot, you can skew your average. And then the blue box that's appearing now is the minimum, being 0.3 or 0.8, so it's possible to have a high average and actually a lower minimum than another sight that has a lower average but a higher minimum. And the reason why I'm starting with illuminance is because that's really what people want in their parking lots. Most specifications by companies or other programs say how much light they want, and again, a lot of companies require an average and so often neglect the minimums, even though that's what the IES recommends. And I also like to highlight, this is what we're showing here, is that you can actually still get a lower average but a higher minimum through better distribution, and the specification tries to encourage that.
And I'm moving into the next slide. So the illuminance requirements in the parking lot specification, they're done by ambient zone, so our lighting condition or lighting zone and this is becoming more common across the United States in that you don't need the same amount of light everywhere. I'm located in the greater Boston area, so we're a more dense, urban area versus someplace, let's say, in Lawrence, Kansas, where it's less dense and less urban. You don't need as much light there because there's less people, and there's less sky glow and other competing sources, and so the idea here is that your lighting zone is based on essentially population density and some other factors, and the lower the number, the less dense it is, and therefore, you need less light. Lighting zone 0 and 1 are residential, and 0 is actually a national park zone. So of course, you wouldn't put a retail center or a commercial enterprise in a national park zone, and that's why it's a 0 because it doesn't apply.
But again, the specification sets it as a minimum, so in this case, if you're in a very dense, urban area, the specification is just calling for the main parking area, a minimum of 1.0 footcandle horizontally and vertically, 0.5, so it is acknowledging that it is also exceeding the IES recommendations. That is what the end users of the specification did want. They did want more, but they also tightened down the uniformities to prevent significant over lighting. The specification also looks at different parts of the parking lot in that it treats the parking lot differently. Rather than requiring a 1.0 footcandle everywhere, it actually says well wait a minute. The perimeter, those parking spots way out on the edge that are used only high traffic days or in the worst case scenarios, we don't really need as much light there, and they're often bordering property lines, so it's cumbersome actually to light them to high light levels because of light trespass issues. So it actually recognizes that's a place they don't want as much light. They still want good quality light, but not nearly as much as somewhere else.
The front aisle, this is the crux of lighting issue here is that people are leaving the parking lot and entering the building, whether it be a grocery store or a commercial building or an office building, this is the highest place for an accident because the vehicle/pedestrian conflict is high. You have someone literally leaving their car on foot, and you have a lot of cars crossing through there at the same time, and so that actually is where you want more light. The most amount of light in the specification is for the front aisle zone to prevent and reduce any vehicle/pedestrian conflict.
And then the other part of the parking lot that is separated is the loading area or the rear drive. Most buildings have a backside, or a side and then for sure a backside. This is where deliveries are made, their trash dumpsters are located, sometimes employee parking, and depending on the site, it's a narrow area usually because most of your parking tends to be on the front of the building. And so because of that, they don't need as much light, and sometimes they don't want to encourage people back there, so they don't want as much light back there. So the specification separates that zone as well.
So this breaks down the specification into the other zones. So in this case, the specification, again, requires different light levels again by zone and in part of the parking lot. So in the front aisle where there's high conflict with cars, if you're in a dense area, it's actually 2.0 footcandles minimum because that's where you want the most amount of light. And then in the periphery areas, such as the perimeter or the rear area, less light is required.
The next element is power density. So power density essentially is as straight forward as it sounds. It's watts per square foot, or meters if you're in Canada or other places. So how you calculate it is pretty straight forward. It's the input wattage of the fixture times the number of fixtures you have and divided by the total area of the sight. The first thing I'd like to point out is the two codes on the far right, ASHRAE/IES Standard 90.1-2010 and California Title 24-2008. So Standard 90 is a code that is adopted by many states. Some states also have an IEC code, which is very similar to Standard 90.1. They're adopted at different time periods, so some states have adopted Standard 90.1-2001. Other states have adopted 90.1-2004 or 2007. Some states will start adopting 2010 code soon. California has their own energy code called Title 24, and the most current version of it that is in effect right now is the 2008 version.
The thing that I like to show is that energy codes, like our specification, are recognizing that you don't need as much light in less dense places. So if I'm in middle of nowhere Illinois, you don't need as much light as if you're in Chicago, and so because of that, they're starting to actually restrict the allowed power density for those sites, again, based on ambient zones. So that's where the zone number decreases, the power density that's allowed also decreases. In California, they do give a little bit of wattage adder, that's separate from the power density. Standard 90 does it on a larger level, that's why I recommend consulting the Code for all requirements.
So for the first zone, the specification is lower than either the two codes, and it's for the first two zones, but when you start getting to zone two or zone one, the specification approaches the same numbers as the codes do or virtually do. It should be known that the energy savings are not coming from the LED technology. They're coming from limiting the power density. You can actually have high energy use with an LED system as well where the difference is is that an LED system allows you to you don't lose as much light within the fixture like you do with conventional technologies, as well as LED systems can allow you to use lighting controls such as maybe dimming or occupancy sensors that you can't use with conventional controls. But to guarantee some level of energy savings, the power density was limited rather than trying… Another methodology would be to use something like kilowatt hours, but there are a lot of problems with that. Parking lots are not metered as much, and it's hard to separate the parking lot lighting from some other types of lighting, and so that's why power density was chosen to guarantee a level of savings and then specification encourages the use of lighting controls as well.
Continuing on into actual applications; this is a computer rendering, so why it's a performance specification. So this is an actual rendering done to lighting software of a parking lot for a hotel in Tennessee. They were considering possibly using a pulse start metal halide system, and their average illuminance here is 4.0, almost 5.0 footcandles, and the minimum is 0.2. What I always like to show is that max to min is a 75 to 1. So as you see in there, they have those high, bright spots directly under the fixture, and then dark everywhere else, kind of like a checkerboard effect. And that's typical of some conventional lighting technologies depending on the layout and the fixture selected. And the power density here is 0.1 watts per square foot, whereas, the hotel chain in Tennessee was considering possibly going with LEDs, and we applied the LED specification to their site, and the power density, again, was much lower here. To meet it was 0.06. The average did decrease. The minimum was slightly greater than the previous one. The minimum could have been better, but we were limited by some issues with the site lighting and where we could put it, but the max to min was much lower. In the previous slide with the metal halide, it was 75 to 1. Here it's more like about roughly 12 to 1, and that's because the optical characteristics of the LED fixture selected allowed for more of a uniform light. It essentially looks like a blanket of light, very uniform and not hot spots and dark spots as conventional technologies tend to be.
