U.S. Department of Energy - Energy Efficiency and Renewable Energy

Building Technologies Office

CBEA High-Performance 2'x2' Troffer Specification (text version)

Below is the text version of the webinar titled "High-Performance 2'x2' Troffer Specification," originally presented on July 19, 2011. In addition to this text version of the audio, you can view the presentation slides and a recording of the Webinar (WMV 10 MB).

Linda Sandahl:
And all of you know that we now have high-efficiency 2'x2' troffers back. That was finalized at the end of June, so we're real excited about that. I'm going to turn it over to Jeff McCullough, who was the technical lead on that spec, in just a minute, but first, I want to go through just a few things to keep in mind for the webinar. And did someone else just join?

Kristen:
Yeah, this is Kristen Sandealy.

Linda Sandahl:
Oh, hi, Kristen. We are just getting started. Well, first of all, if you could put your phones on mute, unless you have a question, that would be great. That way we won't get any background noise. But Jeff assured me that he would like to take questions during the presentation, so feel free to take your phone off mute and ask a question when you want to. If you use the F5 button, you can enlarge the image in Live Meeting. So if it's small right now, you can make it larger, and you can also download the PDF directly from Conference Plus if you go up to the little yellow shared notes icon up in the right-hand corner.

A yellow box. We'll also post this on Plone after the event. And a couple more people just joined. Would you like to announce yourself?

Jason:
Jason here, In and Out.

Linda Sandahl:
Hi, anybody else?

Brian:
Brian Delfin from Walgreens.

Linda Sandahl:
OK, anyone else?

Rick Gabriel:
Rick Gabriel, Health Care REIT.

Linda Sandahl:
OK, Rick. And when you dialed in, you heard we are going to record the call. That's going to help us capture questions at the end, and we're also taping this so that we'll have a WMV file, so if there's anyone in your office that didn't have a chance to listen in, they can listen later when they get a chance. And I think that's all that I needed to make you aware of before I turn it over to Jeff.

Jeff McCullough:
Very good—Jeff McCullough here. I offer my welcome to you all. For some of us, it's still West Coast time, so I'd say good morning. For a lot of you, it's East Coast, so good afternoon. What I'd like to do today, and I'll go through the topics here shortly, but I very much would appreciate a two-way dialogue. We're a small enough group that we can kind of roll up our sleeves a little bit. I assure you, some of the material that I have put in front of you today will provoke some reactions from you, but I'm more than happy to roll up our sleeves and chat about them and why some of the decisions as far as—especially as we talk about rated life vs. lumen maintenance.

So I know that one will be one that will get your attention. We're going to go for about, oh, 30 minutes. I recognize your time is valuable. I don't plan on doing a death by PowerPoint, so we'll go for about 30 minutes with slides and an overview of the presentation itself, and then we will have a couple of questions for you, and that'll be your opportunity to kind of chime in and let us know how we can help you, how you can see the specification used in your facilities and in your projects, and then we can just have an open dialogue at that point.

I don't see it going a full hour, but we certainly can go up to that time if you like. So with that said, let's get started, and I've advanced to Slide 2 for those that are not on the Live Meeting. Here are today's discussion items. We're all alliance members, so I don't need to give you too much background on who we are and what it is we're trying to do, but we are going to then move into what's the installed base of these products, and with that, there'll be a little bit of a primer on the various types of 2'x2' troffers that are out there.

If you're a lighting geek, you probably walk into your stores and look up and know what these things are, but we'll take you a little farther so you fully understand why we might choose one type over another. I'll give you some insight as to how the specification was developed and how we went about establishing some of the key metrics associated with it. We'll talk about some of the CALiPER results, and I think all of you are familiar with CALiPER, but it's the DOE's Solid-State Lighting program, commercially available LED product evaluation reporting program.

I think that most of you will know that CALiPER goes out, purchases products on the open market, tests them, and makes them available to anybody to see what the actual test results are. So we've used some of the CALiPER results, and not only is it kind of crafting the specifications, but also in doing some of our benchmarking activities. We'll then talk about the specification details, some thoughts on DOE's next step now that we have a final version of the specification. We have some ideas on how we might leverage that and perhaps expand it.

Then the Q&A will follow thereafter. So with that started, let's get going. OK, so by virtue of the title, we are talking today about 2'x2' performance-based troffers. What's kind of interesting about this project is it tends to be crosscutting, so we have all three of the alliance membership involved in it. We all have these 2'x2' products in our facilities or our buildings in one form or another, so it crosscuts across all of them; it was requested by alliance members.

You know, recall there's been several calls for in what would you like to see work on. This one worked its way up to the top—about last, oh, Linda, I would say it was September or October. We formed a committee, and Kevin Powell of GSA is the chairperson. He's not able to join us today, but he is the chairperson that essentially has led the effort. DOE via PNNL has been working behind the scenes, crunching the numbers, and kind of running things as we move forward.

