Below is the text version of the webinar titled "National Fuel Cell Technology Evaluation Center (NFCTEC)," originally presented on March 11, 2014. In addition to this text version of the audio, you can access the presentation slides.

Alli Aman:
I'm going to go through a few housekeeping items before I turn it over to today's speakers. Today's webinar is being recorded. So a recording along with slides will be posted to our website in about 10 days. I will send out an email once those have posted to our website, so you will know when they are up and ready for your viewing. Also everyone is on mute. So we encourage you to submit questions throughout the webinar via the question function box. So definitely if you have questions as today's speakers are speaking, definitely I encourage you to submit those during the webinar. We will get to those during the Q&A portion of the webinar at the end of the presentation. Also, I encourage you to check back to our website for future webinars. This is a monthly series, so we have different topics every month and I encourage you to check back to our website often to see what is going to be presented the following month. Also I encourage you to sign up for our monthly newsletter that we do send out; that will keep you up to date on all webinars as well as other things going on in the Fuel Cell Technologies Office. So I encourage you to sign up for that on our website if you haven't done so already. And now I'm going to turn it over to Jim Alkire who will be introducing each of today's speakers. Jim is a project manager for the technical validation and market transformation sub-programs in the Fuel Cell Technologies office. Jim?

Jim Alkire:
Alright, thanks Alli. And thanks again, everybody, for joining us for today's webinar. We have two excellent speakers today from the National Renewable Energy Laboratory. Both Jennifer Kurtz and Sam Sprik work, among other things, in the National Fuel Cell Technology Evaluation Center. So I'll give a quick bio and then we'll get started. Jennifer Kurtz manages the hydrogen analysis group at the National Renewable Energy Laboratory, which includes hydrogen and fuel cell activities in technology validation, safety codes and standards, market transformation, and analysis modeling. Prior to joining NREL in 2007, Jennifer worked for six years at UTC Power, primarily in fuel cell system design and components. So Jen will be speaking first and then Sam Sprik is a senior engineer at NREL and analyzes real world data from hydrogen infrastructure and fuel cell operation. He currently leads the hydrogen infrastructure task at NREL, which includes data from DOE technology validation projects. He develops software code in MATLAB to automate the analysis and has been working on hydrogen projects since 2005. Prior to that, he modeled and analyzed hybrid electric vehicles. So Jen, go ahead.

[Next slide]

Jennifer Kurtz: 
Thanks Alli and Jim. Sam and I are going to give you an overview of the National Fuel Cell Technology Evaluation Center, NFCTEC. We'll talk about a couple of results from the different applications that we look at and how those benefit the fuel cell community. And then we want to wrap up with a discussion on a fuel cell cost price aggregation projects that we're starting. And there will be a request that we've got in terms of getting people to volunteer and supply data for that project.

[Next slide]

So with that, I'll just start giving you the overview of NFCTEC to start.

[Next slide]

In September of 2013 there was a press release from DOE—the link is included on this slide—the "Energy Department Launches National Fuel Cell Technology Evaluation Center to Advance Fuel Cell Technologies". If you are familiar with the work that we had been doing prior to September 2013, that work was actually done with the Hydrogen Secure Data Center. So we went through a rebranding of the Hydrogen Secure Data Center to the NFCTEC, and on the bottom of this slide, it identifies a national resource for specifically hydrogen fuel cell stakeholders supported through EERE's Fuel Cell Technologies Office. The icon on this slide identifies the different applications that we look at—and we'll go through those in a little bit of detail in future slides.

[Next slide]

Our objectives at NFCTEC start with the independent secure analysis of data. The majority of the data that we look at is from real world operations. Again, the majority is also from DOE funded or co-funded activities and range in a number of different applications. Some of the things that we look for specifically in this analysis work would be benchmarking, confirmation of component and system technical targets—so where are these systems performing against targets for out years, generally like a 2015 target or a 2020 target. We also look at the optimization, demonstration, and some of the more basic deployment-type factors that help with the education of how these systems are being used, where they're being used, and how they're performing. A big piece of making this all work is the collaboration with industry and other organizations that actually supply the data, are using the fuel cell systems, the hydrogen systems. So that collaboration is quite important to us actually achieving these objectives.