As I do the specification presentation and I get questions, people always say, "Well where do LEDs make sense?" This is a graphic of the average retail electricity price by state in 2009. The darker blue - - the darker the color, the more expensive it is, and then over on the left hand side is a ranking of the top ten from highest to lowest. So if you have a site in Hawaii where you're paying a lot of money for electricity, that's a great place to start. The next many sites are in the northeast, Connecticut, New York, Massachusetts, New Hampshire, Rhode Island, New Jersey and Maine. They're all expensive as well. Only California and Alaska are outside that. Alaska also has high electricity rates because of the way they get their energy source.
So the first site is if someone's looking at and saying, "Well, I think I'm interested in LEDs, but I hear they're expensive." Well the first place to try them out is probably in place like New Hampshire where you're paying $0.15 a kilowatt hour, or New York or Connecticut, $0.18 a kilowatt hour. Any reduction in energy use can help pay for itself at that rate.
Continuing on with economics, where do LEDs make sense? The next thing you need to think about is hours of operation. Some sites operate their lights pretty religious. It gets dusk around 7 p.m., and the store closes at 10 or 11, and therefore, they turn off the lights at that time, and that's really four hours of operation. Well if you don't operate your site for a lot of time overnight, that's not really going to help you because your hours of operation are going to affect your energy case, and so that's something to think about. I mean from a pure energy point of view, it's actually better to just turn off the fixtures than to operate a lower wattage for a longer period of time, and so that's something to think about. Another thing to think about with LEDs, is that you can actually dim them or control them with occupancy sensors, and that is something we'll discuss a little bit later where we've actually done sites that have done both of those type of things.
Other economics that need to think about is maintenance. So lamps and ballasts have to be monitored and replaced, and I say monitored because they're going to fail at different rates. And so some companies actually specialize in this. They drive out to your parking lot, and they drive around it one night, and then they come back in a week and they take notes and see what fixtures are burned out, and then come back and replace them, that type of stuff. So maintenance is a real cost, but not a lot of companies just consider it a cost of business. Other people have onsite electricians and don't actually know how much they're spending to maintain their lighting, and it's going to vary by region. Certain places labor is more expensive. As your poll height increases, it's going to be more expensive because it means a bigger bucket truck or more insurance, et cetera, but below are some real numbers. When you factor in things, we've heard that a real industry average from actual lighting manufacturers is about $225. It's a real cost to replace just one lamp, because if you've got to bring out one, maybe two, people, a bucket truck and so on… RS Means is a national construction estimating reference book. They estimate it's $65 per hour per person to do any maintenance in site lighting, and so if you think about how many fixtures you have on site, that can add up. The rest of the numbers are real numbers from real projects.
And why bring up cost of maintenance is that some sites actually calculate how much they're spending in maintenance and factor that in to the economic situation and they say, "Okay, well if this LED fixture is going to last ten years, I don't have to do that maintenance that I was doing every two years. I can save that money, and that changes the economic story," rather than just the current situation.
Continuing on with cost and economics, where do they make sense? Well LED lumen output correlates to cost, so the more light you need from an LED fixture, the more expensive it's going to be. Unlike conventional technology, that is not the case. So conventional technologies, which is the center column, metal halide shoebox, which is this image down at the bottom is a typical shoebox, and these pictures came from Grainger's Website because it's publicly available. You don't need a login, and the prices are pretty stable based on geographics, lighting industry. You do pay more depending on if you go through a manufacturer's rep. If I'm in New York City versus Kentucky, the prices are different. So there's definitely a geographic component to pricing, but I just used Graingers as a quick idea to show how pricing works.
So metal halide, you can go from a 250 watt nominal metal halide to a 400 watt metal halide. This is what the actual output is. It goes up. Yet the price didn't really change much. You're looking at a $70 adder, so your lumens per dollar is much higher, so there's cost related to light output is not really a factor in terms of lighting technology. In terms of conventional technology, the appearance also matters. So this fixture over here is what some people call an architectural feature. It has smoother lines. It's cleaner than the shoebox over here. So because of the manufacturing aspects related to that, it is a more expensive fixture, which is what you're seeing over on the right hand column. So the appearance of conventional technology affects the price sometimes much more than the actual light output of conventional technology.
In contrast is the LED, so these values here. So luminaire output, this product is producing 5,600 lumens, and this one's producing about 11,000 lumens, and yet the price is almost double, and that is because you're paying per LED that you put in the fixture, and there are ways to mitigate cost. You can change drive currents, some other aspects of it, but again, the idea here is to show you that if sites continue to want a lot of light… I've heard from fast food chains or quick service food places, they require 10.0, 15.0 footcandles, and when I start hearing these high numbers, it's like well LEDs it's really hard to make them work there because as you keep increasing the output, the more expensive your fixture gets. So that's something to think about. How much light do you really need? At some point, keep adding light doesn't mean that people can read better, doesn't mean people can see better, so that's something to think about.
Moving into the next slide. Part of the specification had to do with reliability, and the specification does carry a warranty. This is because, I just showed you in the previous slide, LED fixtures do cost more than conventional technology. Yet when we started this process, they were not getting warranties, so a warranty on a conventional light product is usually at one year for the lamp, two to three years maybe for the ballast, and then the fixture, that tends to carry maybe a ten or longer year warranty, but that's because it's on the finish paint or something else. Well two years ago when we started this whole process, retailers were still giving that same type of warranty, and commercial enterprises were saying, "Wait a minute. I'm not going to spend 2,000 or 3,000 more dollars when the warranty hasn't changed." "You're telling me this fixture's going to last five, maybe ten years, maybe 20 years, but yet you're only warranting it for two years." So the specification puts in requirements, and very specific requirements, on the warranty. It says what it needs to be, what it covers and that type of thing, and not because of anything entirely that we've done, but I think because a lot of people in the industry have been voicing complaints about this, the lighting industry, many good LED products now carry a five year warranty. I actually saw recently a 20 year warranty from a product. I was a little shocked by that because I don't know of many other products that have a 20 year warranty, and the warranty should cover light output and color. It's not just how much light is being produced, but depending on the quality of the LEDs, they can change color over time, and it should factor in that. The warranty should also cover the appearance and the mechanical stability of the light fixture itself.