So the committee kicked off in November. Here we are in the middle of July. We have a final version. It's been vetted by the alliance membership. It's been vetted by industry, and it's being shared more broadly, so it is initially an alliance request. It is supported by utilities and EEPs, or energy efficiency programs, and we'll talk a little bit here about the design by consortium, and what role that will play. But it's clearly something that we're setting a performance level that's quite high and quite challenging, and it will be available for incentives through your utilities as they kind of join on board, particularly under the banner of the DesignLights Consortium.

OK? So let's talk about that, and the reason we bring up the DesignLights Consortium. I take most of you are aware that the ENERGY STAR® program, for the most part, covers residential luminaires and residential fixtures. There are some commercial products in there, but for the foreseeable future, EPA does not intend to develop additional specifications for commercial products.

So filling that void for however long necessary is what's called the DesignLights Consortium. It's based up in the Northeast with the Northeast Energy Efficiency Partnerships. There are quite a number of market transformation groups across the country that support the DesignLights Consortium, either directly or indirectly, and what they do is they've developed not only criteria, and I'm careful to make the distinction between criteria and specifications, and we'll get into that here a little bit later. But they've developed performance criteria for, as of today, 11 different categories.

As of tomorrow, they are going to expand them to 19. So some of the information I'm sharing on the DesignLights Consortium will be expanding as they roll out their next revision to their criteria. And you can see the link there, http://www.DesignLights.org, that you may want to look at here at some point.

OK, let's shift gears here, and let's look at what's the installed base of 2'x2' troffers. What's interesting about this is that the manufacturer's industrial reports used to track the products made in the U.S. Not only lighting products, but all types of manufactured products. They ceased to capture that information in the late '90s, so as far as what percentage of the installed base is 2'x2', 2'x4', 1'x4', and various other luminaires, you can imagine that back in '99, the proportions were about the same. But now, of course, the number of installed products has changed, and then it's shifted from, say, older technology, T12, to T8, and now LED.

We do have a new source of information, and that's what this slide shows, and that is from the Niche applications document that the DOE provides for the Solid-State Lighting program, dated January 2011. And from this, it's a snapshot of where the 2'x2' troffer technology lies. So you can see there, and it's kind of surprising to me, actually, that 17.3 percent are still U-shaped to T12 technology.

I would have thought that just about all, if not all, were at least T8 or some of the PL-based products. But there are apparently some jurisdictions, some states that allow T12 in their mix. But by far and away, T8 is the predominant use, about 75 percent. There are other fluorescent versions, some of the pin-based products, for example. And as of January 2011, about a tenth of a percent are LEDs, or about 50,000.

Francis Rubinstein:
Jeff, are you taking comments from—

Jeff McCullough:
I sure am. Please, yeah.

Francis Rubinstein:
OK, Francis Rubinstein here from LBL. Thanks for putting together this information. A question about the other 2' fluorescents, 8.5 percent, and that's 4.5 milli-units. Would that include like T5s and the PL, the long PL lamps and stuff like that?

Jeff McCullough:
That's my—this is actually a great point. I probably should have rolled up my sleeves a little further on this, but frankly, I would have expected more of the T5 and PL—

Francis Rubinstein:
Me, too. Me, too.

Jeff McCullough:
A higher percentage. All I can draw upon because I have to go back and look at that document and see who was it that put it together and over what time frame. As I mentioned kind of at the outset, the manufacturer's industrial reports that used to be required for luminaires ceased to capture the data back in the late '90s, so I'd be surprised to learn—or not surprised, but I'd like to learn, you know, where the contractor for DOE found these numbers. I don't know if they extrapolated or if they did some surveying, but traditionally, it's kind of difficult to give this information.

OK, let's move on. Bear with me here a second. OK, so let's talk a little bit about the energy savings story. This is just some simple math if you assume your spaces or offices are occupied ten-and-a-half hours per day. There's a range of input powers for these products. It depends on lamp type. It depends on ballasts. As you know, you can reduce ballast factor. There are various permutations of lamps that are available.

For this particular number, I just grabbed what was in the middle, so about 59 watts, times the number of installed fixtures, and what we're talking about is about 12 terawatt hours per year of consumption. Certainly, that is significant. There are potential—and via the spec, where if we set a level of performance that is 20 percent better than the average of the incumbent technologies, obviously, you can see 20 percent savings from that number. And that's just on one-for-one basis.

If we start to integrate controls, daylighting, occupancy sensing, those types of things, we can do far better than that, and achieve up to 50 percent savings. So on a simple one-for-one basis, we're targeting—excuse me, 20 percent, but we can go up to 40 and 50 percent by using controls, and that's one of the benefits of this spec, and that's one of the things that the committee really wanted to move forward with was including the requirement or the—including the capability for controls and/or dimming.

OK, so with that said, let's take a few slides and kind of look under the hood as to what are these troffers, what permutations are out there, and I guess what I should start out with is one of the activities that DOE undertook at the very outset was to go out and do benchmarking. So we went out and did a significant amount of data mining of the incumbent technologies. And so from that, we were able to develop all kinds of statistical analysis as far as what's the range, what's the median, what's the model—those types of things.