[Next slide]

At NREL or the National Renewable Energy Lab in Golden, Colorado, we just dedicated a new facility last year called the Energy Systems Integration Facility. NFCTEC is housed in this new location. In the upper left hand corner is a map that identifies the different labs, the office space, and the high performance computing center. There is a link, if you just search on NREL ESIF there is a lot of information available about the different labs, the functions of the labs, functions of the overall building, but what I want to point out is two circles that are relevant to NFCTEC in that upper left hand diagram. The circle that is in the right hand side is where the NFCTEC work area is located. An image of that work area is on the upper right hand corner. Our computer information or hardware is actually stored in the High Performance Computing Center, which is in the middle module of this building identified with the circle on the left. The—excuse me, the hardware is connected across these two modules of the ESIF through a dedicated fiber network. It is not connected to any NREL network or external network here and that's part of the security procedures that we'll go over briefly. The bottom image is just a snapshot of ESIF that also identifies where NFCTEC is, the high performance computing, and the laboratories in the back of the building.

[Next slide]

Our security procedures are in place and have been in place for a number of years now to appropriately handle commercially sensitive information or proprietary information from our data partners. The image, the Word document image, is the title page of the security procedures. These are reviewed on a regular basis, if not more frequently at least every two years. It's shared with all of our data partners and it addresses things like the physical room security, how we transfer data, what we do within the NFCTEC secure work area, and the responsibilities. We're also connected to NREL's house security; the room does have alarm systems and is only allowed access through badge access, and that is with eight people here at NREL and that's all. So how we handle security procedures for this data is quite important and integral to how we are able to achieve the independent analysis of the proprietary commercially sensitive data.

[Next slide]

In general, data is delivered to us on a regular basis—most often it's quarterly, every three months. That data is put together, it covers operation, maintenance, and safety. It comes from developers, end users, other suppliers—they put it together and send it off to us here at NFCTEC. We complete an internal analysis at least once a quarter and then every six months, generally, we go through a publication process. That publication process has two features to it. It has detailed data products, or DDPs. These DDPs are individual data results that go back only to that data provider. It identifies some of the benchmarking of their systems with the rest of the systems that we are looking at. It also has a lot of additional individual system analysis, for a vehicle for instance it would be every vehicle fuel economy, the range, durability of those vehicles. That only goes back to the individual data provider. They use that to help with the review of the composite data products, or CDPs. This is the data that is published, it's available on our website through presentations, publications, and reports. It's aggregated in such a way that commercially sensitive or proprietary information is not revealed. And as I mentioned, it's generally published every six months; we do have some projects that are a little bit more frequent or less frequent.

[Next slide]

In order to do this consistently across multiple data providers, we utilize data templates and different tools. A couple of the snapshots, the Excel snapshots on this slide, are examples of vehicle operation, maintenance, safety, and general parameters for vehicles. The data providers send all of this information so we can make sure that when we analyze a particular performance metric, we're doing it with consistent data. The image in the upper right hand corner is our fleet analysis tool kit or NRELFAT, and we use this for all the applications. It's all set up in MATLAB to automate the processing as much as possible so that we can quickly go through a large amount of data consistently and efficiently.

[Next slide]

We've been working on technology validation analyses for a number of years now. This is a timeline on the bottom of the slide. We started in advance of—or before 2005, started with buses and cars and infrastructure. The stars identify when we've done publications or reports, some version of reporting on what the status of the technology is. So you can see there's a pretty regular drum beat of getting the analyses out there and making sure that the public—whether it be fuel cell and hydrogen developers, governments, or other end users—can see how these technologies are proceeding. One note that I also want to make is that all of this information is available for legacy too. So you can go in and see how we started with some of these reports with just a few—a few cars on the road, just one or two hydrogen stations out there that we were taking a look at, and how it's progressed. We've analyzed now over two million hours of fuel cell operation and we have applications in prime power, backup power, laboratory stack data, forklifts, infrastructure, vehicles, and buses.