Continuing on with reliability. Again, so I've mentioned that LED fixtures can last a long time. So as we started developing the specification, that's where end users really wanted help in mitigating risk, and well how long is this fixture going to last because depending on the manufacturer, the light ratings are very different. And also right now there's not entirely an industry standardized number - - or methodology for testing luminaires to estimate life. This graphic is from the DOE CALiPER Program and for those of you that are not aware of the CALiPER Program, this is a program the Department of Energy manages, and they buy products anonymously. They send them to an independent testing lab, and they test them, and then they publish the results. So manufacturer X claims something, and the testing showed this. Sometimes the performance was better, sometimes it was the same, and sometimes it was worse, but what it's trying to show is that not all manufacturers are testing the right way. Not all manufacturers are being entirely honest, and so it's just adding validity to the market testing.
So CALiPER tested a number of different light sources over 6,000 hours, which is about three quarters of a year if it's left on continuously. And I really like this slide because it shows you what LED output looks like over time, and some of these numbers look weird, the fact that it actually gains in output and then it slowly begins to decline. And then other ones are kind of a bumpy roller coaster ride, and then it kind of levels off. That's somewhat normal depending on the LED chip manufacturer. Other products like this might be a little bit of a warning sign how quickly it's going down the line. So what this specification does is… And actually, the specification, you should know, is a document that we continuously update. I received a call just this week from a manufacturer saying the methodology we're using works for certain manufacturer LED chips, and they wanted to propose a third methodology for estimating life because you could use something called an exponential decay. You could use another binomial function or you could use some much more complex math to characterize some of these distributions because it's not a straight line. It's not… The fact that this output actually gains in time and then reduces, is - - you have to use more sophisticated math, and so the specification currently as written, actually only allows exponential decay, but we're in the process of revising that for new methods that are being adopted by the industry. But again, the focus of the specification is to limit the risk on the end user side.
So moving into the specification in practice. This picture is of a Walmart installation by using GE lighting in Leavenworth, Kansas. It is the first location to test drive the LED site lighting specification. Walmart was very interested in using LEDs. They had many vendors come to them, and they were a member of the Energy Alliance. One of the things they do like about the LED technology is that because of the optical design characteristics of the GE fixture, but also a number of other LED fixtures, is that they don't have as much dark sky issues or good neighbor problems. So as you can see in this image how dark it is, how close it is to the fixture. This Walmart site, there's a community over here to the right that is pretty close to the site, and they definitely had concerns about a retail center moving in, and they were actually quite happy with the cutoff. Good cutoff is still possible with conventional technology, but sometimes it's a sacrifice of output, but LEDs don't always take in less of a hit in output.
So this is a highlight of the numbers. Everyone's always very interested in what are the numbers so they can compare it to their site. It's about half a million square feet of surface. It is concrete. It is not asphalt. It's roughly about 92 luminaires. They're the GE Evolve, which is here. The mounting height is 37 feet above grade. People are always asking, "Well how does it work for my site". The average footcandle is about 1.2 footcandles. The minimum is 0.8 footcandles. They did see a payback between six and ten years. It is a range because we don't know the exact pricing. GE and Walmart have their own special relationship, as well as many commercial enterprises have with specific manufacturers, so based on some numbers, we just put a range. In this case, it's Kansas, so the electricity price is pretty low. Being in the Boston area, that's very low for us, that definitely affected the payback. If, of course, this site was in California where it's twice that rate, the payback would have been much shorter. If it was in Hawaii where the payback's three times that rate or almost four times that rate, payback would be even shorter. They were considering a standard design, which is 1,000 watt metal halide, and this was about 63% energy savings. It is 0.04 watts per square foot. The 1,000 watt metal halide system did provide more light in terms of an average basis, and in some cases, on a minimum basis. Walmart was very interested in taking this specification for a test drive, and has been happy with the site, and they did a comparable study in Peoria looking at the conventional technology. And the way they characterized it, which is always the way we hear it in lighting, is they heard no negative complaints. We always know there's a problem if someone complains, but if no one complains, the best we can say is no one complained and kind of move on with our day because rarely, unless there's a lighting person there saying, "Oh, this is great lighting," do people comment on the lighting, especially something as benign as parking lot. Very few people are looking at parking lot lighting.
The next test drive of the specification was a retrofit application in Manchester, New Hampshire. This is an open air strip mall, and it is using beta LED luminaires. These fixtures are controlled by an occupancy sensor, so there's a little eye up there in the luminaire looking down on the pavement and when a car comes by, it goes from a low setting to a high setting, and we're doing a follow-up study right now. This is part of the DOE GATEWAY Program that was mentioned in the beginning. This is an actual installation, so we published a report on that, and we have a secondary report coming out that's focused only on the use of the occupancy sensors.
So how often were they on? How often were they going from high to low? What were some of the pros and cons of the technology? Did people like the up down up down effect of them? Did they cover a sufficient area? Were they affected by wind? What are the issues with a car triggering the sensor versus a person triggering the sensor? That'll all be in Phase II of the report because there's some great promise… This is a great example this parking lot has a T.J. Maxx and a Casual Male and a Payless shoe store. So it's a typical strip mall that most of us have in towns where we live. What's interesting to think about is most of these stores, except around the Christmas holiday period, close around 9, maybe 10 p.m. So why was the site lighting to 100% from 10 p.m. on through the middle of the night? At 2 a.m., this place is entirely deserted. Do they really need as much light? No, and they were concerned about safety and security, so they still wanted to provide some light. Methods that you can do that still provide some light. You can turn off every other fixture. The downside of that is you have dark and bright spots again. You could dim, or in this case, they went with occupancy sensors. Dimming and occupancy sensors are much easier to accomplish with LEDs than they are with conventional technology.
Again, here are the highlights of the T.J. Maxx site. It's about 150,000 square feet. It was 25 luminaires. The mounting height was slightly lower than I showed you at Walmart. It was 33 feet. Here the average was about 2.0 footcandles. The minimum was 1.0 footcandle. The payback here was a very large range. The reason is that while the electricity price is much higher, here it's $0.14, as well as they were paying a fair amount for maintenance in that they had some power quality issues, and so when they installed the new LED fixtures, they also installed a devise to literally clean the power. And so if you cut their old maintenance numbers, the payback would have been this low, but the ten years factors in a more traditional maintenance number, as well as a more traditional national average for electricity, and so that's why it's so broad. So there are a number of factors that affect ultimate payback.
Energy savings here, again, are about 63%. It just happens to be a coincidence compared to the previous Walmart slide. That is an estimate that we're verifying through the use of tracking the occupancy sensors in how often are they in the high state and how often are they in the low state. And the watts per square foot again come in at 0.04 watts per square foot.