And we used that, the committee used that in establishing what the criteria ultimately would be. So there are three—one could argue three or four main types of troffers that we see in our buildings. This one is probably the most common. It's a bit dated. I don't think it's the most prevalent, but nonetheless, they are fairly common. This is a lens troffer. It's typically a prismatic, fully enclosed, semi-opaque lens. Fixture efficiency ranges anywhere from about 65 percent to 90 percent.

You can see there's a range of lamp options, and to Francis' comment, there are some T5 HOs that are certainly in that range, and some of the brightest products would be F17T8 is a 2' linear product, and then you have the U-bend T12—sorry, T8F32s. Again, because we don't have the most current proportion, you know, we subjectively, based on our knowledge of the industry, are going to assume what we know that there is certainly a pretty good portion of these products out there, but we don't have direct shipment data to report that.

OK, so moving on from lens products to—or I should say prismatic lens to parabolic troffers. These guys are fairly common. They were fairly prevalent in the late '90s and the 2000s. They're still used quite a bit today. Efficiency ranges, at least from the products that we surveyed, 50 to about 72 percent. Pretty much the same lamp combination that you saw with the fully lens prismatic products. And of course, they are parabolic. They tend to give us more of a bat-wing-type distribution, and we'll look at that here shortly.

You can order any range of cells, anything from 4 to 16 to ice cubes. Ice cube trays, which would be probably 64 or so. All kinds of those products out there, and we believe this will be the majority of installed 2'x2's currently out there. OK, so moving on from that, let's go to some of these we call a basket or indirect-type troffers. The efficiency range is much broader, as you might imagine. You have fully veiled lamps at this point. So ranging from 33 to about 80 percent.

Similar lamp permutations from the prior two, and the baskets tend to either be perforated—as you can imagine, there were a bunch of series of holes that allow some diffusion of light, but they tend to take advantage of the reflectance off the ceiling, if you will, off the top of the luminaire, and be more diffuse in their distribution. There are types that are side mounted and center mounted. I would argue they are becoming much, much more prevalent and much, much more popular. So while it's probably a small portion of installed products, it'd be very interesting to know what the current snapshot is.

I would suspect that they are certainly approaching rivaling the parabolic that we saw on the prior slide. OK, and then kind of an iteration of that, and this is where we can argue are there three main types or four, are this kind of new generation of high-performance troffers, tend to be a little bit more optically efficient. So you're seeing here about 67 to 90, 91 percent fixture efficiency. Using some of the premium lamp options, they tend to give you—it's got kind of the best of both worlds.

You get not only that diffusion that you might see in an indirect, but because of the lamp spacing and location, you also get a little bit more direct emission, and so the overall fixed efficiency moves up a little bit as well. OK, so let's now take a look at how we began the process. So what you have before you is a table that was developed that's based statistically on all the products that we have surveyed. And so on the right-hand side, you see fluorescent survey data, and those are the typical statistics you might expect.

You know, so this is what you're looking at from an initial lumens, and by the way, this is luminaire lumens—this is not the lamp. This is net lumens from the luminaire. Minimum of 1,176; maximum, about 88, 90—a pretty serious fixture. But the one that occurred the most often, and that's the mode, is about 3,000 lumens. The mean, about 2,885. And the reason we're showing the DOE alliance specification and contrasting it a little bit with the DesignLights Consortium-qualified product list is just to show you where the thresholds are set.

So the DOE criteria for LEDs is 3,200 initial lumens. The criteria for the DesignLight Consortium that should be released tomorrow is 3,000 lumens, so they are very close, but the DOE one requires a little bit more initial light up. Black light is on the luminaire efficacy. Our criteria is set at 69 lumens per watt. The DLC, as of today, is 55. As of tomorrow, it will be 60. They're not coming up farther for their initial entrance to this category, but nonetheless, the specifications can be used and some of the resources used within the DLC, you can search their database. You can use the Lighting Facts website and seek out products that meet both the minimum light output of 3,200 lumens and the minimum luminaire efficacy of 6,900 lumens. Oh, I'm sorry, 69 lumens. OK?

So now is a good point to share with you, how did we arrive at the 3,200 and the 69 lumens per watt? So you can envision after doing all of this data mining, we have deployed a methodology, and the committee decided that the best way to go about this was to do all this data mining, establish the performance of the average, in this case, fluorescent luminaire, and then make the conscious decision that the specification would target those products that were at least 20 percent greater than the average. So that was the starting point, and these numbers reflect that, for the most part.

It certainly goes without saying that we also surveyed the current solid-state lighting products on the market to see, well, are there products that could meet this specification? Let's look over the horizon and see what's coming out over the next three months, six months, nine months, and lo and behold, I think some of us are aware that there has been a whole slew of 2'x2' products that have come out, oh, arguably in the last four to five months that are significantly better than even a 69 lumens per watt that we're setting.