At this point in the presentation, I'm going to turn it over to Mark Peternell of Regency Centers to talk about their experience as a site owner. He's got some great slides on their operational issues, as well as their successes with LEDs and some of the challenges they've dealt with. Mark Peternell is a Vice President of Sustainability for Regency Centers. They're a real estate investment trust specializing in developing, owning and operating grocery-anchored and community shopping centers. Under his leadership, Regency has cost-effectively reduced their use of natural resources by implementing operational process improvements and new technologies, including smart irrigation controllers, LED parking lot lighting, and networked lighting controls. Prior to Regency, Mark worked for Forrest City Enterprises where he was responsible for the development of the company's retail portfolio in Denver, Colorado, including the first open-air lifestyle center to achieve LEED certification. Mark serves as a member of the USGBC: LEED for Retail Committee, International Council of Shopping Centers Environmental Subcommittee, and Real Estate Roundtables Sustainability Policy Advisory Committee.
Today's Webcast will provide an overview of the LED parking lot performance specification in action. Mark Peternell of Regency Centers will discuss their approach and experience incorporating the specification into their parking lot portfolio.
Mark Peternell: Thanks, Michael. Good afternoon and good morning, everyone. It's a pleasure to speak with you today about Regency's outdoor lighting initiatives, specifically our participation in the Department of Energy's Solid-State Lighting KOA Demonstration Project at our Willston II shopping center in Falls Church, Virginia. As Michael said, my goal is to provide a real world application of the specification and LED technology, including both some of the positives as well as some of the challenges that we faced during this project.
But first, let me take a step back and give an introduction to Regency Centers in our broader sustainability and energy efficiency initiatives. As Michael said, we're a national owner/operator and developer of grocery-anchored and community shopping centers. Our portfolio consists of 396 properties, encompassing 53 million square feet of gross leasable area. We're in 28 states, plus the District of Columbia, and we support that portfolio with 17 offices across the country. We are a real estate investment trust publicly listed on the New York Stock Exchange, and our market cap is 5.9 billion.
GreenGenuity is our tagline, if you will, for our sustainability and energy efficiency initiatives. Essentially, this is our commitment to develop and operate environmentally sustainable shopping centers that utilize an evolving set of best practices and building design construction, operations and maintenance. This program was formally initiated in 2007 that really follows a long history of quality and innovation, and I want to stress that point. When we talk about building green buildings and sustainability, really what we're talking about is a quality initiative, and I think LED is a great example of that. Also want to stress the importance of industry sustainability leadership and some of the indirect benefits that Regency has seen as a result of this initiative. For one, as a publicly traded company, we've seen increased interest in sustainability initiatives from both our current and potential investors. We've had direct development opportunities created as a result of our commitment to sustainability. You also have the benefit of other stakeholder interests such as some of our key retailers that are also a part of the Retail Energy Alliance, as well as interest in our employees in working for a more sustainable company. And then lastly, as Michael's slide showed, you see the various increasing amount of green building and energy code regulations and compliance issues. But most significantly, is the opportunity for utility expense savings and the value proposition associated with those drivers.
So for Regency, why outdoor lighting? Obviously, energy is a significant and increasing operating expense for our portfolio. We spend approximately $9.5 million each year on our utility expense strictly for outdoor lighting. As an owner of primarily neighborhood and community shopping centers, the vast majority of our overall energy expense goes to outdoor lighting. We also have seen over the past decade that this line item is increasing at a faster rate than our other expenses at about 4 to 5% per year. The other factor is our lighting maintenance expenses. With a portfolio of 53 million square feet, the expense each year to maintain our lighting systems is quite significant at about 2.3 million per year. When you combine our electricity expense for outdoor lighting with our average annual maintenance expense, together these represent about 11% of our portfolio's total operating expenses.
Another component of our maintenance is our capital improvement projects where we've been spending over the past two or three years approximately 3 to $400,000 on lighting retrofit projects.
So what have our lighting practices been historically and how are those evolving as a result of this increased emphasis on energy efficiency? We really do not have a national lighting specification or standard. There are a couple reasons for that. One is for a development business, we typically design to our anchor-tenants, to their requirements, and often times, we also are forced to design to jurisdictional requirements. Those two are often times in conflict. From my experiences, sort of the anomalies I see in the industry is that you see huge differences in light levels from one property to the next. So when I think of the idea of a standard, it's really hard to get my arms around that because some of our sights have an average footcandle as low as 1.5, and others are in excess of 10.0 footcandles. I really think that this begs the question how much light is the right amount. So the above factors make it difficult for us to adopt a lighting policy. That's one of the nice things about the DOE specification is it really takes into consideration differences from the lighting zones, so that's certainly a step in the right direction.
So what are our decision making criteria for lighting retrofits? And we're looking to replace older, less efficient technology, certainly looking at sights that either have higher historical energy consumption and/or higher electricity rates. Equally important is the maintenance expense history and when I move to talk about our Willston II project in Falls Church, that was a big factor in our decision to move to LED technology, was the history of high maintenance expenses. But I also want to add I get the question all the time, "Well what's your payback criteria? What are your ROI?" The lighting decision is not always an economic or payback decision. A lot of times when we're looking at a lighting retrofit project, we're really looking at right sizing. So for example, if a project has an average footcandle of 1.5, we may increase that by two or three times to get to that right amount of light based on whatever the jurisdictional requirements are or that anchor-tenant's requirements, and that situation may be very difficult to save any energy. At the same time, in the case of Willston II where you had an average footcandle closer to 8.0 or 9.0, there's tremendous energy savings opportunities regardless of whether using LED or conventional fixtures, so again, emphasis on right sizing. It's not always a purely ROI or payback decision.
Also aesthetics. Clearly we're looking at replacing older and less efficient technology, but again in the case of the Willston II case study, you'll see that that project was undergoing a major overhaul, and so the decision to replace our lighting fixtures was embedded into that redevelopment program because we wanted to go with a newer, more modern fixture. I think for us, the economic criteria is not so much about whether or not to retrofit, but whether or not the technology makes the most sense for your retrofit.
So moving on to our experience with the specification. Willston Center II is a shopping center that was originally constructed in 1986. Regency purchased this property in 2005. It's a very sort of traditional neighborhood shopping center anchored by Safeway and includes a mix of national and local restaurants, office and retail tenants. As you can see from the before picture, ever since we acquired the property in 2005, we knew that Safeway had a desire to both expand their store to a more modern footprint, and obviously, upgrade the façade of the center. As a result of Safeway's desire to expand, this led to a complete redevelopment of the shopping center, including a new exterior façade, new signage, a new parking lot and landscape, as well as exterior lighting.
Here is just an aerial photograph of the project. It's the Center here in red. The interesting thing about this site is there's an adjacent shopping center anchored by Target, and as Michael will attest in visiting the property after the installation of the LED, when you look at those two projects side-by-side, you really can't see much difference in the levels of the light quality between the two projects. And just to give you a sense of geographically where this project is located in Falls Church, Virginia, which is within the D.C. metro area beltway.
So the lighting conditions at the site prior to the redevelopment, it was a really old system, about 25-year old system that used 1,000 watt probe start metal halides. They were 30 foot polls, and the average footcandle level at the site was 8.0, and the max to min ratios was simulated. We don't know exactly what it was, but we estimated it was at least 40 to 1, so much like Michael showed in his computer rendering. Average footcandle really high, but the uniformity was actually quite bad. So we had several areas on the project that were hot spots, and then several dark areas. Financially it was an expensive system to operate. Our kWh rate in this territory is about $0.095, so probably about the national average. In total, we were spending about 18 to $20,000 per year in electrical costs to run the lighting system. That does include other exterior lights like wall packs and soffit lights. And then our maintenance, like I said, this was a huge sort of eye opener for us. We went back and did a detailed review of the maintenance history, and on average, we were spending about $500 per month in maintenance, lighting maintenance, for this system. That included both monthly night sight inspections, with the failure rate we were having for lamps and ballasts, that was a requirement, and then additional labor costs to send out a bucket truck to replace the technology, and then obviously, your material costs.
As I mentioned, and this is an important point to make when we start looking at the economics, the decision to replace the site lighting at Willston Center II was a foregone conclusion. This was part of our overall modernization of the center. The original plan before we got involved with the DOE GATEWAY Project was to replace the 1,000 watt probe start metal halides with a newer 400 watt pulse start metal halide. So I think energy savings and maintenance savings were going to be a foregone conclusion, and the justifications for considering LED is several factors involved here. First one is I just want to acknowledge that we probably would not have done this without our involvement with the Commercial Building Energy Alliance and really having the technical support from Michael and the team at PNNL because there's a learning curve associated with adapting new technologies, so that was a big driver. The other thing that I want to acknowledge is this was a $2 million plus redevelopment, and anyone who's in the construction industry realizes that regardless of what the ROI might be, first costs are extremely important, and had we not had savings in other areas of the construction project, the thought of increasing the site lighting budget by 40 or $50,000 to adapt a new technology, probably would not have occurred simply because the construction budget was locked in and we had savings. That opened the opportunity for us to explore new technology.
That being said, from a life cycle cost standpoint, both the energy savings and the maintenance savings were significant and I'll share some of those numbers with you in just a moment. We talked about the importance of improving the uniformity and the color quality, and then also we looked at this as a test site, much like the Walmart project in Kansas. And we have a large portfolio. We're spending all this money on capex projects each year. There's a lot of talk about LED, and we really wanted to get our toe in the water and have our own experience with this new technology.
So we move on now to our selection process. This is really kind of where it gets fun and interesting. With Michael's assistance, we reached out to several national manufacturers, as well as their local product reps, and from my experience, I'd say that the response was very surprising. We didn't get a great response initially from anyone. Some didn't really respond at all, and some were very slow to respond, and that was just sort of mind boggling to me given all this R&D that's gone into investing in new technology, and maybe that's because our type of development is very fast-paced, and we were sort of demanding a very quick turnaround and a quick response. But I'll have a little bit of color commentary on that here on my conclusion slide, but that was sort of our first source of frustration.
And not having a lot of experience with lighting procurement in this fashion, I'd say that the number of different players involved really seem to slow the process down. In some cases, we were dealing directly with the manufacturers, and in other cases, we were dealing with the manufacturer's rep. In other cases, they sent us to their distributors. It was really hard to get quality or a timely response. It's like anytime we wanted to make a simple change to the photometric and go from a three-fixture layout to a two-fixture layout, it seemed like the time to get that sort of simple type of tweak done to the photometric plan really added a lot of time. But in the end, Cooper Lighting and their local rep really stepped up, and that was a big factor in our decision making process was just their responsiveness and willingness to work with us and our quick schedule. I also want to say that I think applying this specification or any lighting design is significantly more complicated in a retrofit or redevelopment situation. We had several different issues that came up throughout the project that just slowed this process down, everything from having misinformation about the voltage on this system to not being sure if when we replace the polls the bolt pattern on the existing concrete bases was going to work with the polls. So a number of different factors that complicated this particular process, nothing to do with LED, but just sort of a general observation.
Other factors that were critical in selecting Cooper Lighting. As you can see here, we really liked the modularity of their system. So what you're looking at here are different LED bars, and we actually had to switch from a three-fixture eight bar configuration to a two-fixture 12 bar configuration because of this voltage issue and just the modularity that this system gives us was really useful. And then lastly price. We did finally get pricing from, I think, three of the four manufacturers that we reached out to, and the Cooper fixture had a very competitive price at that time.
Some of the project highlights. We ended up installing 55 Cooper luminaires. They are 206 watt fixtures. I mentioned that we did change the configuration sort of at the last minute. The average footcandle was just under 3.0. The minimum was 0.9 and the maximum was 9.3, so we took that max to min ratio from prior conditions, which was approximately 40 to 1 or maybe higher, down to more of a 10 to 1 ratio, 10.3 to 1.0. When you look at these payback numbers, I just want to, again, highlight the fact that the decision to replace the older technology was a foregone conclusion. So I've excluded the cost of the poll and the labor from the payback analysis because this is really a sub-cost. And what I've done here is I've compared both the energy savings from the existing baseline 1,000 probe start metal halide to the first design option of pulse start metal halide, and then our LED. You can see that our energy savings was going to be significant in either case, about 59% with the pulse start metal halide, and what we estimate to be close to 75% with the LED fixtures. And then if you look at just the incremental savings to go from the pulse start metal halide, which I'm comparing here to the LED, starting to serve an additional 40% incremental savings to use the LED. Again looking at payback, excluding polls and labor, you had about 3.5 years for the LED. Again, we were paying about $0.095 per kilowatt hour. Another big factor in this is we did, and we still do, run these lights from dusk to dawn, so you had long hours of operation which helped the payback. Interestingly though, if you look at the payback compared to the initial design to the LED, obviously the increased first cost of the LED meant that the incremental payback for that technology was quite a bit longer compared to the 400 watt pulse start metal halide.