So there are indeed a significant number of products that can now meet this specification. The next slide, and this is kind of just an addition. So in addition to going out and surveying the manufacturer specifications and going over spec sheets and IES files and gathering all that data, we also have the benefit of the DOE CALiPER program. Well, we've actually gone out and done our own testing on products, and these are representative of those products that we had surveyed.

The way the nomenclature works, so if you look at CALiPER number, the first two digits are what round of CALiPER testing. So the first one is '09, so that's the round nine, and then it's luminaire 81. If you wanted to, you could go to the DOE website and actually download that actual product's IES report and/or test result, so you're welcome to do that, but this was just added information of independent third-party testing that tended to validate what our assumptions were, as well as help us formulate what to set the level's performance for.

OK, this next slide shows us—basically, it's all the data put on one slide. And so on the horizontal axis, you've got initial luminaire output, so that's net lumens from the luminaire. On the vertical access, you have luminaire efficacy, or LER. And you can see by the keys here the diamonds are basic catalog mind data. The shaded squares are LED catalog data, and the triangles are CALiPER data. And so I guess what I would draw your attention to, they seem fairly tight. For example, if you drew a line across this at about 70 lumens per watt, you see that we're capturing arguably probably the upper 10 percent of products, and there are indeed several LED products that meet this. Likewise, if I drew a vertical line at about 3,000 or 3,200 lumens, you can see there's a significant number of products over to the right there that meet the minimum light output requirements.

So this is just useful to kind of see well where is the industry—what's the range of performance, and indeed, there are some high-performance products that are both fluorescent and LED based, and I should have mentioned early on that this specification is actually technology neutral. There are reasons for doing that. I know a lot of you are LED centric. But there are also some arguments for developing a specification that perhaps a high-performance, high-quality fluorescent product might need as well.

So they are technology neutral. Ultimately, if you begin to use these specifications or this specification, you can remove one or the other if you choose to, so we've left it kind of open ended, and we'll talk about a little later what your preferences might be. OK, let me scroll down further. Let's look now at some of the distributions that are out there. So we've gone over those four types of products that are out there. For the lighting designers or the techies within us, we need to look at the distributions of the light emitted from those luminaires, and so these are some typical distributions that you might expect.

On the left-hand side of the lens, prismatic, the very first one, you know, somewhat of a teardrop shape, if you will. The parabolics give us that kind of a bat-wing distribution, tend to throw more light horizontally, have high cutoff angles, etc., etc. Both the baskets and our non-planar tend to be more diffuse, more uniform if you prefer. And more—and then the term volumetric oftentimes is used to describe this kind of a distribution.

The reason we show this to you is that initially, the committee was looking at how do we specify distribution, and more importantly, uniformity. And that's one of the things that we need to talk about here shortly. And the way that you can get at that is we hypothesize based on the data mining that there are really three different distribution types that could be captured or categorized as narrow, medium, and wide distributions. Ultimately, the committee decided that because of you as specifiers, you as building owners, are going to select products. We didn't want to get into the business of telling a manufacturer how to design or how to make their products, so we put the onus on the lighting designer to grab the photometric report, do whatever design is necessary to meet the intent of the application or the space.

So rather than being prescriptive with spacing criteria, and by the way, spacing criteria is the ratio of space in between fixtures to mounting heights, and that's used directly from an IES report. So if you brought an IES report in there, you would see what the spacing criteria is, and essentially, it tells you that ratio that would give you equal illuminants on your work surface, whether it be on the floor or on a work plane or the desk.

We're also using spacing criteria as somewhat of a surrogate for uniformity. So I can say that if you installed luminaires and you follow the spacing criteria, you would have at least a minimum level of a luminance, especially as it relates to where these fixtures overlap, or as they "butt" each other. What it doesn't get at, and this is kind of a tricky thing, is it says nothing about the uniformity within the beam spread or the beam angle or the distribution itself. So we didn't go that far, but we at least felt that by using spacing criteria, it goes a long ways towards ensuring uniformity of illuminants on the work plane in our spaces.

So the alliance, rather than being prescriptive here, we've actually gone out and specified a range of 1.1 to 1.4, and that's in both planes, the 0 to 90 and 180 to 270 degree plains. I've got the DLC as taking a different approach, and they've actually used zonal lumen densities to kind of get at this. The version of the DLC-qualified product list that comes out tomorrow, this will go away and they, too, will be using spacing criteria very similar to DOE's. OK?

And so just as an overlay, you know, those are the same four distributions we just saw. Here are some of the CALiPER-reported distributions of solid-state lighting products, and you can see they tend to be kind of a cosine distribution or circular, if you will. And so it's important that the lighting designer be equipped with this information as they lay out their luminaires. You cannot just go and assume a one-for-one change out. You need to look at the IES reports. You need to do whatever modeling is necessary to ensure that your system meets the intent of space or the application.

OK, so we've already seen that slide. OK, let's look at the specification itself. I shared earlier that it is technology neutral, either fluorescent or LED, nominal 2'x2' troffer. Here is where it gets a little bit interesting. So we've specified minimum initial lumens, and this is at the luminaire level of 3,000 for fluorescent and 3,200 for LED, and we'll talk about why we've done that here in the next couple of slides.

But we actually have separated them out initially to account for the long lumen maintenance and the issues we have with rated life vs. lumen maintenance that's associated with solid-state lighting. You know that we're setting the bar quite high at 69 lumens per watt. We've chosen all sorts of cap, the maximum input power of 54 watts. The correlated color temperatures, as you might expect, 2,700 to 5,000 K. This particular group, the committee, chose not to go beyond 5,000 K, and they also chose not to include the LED option.

There's another color bin, if you will, another defined CCT of 4,500 K. This group wanted to mirror the fluorescent offerings in their building and didn't want to introduce yet another color into the mix. CRI, so color rendering of 80 or greater, and the requirement of an R9 value. You're starting to see that used more and more. R9 is the red pastel, if you wish.

And so a red content is seen by lighting designers as being important so that we fill out the spectrum. So the requirement here is that the R9 value be greater than 0 at the minimum. We've already talked about the spacing criteria, 1.1 to 1.4, and that was in either the 0 to 180 or the 90 to 270 planes. OK, THD for the products, less than or equal to 20 percent, power factor of greater 0.9, you would expect that for commercial applications.

There is a requirement that these products offer continuous dimming from 100 percent down to 10 percent, and that's considered kind of a midrange dimming. You can go down to 5 and even 1 percent. You can go up to 20 percent. But down to 10 percent is seen as kind of a midrange option, not only from a kind of performance standpoint, but also from a cost and/or complexity standpoint.

We've gone so far as to specify a driver efficiency of at least 85 percent, and just so you're aware, in a solid-state lighting system, that driver efficiency factors into the overall performance of the luminaire. We didn't want to leave it unaddressed, and so the—they decided to establish at least a reasonable performance, based on products currently out in the market.

Accessibility—this was a huge issue, especially as we talk about solid-state lighting. Certainly for fluorescent, you would expect that you should be able to replace the ballast, and likewise, for a solid-state lighting product, the driver should be equally as accessible and replaceable. One of the things that is in the specification is that at some point in time, you will need to replace the light-emitting portion of this luminaire. We certainly don't want to get into the business of disposable luminaires where these things reach 50, 60 thousand hours, and we have to replace the entire luminaire. So the committee wanted to have a product that that light engine, that light-emitting section could be replaced. And when it was replaced, could be done without the use of a soldering iron. As you can imagine, that would just be a disaster to have people up there, electronics technicians, unsoldering boards, etc., etc. So that is a requirement.

Now I'm going to get your attention a little bit. So I've kind of set the stage when we talked about OK, we have a technology-neutral product. For fluorescent technologies, we have a minimum rated life of 30,000 hours.

Rated life for those—I think all of us are kind of aware, but rated life is defined as the point at which 50 percent of a large sample is still operating, large sample of lamps is still operating. And so it literally is a B50. It's a mortality rate or a mortality point. For solid-state lighting, it gets a little bit precarious because while it is possible for them to fail catastrophically, they don't usually do that, especially in a well-designed, well-engineered luminaire, but they do exhibit long sustained lumen depreciation.

So as you think about, all right, how do I equate those two so that we are at least on an equal playing field, the way that we've done that is to use LM80, TM21, and what's called an in-situ temperature measurement test. I'll come to those in a little bit of detail, but not too much because I think most of us have heard some of those buzz words. So with that said, here was the thinking. This is kind of a simple diagram, and what I probably should have done is done it over two lifetimes.

So if I took a look at the lumen maintenance of the fluorescent product, which is the green line vs. an LED product, which is the yellow line, you know now that we've set the minimum initial lumens for the LED at 3,200 lumens, and for the fluorescent 3,000 lumens, it's fairly common practice. So if I have a lamp, for example, that has a 30,000-hour rated life, it's pretty common practice. I suspect in your facilities that you do a group re-lamp at about 70 percent of rated life. So this is at I'd say 21,000 hours. This is what this white dash line represents.

So the reason for setting the LED product a little higher is to hedge our bets, so to speak. What do I mean by that? You can see that about 10,000 to 12,000 hours, the light output is equal. But after that point, the lumen depreciation of the LED is greater than the lumen depreciation of a fluorescent product. What becomes tricky is so if I do a replacement at 21,000 hours or at 30,000 hours, I now jump back up to full light output, to that full 3,000 lumens.

And so you can envision if I had to carry this graph off for another 30,000 hours, kind of a saw tooth function where you'd have a higher sustaining light output with a fluorescent product vs. an LED. And at some point, we may come up with a metric on how one actually kind of captures that, you know, sustained lumens—percentage of sustained lumens over time, or something along those lines.