I'll come back and make some final comments on this project. I did just want to quickly touch on one other project. This is another site that was originally developed in 1965, Grenada Hills, California, just outside of LA, another redevelopment that we engaged with Pacific Northwest National Lab and the Department of Energy to look at applying the specification, the LED specification. It's currently undergoing a redevelopment and again, replacing the lighting fixtures was a foregone conclusion. So we looked at this site as a possible candidate for the LED replacement technology, and ultimately concluded that it was not a good fit. The existing technologies were also 1,000 watt probe start metal halide cobra fixtures, and here is a perfect example right there. You can just see huge differences in light output. At least within our portfolio, we had about a 1.5 footcandle average. Really old technology. Really inefficient. A lot of trespass. Presumably really high max to min ratio with 40 foot polls and the space in between the polls was quite significant. As I mentioned, we considered both LED and pulse start metal halide. The poll height of 40 feet, as well as the spacing between the polls, really limited our ability to use the LED fixtures. And on a first cost basis when Michael showed his pricing slide, the incremental cost to get up to the lighting level that we desired, the cost for LED nearly doubled, where to go from like a 400 watt pulse start metal halide to a 750 watt pulse start metal halide was a nominal increase. So from a first cost level, it was very, very difficult to compete with the higher wattage conventional fixtures.
So in summary, some of the lessons learned and just our conclusions at Regency, really a lack of industry consensus on light levels. Obviously everyone's concerned with security and want to have a center that's well lit, but you see all these different discrepancies and then you have the conflict of what the jurisdictions requirement, make it really hard for us to develop a national standard or specification. Another thing is managing perceptions about LEDs. Just some of the literature out there talks about just the great coloring. Well metal halide is also a very white light source, and I think some stakeholders at Willston II really expected to sort of drive down the street and see that property just glowing. The reality is we have, like Michael said, the response has been neutral I would say, which is a positive thing. No one's complaining that it's too dark, but there certainly was that sort of misperception that this property was just going to stand out and be brighter than all the adjacent properties.
I talked a little bit about the design process and our frustrations with the procurement and just some of the inefficiencies that exist in the value chain, and something that I think we should all be interested in seeing improvement on. LEDs are more feasible with lower light levels, shorter poll heights, and are shorter distances between polls. And I already made the comment about it's very difficult to compete on a cost basis with higher wattage and metal halide fixtures. One of the things that I wish we would have done differently at the Willston II Center is integrate lighting controls. Once we made the decision to go with LEDs, our budget was pretty much exhausted and then just based on our timing, I really didn't have the resources to really think about what kind of control system do we want to use on that sight, whether it be dimming or motion sensors. That's a huge area of focus of ours now, and our retrofit program is looking at how can we not run from dusk to dawn on all these properties and I think there are huge cost savings opportunities with lighting controls. And we'll continue to consider using LED and the DOE Specification, but first costs are still a big driver in our decision making process.
Michael, I'll hand it back to you.
Michael A. Myer: Thanks, Mark. So I received an email during your presentation that we have a number of questions, so we'll make sure we have enough time for that. The last couple of slides are high level anyway, so it's really easy to go through them. And again, as Terry said in the beginning, we are going to make this presentation available in via PDF in the near future. Here's a quick link to the specification. I'm going to leave it up here for a second more in case people are not familiar with where to find it. This is a cover page of what it looks like. I did include PNNL, which is Pacific Northwest National Lab; hopefully at this point you know who we are. We are managing this for the Department of Energy. Linda Sandahl's contact information is here. She is the overall Program manager, and I am the technical lead, as I mentioned earlier. So if someone has a technical question or if they wanted to propose a modification to the specification, it's not just me making the changes, we do have a group of end users who hopefully can evaluate it and say, "Yeah, we think that's a good change." The specification's been vetted a couple times before manufacturers, so we didn't want to just have it without their industry opinion on it as well.
Other programs I'd like just to mention quickly, so as I said that the Commercial Building Energy Alliance is made up of three smaller energy alliances. Here are the links to those energy alliances in case you're interested in knowing more about any one of those specific energy alliances. And then because we have been talking about LED parking lot lighting for an hour now, I just want to mention DOE's SSL Commercialization Web pages in case someone's interested in more information. The homepage links to everything else, so if someone's interested in knowing more about CALiPER, as I mentioned earlier, you can go to the CALiPER Website and download information about products that have been tested and see how they faired against what they claimed. Also how they compared against other tested products. It is not entirely comprehensive because it's just impossible to test that many products in any given time, but it does give you a sampling of what the industry looks like, not only as currently but as over time. They have great graphics that show since 2005 or 2008 how products have been trending. Improvement is the short answer, but you'd see in what ways they've improved.
And then the final slide is more different Websites that I mentioned. Standards of Development, there's a lot of new LED standards coming out. The GATEWAY Program, which is what a number of the installations that you've see are part of. These are real world installations that you can read all about. A lot of different things there. And then Next Generation Luminaires is also a luminaire design competition. The 2010 winners are going to be announced shortly, and this program looks at not only aesthetics but output and serviceability and other aspects of it. And that GE product that Walmart used was a previous winner or a recognized status. I don't entirely know all the distinctions.
At this point, Eric Richman from PNNL, is going to come on and moderate some questions that we received during the call thus far, and Mark and I will attempt to answer them in the remaining time period.
Eric E. Richman: Thanks, Michael, and thanks to both Michael and Mark for the wonderful information they've provided. As Michael mentioned, there are a lot of questions. We'll try and go through these as quickly as possible to answer as many as possible. I'll try and direct them where I think they go, but Michael and Mark, please feel free to chime in when you have some responses.
The very first one we got about measuring footcandles, vertical footcandles in a parking lot, the simple question was: Do you still hold your mirror horizontally? Michael, do you want to provide a little more to that answer?
Michael A. Myer: No, so depending on how you're taking the measurement. You don't hold it horizontally or parallel to the ground. You actually hold it vertically, and we've actually discussed doing another document that actually talks about some of the verification stuff. I actually developed, because I take a lot of vertical measurements, I developed a telescoping pole from a painting handle and allows me to go up and down and it's on a base. But no, when you take vertical measurements, the meter should be perpendicular to the ground surface, and there are specific heights you want to take them at. The IAS talks about them as well. You want to be at a certain height based on an average viewing position of a car. If you take them horizontally above grade, you're just taking a horizontal measurement four feet or five feet up, or whatever height up, and while that is informative, the actual true vertical ones are sometimes more informative.
Eric E. Richman: Thanks, Michael. Here's another one I think you can answer on the spec dealing with the four zones. The question really is, there are a couple of them: What was the basis for the zone 0 to 4 densities and the associated illuminance requirements, and can you provide a little more or provide a reference to where you can get more detail on what's actually included in zones 0 to 4?
Michael A. Myer: Well the specification does, I believe, have an appendix that outlines the actual definitions of them. And what the rationale there is, as I said, in the turn of the new millennium, everyone kind of started realized that one size doesn't fit all, and so the IES and different agencies started looking at it from different points of view. So one thing they did is also the IES changed their ratings. A lot of people are familiar cutoff classification. Well those have been superseded with what we call BUG ratings, and now a BUG rating refers to backlight, uplight, and glare. And so now those are starting to say well if you're in a certain environmental zone, less uplight is permissible, but if you're in environmental zone 4 which is essentially New York Times Square, that much more uplight is not as much of an issue. So the industry, whether it be the lighting industry through the IES or energy efficiency programs all started realizing wait a minute. We don't need as much light. If you think about Standard 90.1-2007 and earlier editions, they allow 0.15 watts per square foot per parking lot, and so you have anything from a small, very small, fast food place out in the middle of nowhere to a very dense place in downtown Los Angeles all getting the same allotment, and it's really not as needed. So because a lot of people are moving that way, we tried to harmonize with them. So our definitions look at similar definitions whether it be LEED, whether it be the International Dark Sky Association and the IES drafted a model lighting ordinance and we looked at their definitions, and we looked at what ASHRAE did. We also looked at what TEL 24 did, and we tried to harmonize with their approach and also their definitions.
Eric E. Richman: Thanks, Michael. Here's a couple for Mark about the information he presented. The first one: Please restate why the California project didn't work for Regency?
Mark Peternell: Sure. So the alternate technology, the technology that we ultimately went with, was a 750 watt pulse start metal halide mounted at 40 feet, and the polls are a combination of triples and doubles, I believe. To get not even the same light level, but to get to the light level that we desired with LED was going to require significantly more fixtures. And so the first cost premium to go for the LED based on the parking lot configuration at this particular site was too significant to overcome and incorporate the LEDS.
Eric E. Richman: Okay, thanks. A couple more for you, Mark. One here: In Regency's experience, any problems getting replacement parts?
Mark Peternell: We haven't gotten there yet. I assume you're talking about with the LEDs. This project is our first, and installation was completed last fall, so to this point, no, we have not had any issues.
Eric E. Richman: Okay. Another one: Any less of a cost done on your LED installations? Any results of that to share?
Mark Peternell: Yeah. We did, with Michael's assistance, we did sort of a very detailed analysis looking at all three systems, the baseline system, the proposed design alternative one with the 400 watt fixtures and then the LED fixtures, looking at both historical electrical costs and maintenance costs, and then projected maintenance and energy expenses going forward. So yeah, we did do a detailed analysis, and I don't know if there's anything in there that would be proprietary or not, but if you want to follow-up with Michael another time and we can see if there's something we can share publicly.
Eric E. Richman: Okay, great. Here's one, kind of a general area I think both of you can address, having to do with security needs and outdoor lighting. The question really is: How is the need for security at perimeter areas balanced with potentially less overall use, reducing for security issues? Mark, if you maybe could provide your perspective from Regency, and then I've got another question I'll ask that relates to that too.
Mark Peternell: Yeah, well and I completely agree with the diagram that the - - what the specification, in fact, that the most important place for us to have light is in front drive aisle and the parking spaces closest to the storefront. And I think the idea of having lower light levels in the perimeter in the far reaching areas of the shopping center, of course, needs to be balanced with security needs, but I think the IESNA recommends a minimum of 0.5 footcandles. And we're looking at the footcandle averages in the perimeter areas, even though they're decreased, they're still well in excess of the recommended minimum security values. So that's really, I think, something that as an industry we should all be looking to do a better job of is really yeah, if you want to have 5.0, 6.0, 7.0, 8.0, 9.0, 10.0 footcandles in the front part, in the high traffic areas of the shopping center, that makes a lot of sense. But certainly, there's opportunities to decrease that in the rear and the far reaching areas of the project.
Eric E. Richman: Okay, thanks. And a follow-on for Michael. A couple items here on… The question was: Is there a lumen limit on LEDs, speaking to the issue of can you get the light levels you really need if a site owner wants more, and then specifically, what footcandle level was achieved when the T.J. Maxx lighting was dimmed?
Michael A. Myer: Oh, okay. Is there a lumen limit? Well right now, I would definitely say yes, but with conventional technology there's a lumen limit as well. I mean metal halide caps the lamp wattage… I mean you can get beyond 1,000 watts, but they're very specialty lamps usually for sports lighting, so they cap out at about 110,000 lumens. For LEDs, it's much different depending on how you do… So let's pick on the Cooper product that we talked about. So they designed theirs in an LED array, so at some point… There's 20 LEDs per array, and their fixture can only tolerate up to 12 arrays, so you have 240 LEDs per array. And so as Mark said, one of the problems we had is we had originally tried to go with a 12 LED array, but because of the voltage, we went to 480 volt, there was a glitch and they couldn't go with a 480 volt at that time so they had to go from a 12 array from two 12 arrays to three 8 arrays. So total math, 2 times 12 is 24, 3 times 8 is 24, so virtually the same. But going back to my earlier point, LED arrays are getting more efficient over time, and therefore, more light coming out. But yes, for at least in the near term and possibly the long term, there will be an overall limit because you can't just keep growing the fixture. It's going to get a point where your effective projected area becomes so large that it becomes a wind issue or a cost issue. Also, there's some driver issues that the more LED arrays you have in a fixture, the more drivers you need and other stuff like that. So right now, yes, there's probably - - I don't - - I'm not doing the math in my head right now, but there's probably a limit to the amount of lumens you could get out of one LED fixture. That being said, there are things you can do - - you can do things like drive current. As you increase the drive current, you can increase the light output of the LED. So if you really wanted a lot of light, the easiest thing I could say is find a manufacturer that can jam a lot of LEDs in a fixture and that's willing to operate it at a high drive current. That will give you a lot of light. It does affect the life output because the more current you drive through the LED it is going to affect life.