But the reason for bifurcating the spec was to account for that, and at the point of 21,000 hours, we're about 4 percent off, so to speak. And there's a lot that feeds into this. You know, there are some products that have very good lumen maintenance, an L70 value of 50,000 hours or greater. So this intersection here may not be as steep at that point. OK? So let's press on. So in addition to the parameters that I've just shared with you, there are also lists of options or bid alternates, if you prefer. Certainly emergency lighting was something that the committee was interested in, so two levels were required or options required for emergency lightings.

The committee also wanted all kinds of options as it related to dimming and some controls. So they've asked for analog dimming is at 10 percent, step dimming to 50 percent, and as I shared earlier, dimming down to 20 percent is pretty easy to do at a relatively low cost. Going down to 5 percent or 1 percent is more expensive and more involved. There was also a great interest in protocols, either dolly or some of the wired and/or wireless versions that are out there.

An option for daylight sensing—of course, that would be integrative with your control schemes. Folks were interested in some demand response. There's also language in the specification that allows for what we call centralized tower conversion and/or controls, and I don't want to share a manufacturer's name, but I think a lot of us are aware that there are companies out there making kind of centralized power systems where you can literally have a daisy chain or—it's not so much a daisy chain as it is a centralized power conversion.

And with that allows you to do a lot of control interfaces and control schemes, and so there are manufacturers that are starting to introduce those kinds of products in the market, and the committee wanted to leave the door open, so to speak, for those kinds of products. OK, so I have gone through the specification itself. I would like to talk to you about what DOE sees in moving forward. What I should share with you is that the Commercial Building Energy Alliance has a listing of products where they look at them, they look at their viability, and then these committees are formed to go about developing specifications for them.

2'x2' troffers came up; I suppose it was late summer, early fall of last year. But what's interesting now is within the last six to nine months, there are some new products on the market by multiple manufacturers that now make it viable for 2'x4' and 1'x4' troffers also to be considered and directly compete against the incumbent technology. So and by virtue of the DOE Solid-State Lighting program supporting the DLC, we've done a lot of that same benchmarking I've shared with you today on 2'x4' and 1'x4' troffers as well.

So it seems a logical extension to take those specifications. I should say take the 2'x2' troffer specification and add 2'x4' and 1'x4'. A lot of the language, I think, would remain the same, but some of the initial light output efficacy, some of those metrics would be changed to reflect this additional data. So we're considering doing that and then using the DesignLight Consortium-qualified product list as a resource, and we'll engage them on how they might indicate products that are CBEA approved, if you will, whether that be different coloring.

You can certainly do a search on their website based on efficacy and based on product category and get that data, but that's something that we're going to be looking into. There are also some demonstrations via our GATEWAY program. There's one—actually, there's both of these are GSA sites. One in San Francisco, downtown there, and then a second one in Auburn, Washington. These will be fully metered installations that I believe will be going in this fall with products that meet the Building Energy Alliance specification.

Here is a snapshot of the DesignLight Consortium webpage. It's just http://www.DesignLights.org. We thought we'd go ahead and share that with you folks so you can see where it is. At the top, you can see solid-state lighting. As you enter the website, you can actually take a look at not only their product listing but their different categories, and so with that—I guess I'm going to take a break.

I'm going to say please, if you'd like to turn off your mutes, and we have questions that we'd like to ask of you. Now that you've seen the specification, we've got four questions that we'd like to ask and perhaps have a dialogue with you. So based on your knowledge of this specification, how likely are you to use it? Where do you see it making the most sense? Do you have a preference? Do you have some bias towards LED vs. fluorescent, or vice versa? And then what else can DOE do to support you in providing additional information and/or resources? So this is essentially the point in which we end the conversation, or end the presentation, but take things off mute and have a dialogue. So with that said, I'm happy to answer not only questions over the content provided, but then further on, this conversation.

Linda Sandahl:
Jeff, this is Linda. Thank you. Any questions on content?

John Scott:
Hi, this is John Scott, Linda. If you don't mind, I want to thank Jeff for his presentation.

Linda Sandahl:
Sure.

John Scott:
And Jeff, after reviewing the content, it seems to be a difficult case to make if you have T8s or T5s already in your building or at transition. Is that—am I reading that appropriately, or—

Jeff McCullough:
Yeah, I would say yes, and one of the supporting documents that we've proposed is if I have 2'x2' troffers today, and then eventually, we'll be able to expand that to 2'x4' and 1'x4', you know, what are my opportunities? As a new luminaire, as a retrofit activity, does that make sense? And that's my second question out there is can you actually justify doing that? But if you already have T8 and/or T5 permutations, I guess one of the ways to approach this is does your space, does your building, meet IES guidelines?

If you desire to maintain say 30 foot-candles for your office building, if you're overlit, then frankly, there are a number of options you should be considering. You can go with reduced-wattage lamps. You can go with a reduced ballast factor, or you can consider doing something such as these. So you'd have to take a look at the economic justification and the basis for making those kinds of choices. But if you're using pretty good luminaires that are fairly efficient, meaning optically efficient and electronic ballasts and T8 and T5, you need to crunch the numbers before convincing yourself that LEDs in this application are the best path forward.

John Scott:
Thank you, Jeff, I appreciate that.