In terms of the T.J. Maxx, the average was… I'm going to pull up the slide so I can just verify where I'm… So this is a great example of drive current. So these LED fixtures operated at 525 million amps when they were operating their high output, which is what this number was, and they operated at a 175 million amps when they were operating in their low setting. Because, as I said earlier, because you're operating at a lower current, the LED actually gets slightly more efficient, and so it's not a direct comparison. You can't just say take 175 over 525 and say oh, it was this much of a reduction. It's close to that, but it's not - - it's somewhat off. So speaking without my entire notes in front of me, I believe the average was just under 1.0 footcandle or near their 1.0 footcandle is the story.
Eric E. Richman: Okay, thanks. Another quick one for you: Are the power density requirements averaged over a complete nighttime cycle or instantaneous maximums?
Michael A. Myer: Well power density is regardless of uses of time, it is the number of fixtures installed per area. It doesn't matter if you use all the fixtures at once, or you turn off half the fixtures over time. Ideally, it would be great to go to something like a kWh-based metric, but of course, there's a lot of limitations in that in metering and other stuff, but some type of metric that actually looked at time component. I'm a big advocate of time components in energy codes, but they're very hard to implement for a variety of reasons, both on the code side and both on the assumptions side, but that would be the best way to do it. So no, the power density is literally looking at everything that's installed on your plans day one regardless if you're operating all those fixtures all night long or only half those fixtures all night long.
Eric E. Richman: All right. Thanks, Michael. A couple more for Mark. One here: Did you factor in the cost to replacing the LED luminaires at the L-70 point versus mirror lamp replacement for metal halide in your maintenance counts?
Mark Peternell: What… I didn't… I guess it's hard to get a clarification. I didn't understand the comment about L-70.
Eric E. Richman: Well basically it had a point of 70% light output.
Mark Peternell: Oh, okay. Michael, I'm probably going to need your help on that. I don't remember what we did from a replacement standpoint on the LEDs.
Michael A. Myer: Yes. I think we did look at what the replacement point will be. I don't remember if… Also, L-70, for those people on the phone who are not familiar with it, is the current used definition for the end of useful life of an LED fixture. This L-70 refers to 70%, or in this case, a 30% reduction or 70% of full output at a certain point in time. We did look at somewhere down the road that they would need to be replaced, ideally wholesale. I don't know off hand at what point in time we estimated that point was going to occur. I don't have my economic notes in front of me.
Eric E. Richman: Okay. And another one here for Mark about the Grenada site: Was induction technology considered for that project?
Mark Peternell: No. And I don't really have a good answer why, but just we're really focused on pulse start metal halide as our primary design option, and then looked at LED as an alternative.
Eric E. Richman: Okay. There were a couple of other questions that came in about induction. Michael, do you have anything really quick about why or anything else about induction you might want to share?
Michael A. Myer: Well so the reason why induction often comes up with LED is that if you're deferring maintenance, why not also look induction which has a long-rated life. So first, the specification was born out of a group of energy users who wanted LED. They didn't ask for it to be technology neutral. They didn't ask to include induction. They actually limited it to LEDs, so it was not a Department of Energy decision. It was a group of end users part of the Energy Alliance that said, "No, we just want to look at this technology." In parking structures, we include induction. Some of their concerns with induction really came down to, as Mark - - one of the reasons why he went with Cooper, as he said, was the modularity, which is the same with why we like metal halide. "Oh, I like 175, but it's not enough light," so you go to the 250. You go to the 400. You have a lot of similar steps. The concern with some of the Alliance members of induction had to do with one, the fact that there's - - certain manufacturers offer it, but not a lot of manufacturers offer modularity in lamps. So they have certain packages, and then that's it. Another concern with some of the manufacturers - - or members about - - induction members - - induction technology had to do with optical design of the fixture in that depending on the induction lamp, it is a large source, optically speaking, and therefore, it is hard to craft good optics around it, as well as you can lose some of the light absorbed in the fixture. So those are some of the reasons they chose not to go down the road with induction.
Eric E. Richman: Okay, thanks. Question here about occupancy sensors, kind of a multipart: What is the real expected life of exterior occupancy sensors? Will they last as long as the LED fixtures? And when using these sensors in parking lots, how are the minimum light levels maintained?
Michael A. Myer: All right, so let's do light first. So that's a great question. I don't know… I'm going to actually take that as a no. I don't know if we looked at the life of the occupancy sensor. I mean it probably comes down to what the actual technology is and how it's working. I do know that photo cells, conventional photo cells, even if they're on - - whatever fixture they're on, they have their own life, and they often fail differently than the fixture, and so that's something you have to factor in. But I don't know exactly if there's even an industry standard of assessing occupancy sensor life, interior or exterior. I do know that we do have occupancy sensors outside. It should be the similar technology or idea to… My parents have one on the side of their house that turns on a light when you walk by that side of the house. So we've had outside occupancy sensors for a long time. I don't know if anyone's actually ever said, "Well how long do they last". Moving them onto a fixture, I also don't know if they're entirely the same type of technology, but again, we've had them outside for so long, somebody, hopefully, has looked at that problem.
The other one… Oh, so how are the minimums? So minimums in a parking lot… Minimums are interesting because some of the problems with the minimum come down to just geographic location in the way the parking lot's laid out. Sometimes it's really hard to meet that minimum in that one single value, and so you end up over lighting for that. In this case, when the occupancy sensors went down to their lowest point in T.J. Maxx, they still met the minimum everywhere except one point in the entire parking lot after we factored in this L-70, as I said. And that really came down to this pure geometry issue in that the one spot, they just couldn't meet it and they virtually met it. It was just the way the parking lot took a turn, and we kind of said, "Okay, that one spot is very close to what we required. We'll just consider that okay." But it was an assessment we did make for that one spot.
Eric E. Richman: Okay. We're almost out of time. Time for one more question. I've got one here for Mark related to warranties: In your experience, how did that go with addressing warranties with manufacturers? Did you have any problems? Good experience? Bad experience? Mark?
Michael A. Myer: Mark, I know has an adjacent schedule.
Mark Peternell: The warranty question, we'll see ten years from now.
Eric E. Richman: Okay. But did you have any problem with them agreeing to the warranty?
Mark Peternell: No. They agreed to the warranty that was in the specification.
Eric E. Richman: Okay. We have several other questions we couldn't get to. Terry, is there a protocol for addressing these?
Terry Shoemaker: Thank you, Eric. Early next week, we will post the presentation, and we will address all of the questions that were not talked or responded to on today's call, and we will have a Word file out there with all the questions and answers.
Terry Shoemaker: Thank you for participating in today's Webcast brought to you by the U.S. Department of Energy. You may all disconnect.