Jeff McCullough:
Sure, sure. Does anybody else want to chime in either on content or just kind of start the dialogue on how you see this fitting in your facilities?

Pat Lydon:
This is Pat Lydon at Legacy Health. One of the questions that stand out for me is would we be more inclined to use high-efficiency fluorescent or LED? And I think it somewhat depends on cost, obviously. You know, yes, cost of luminaires. So I'm certainly interested in encouraging all the teams to look at the LEDs and start including those in the considerable—already-considered options, but until we start getting a better idea of how the cost differentiates, not only the initial cost, but the cost of ownership as well, a little tough to answer the question specifically.

Jeff McCullough:
Yeah, I wholeheartedly agree, and that's one of the areas that we're trying to focus. Obviously, with the building alliances, you all collectively bring a lot of information and/or market potential to the lighting industry. So it's our hope that by your using the specification and manufacturers seeing it not only from DOE, but from you including the specs in your RFPs and your design documents, that you know, that will impact the market, and ultimately, that volume, that economy of scale, bring the cost down. I can share with you that there are at least two manufacturers of LED products now that would meet this specification.

And frankly, their price points are very exciting. We're now only talking maybe a $50, or perhaps even less, differential between a high-performance LED product and a high-performance fluorescent product. So it's—those two values, the cost of a fluorescent luminaire and the cost of a high-performance LED luminaire, are indeed beginning to converge, and it's hoped that the economies of scale will accelerate that convergence.

Pat Lydon:
And that implies, I think, the fact that you're pointing that out, it implies that the two LED products you're mentioning would fully meet the DOE spec. In other words, they would meet the spec for no soldering required for light engine replacement and all those requirements.

Jeff McCullough:
That is correct. That is correct.

Bobbi Swatek:
Jeff, this is Bobbi Swatek with jcpenney.

Jeff McCullough:
Yes, ma'am.

Bobbi Swatek:
Hi. How—have you figured in this new issue of the rear earths coming into play and the increased cost of fluorescent, how it'll affect—

Jeff McCullough:
[Laughs] That's a—yeah, this is a great point, and for those that kind of read the lighting press, there is projected to be a worldwide shortage of phosphorous. Phosphorous used not only for fluorescent lighting, but frankly, some of the phosphorous used in LEDs as well, and how that impacts things. Are we going to see a significant increase in lamp costs, for example? Is there perhaps more focus on recycling? I don't know to what extent you can harvest phosphorous.

I think it'd be very difficult to do, but nonetheless, those kinds of things might be considered. So to directly answer your question, no. We can still go out today and get a premium low-mercury, long-life or extended-life, high-output lamp in the three to three-fifty dollar-per-lamp range. And we've not yet seen an increase in cost because of the projected phosphorous deficiency, but it definitely needs to be considered.

Bobbi Swatek:
Thank you.

Michael Myer:
Jeff, this is Michael from Pacific Northwest National Laboratory. Can't recycle phosphor.

Jeff McCullough:
Can't?

Michael Myer:
It is not—it can't be reclaimed. There's technical issues with it. In the recent fluorescent lamps, manufacturers did a very in-depth analysis of it, and they projected with current supplies to be about 20 to 25 or later for the fluorescent lamp industry.

Jeff McCullough:
Wow.

Michael Myer:
But you cannot re—it's not like mercury, which you can reclaim quite easily.

Jeff McCullough:
Yeah, I've been to manufacturing facilities where they crush them up, and I suppose you could separate it out, but the phosphors obviously are blended for certain spectral power distribution, and so I couldn't see how it could be done. I just didn't have any firsthand knowledge of that. Actually, this may be an issue that kind of tips things. We're kind of throwing a line out to, what, 2020 or 2025. You can envision, of course, a fluorescent lamp and the sheer multitude of them need quite a bit of phosphor over a surface area, where as with an LED, you would need phosphor over a much smaller footprint, you would think.

So perhaps this—in addition to mercury and rule makings and the general trend in fixture efficiency increasing over time—this may indeed become a factor as we move forward.

Francis Rubinstein:
[Very distorted audio for entire comment] This is Francis. I had a question about the next sort of globally. [Inaudible] ... The point I would make is that it is costing us to do that with fluorescents, where you can't do that with LEDs today.

Jeff McCullough:
Yeah. Yeah. And Francis, I—my headset, you were very garbled. I don't know if others on the call experienced that as well.

Group:
Yes.

Jeff McCullough:
I'm afraid I didn't catch all of it, but I think what your point was that you took a look at our slide where it showed a range of light outputs for 2'x2' troffers, and you commented, I think, that there was a pretty high range up to about 8,000 lumens, and that for high ceilings, for example, you might find them in jcpenney. You'd have some probably nine cell parabolics, and that getting the light output with an LED, whether it be 8,000, 6,000 would be challenging. And I fully agree.

One comment that I'll make to kind of add on to that discussion, that issue, is that in establishing what the minimum initial light output should be, you know, we've been giving a lot of thought to rather than just benchmarking to say the typical light output of a 32-watt T8 lamp, there is traction. There is acceptance in the industry for reduced-wattage lamps. So in the linear products, you can go with a reduced ballast factor. You can go with a 30-watt, Econowatt lamp. You can go with a 28, a 25, and so with that kind of thinking, I should share with you that our initial threshold was actually 3,700 lumens.

By applying that reduced-wattage, reduced-light output mentality and acceptance in the lighting industry, we ultimately went down to the 3,200 or 3,000 that you see today. So I hope that helps. I'm sorry I didn't get all your points.

Francis Rubinstein:
Sorry about that. Sorry. That's fine. Thank you.

Jeff McCullough:
OK. Well, we've got five minutes left. I've heard people stay on. I've heard people drop off. I just leave it open that if there's anything else that would like to be said, if there's anything Linda, if you want to make any comments or anything like that.

Linda Sandahl:
Yeah, one thing I would like to ask, thank you, Jeff, for the presentation on the spec. I think we all have a good sense for what's included in it. And really, the next step is getting—using the specification and building. So I'm wondering what else we can be doing, DOE and PNNL, to support you. Jeff talked a bit about the DLC-qualified products list, and there will be products that meet our spec included on that list, and that's one area you can go to see here pretty soon which products qualify in addition to the 2'x4' and 1'x4' specs here pretty soon.

Are there any other—are there technical questions? Are there other things that we can do to help you move forward? One thing we are doing is completing some demonstration projects. I don't know if that's going to be helpful.

Pat Lydon:
This is—is this Linda?

Linda Sandahl:
Uh-huh.

Pat Lydon:
Yeah, this is Pat Lydon at Legacy. I'm interested in seeing—you mean like in certain sites you'd have demonstration projects, or what?

Linda Sandahl:
Yeah, there's a couple of GSA sites in Auburn, Washington, and I think Jeff said there was one in California.

Jeff McCullough:
San Francisco, downtown.

Pat Lydon:
I was hoping for Portland.

Jeff McCullough:
[Laughs] Well, Auburn is not too bad.

Pat Lydon:
Yeah.

Linda Sandahl:
Well, and that's one thing that those of you on the phone might want to consider is doing a demonstration and trying these out. I know that was initially what a lot of people did with the LED parking lot lighting spec, was just how do these look in your building? You know, try a few of them out, and—

Holly Northey:
Well, this is Holly with Home Depot. We actually have it installed in a couple of stores in the back of the house, like in our office package and break room and stuff like that. Just in a couple of stores just to test it out, and we've gotten really good feedback from both the stores and from energy savings. So if anybody wants to check them out, they're in California.

Bobbi Swatek:
Linda, this is Bobbi Swatek with jcpenney. We actually have a store in Maryland that we installed with the 2'x2's, and it's gotten great reviews, and we are getting ready to open a store here in Dallas with the 2'x2' LED fixtures also.

Jeff McCullough:
This is Jeff McCullough. So are you putting them in relatively high ceilings?

Bobbi Swatek:
Twelve-foot ceilings.

Jeff McCullough:
OK, that's not too bad. I've been in some of your older stores where we're talking, what, 16, 18, 20 feet, and I—to Francis' point, that would have been a challenge, I think, technically for LEDs to do that.

Bobbi Swatek:
Right. A lot of our older stores do have higher ceilings, but our new store prototypes are lower ceilings.

Jeff McCullough:
OK.

Male:
The Dallas store, Bobbi, is that—you say the new prototype? Is that the one that's the single-story building and that jcpenney did as part of a DOE Commercial Building Partnership?

Bobbi Swatek:
Yes.

Male:
Thank you.

Linda Sandahl:
OK, anyone else tried them out? I think that would be something we'd probably want to share once we can get some results and maybe some photos of things to share.

Pat Lydon:
Yeah, that'd be great. Another thing that occurs to me, this is Pat Lydon again, in terms of what DOE could do to help encourage adoption, etc. Anything you can do to influence our regional organizations to offer incentives, not just on the luminaires or the products themselves, but even on programs like encouraging use of lighting designers and actual measurement for compliances, IES specs, and that kind of thing. Trying to encourage that because that doesn't happen all the time in all projects, especially in our environment.

Linda Sandahl:
OK, good suggestion. Any final suggestion? We're just about out of time. I would like to remind everyone there is an LED parking lot lighting team call at 3:00 p.m. Eastern, so that's coming up today. And if you aren't part of that team and want to participate in that, just respond to Terry and let her know, and she'll get you the information on it. But with that, I'd like to thank Jeff for the presentation, and thanks to all of you for being on the call, and we will be scheduling upcoming Project Team calls for this. We open this one up to everybody on the alliances that might want to listen in because we've made significant progress and had a spec completed.

But if you are real interested in this and want to be kept up to speed more frequently, also send Terry or I—my email's on the presentation here—let me know, and we can add you to that list. So with that, thanks everybody, and have a good rest of your day.

[End of Audio]