[Next slide]

This slide is just meant to give you an indication within one application—this is for material handling and the infrastructure that goes along with material handling—the type of CDPs that we have, the different categories that we're looking at. So I mentioned deployment, general operation of fuel cells and the infrastructure—how many kilograms have you dispensed? How many hours are on the fuel cell system? We also look at safety. What kind of safety reports we're getting, what those categories are. Maintenance and reliability are two big areas that have seen an increase as these systems become more standard products. Then we were looking very carefully at the reliability and maintenance of these systems. Durability is also an important one, and cost of ownership, especially for the early market product validation work that we're looking at.

[Next slide]

And with that, I want to hand it over to Sam to talk about some of the benefits with some specific examples of the application.

[Next slide]

Sam Sprik:
Thank you Jennifer, and Jim and Alli. This first slide is kind of an example of real world operation analysis that we do here. One of the examples is a trend over time, another is an identification of a gap that leads to R&D focus, and the third example is end user perspective of cost savings for hydrogen fuel cell technologies versus incumbent technologies. So along the left side, with the car icon there, we see durability of the fuel cell electric vehicle. And this is basically hours projected to 10 percent voltage degradation. So you see for generation one, we had a max projection to 1,807 hours, for generation two we see an increase to 2,521 hours, and then our next generation FCEV durability evaluation is starting now, so we'll see where that number comes out. The second column shows infrastructure reliability, and what we've found is that there's a lot of poor mean time between failures. Basically low, low calendar time between failures. One of the things we look at is what are the major components causing failures? We've found that compressors are breaking down often. And so that leads to our R&D focus here at the lab and other places to do reliability testing of compressors. The third column, cost of ownership of the forklifts—we worked through a value proposition, we compared it to the incumbent technologies, which are battery electric forklifts, and we determined that there's an annual cost savings of approximately 2,000 for the class one and two fuel cell forklifts.

[Next slide]

This is a summary of the learning demonstration and then the carry-on work after that. So we've already talked about the fuel cell durability and you can see those numbers again here. The max team projected hours is 1,807 hours for gen one vehicles, 2,521 hours for the gen two vehicles, and you can see that the target was 2,000 hours in 2009. So we met that target. But on average, the fuel cell durability projection was for the gen one vehicles 821 hours, and then for the gen two vehicles an increase to 1,062 hours. And then after the learning demonstration finished, we actually have projections out to 1,748 hours. Now aside from the voltage degradation, people can obviously drive these things past that voltage degradation, so we also track max hours of actual operation on a stack. And for gen one vehicles we actually had one that went 2,375 hours, and then for the gen two vehicles, 1,261 hours. But they were still in operation so they just hadn't accumulated the hours yet. And then after 2009, quarter four, we increased that to 1,582 hours. For driving range, we had a target of 250 miles, and based on the window sticker, fuel economy, and range, for gen one vehicles we had 103 to 190 miles, and then for gen two we had 196 to 254 miles. That's the actual capacity of the vehicles but in real driving that number came down, the actual driving between fuelings was averaging 56 miles for gen one and increased to 81 miles for gen two and then after that it went to 98 miles. So this is probably just showing that people are more confident being able to get to a station to fill their car. But their car is actually capable of a lot more.

For fuel economy, this is the 42 to 57 miles per kilogram on the gen one vehicles, 43 to 58 miles per kilogram on the gen two vehicles. And then the efficiencies of the fuel cell at quarter power—the target was 60% and you can see that we're just shy of that in the gen two vehicles. And then the fuel cell efficiency at full power, the target was 50% and the range shows that we met that, so 42% to 53% on the gen two vehicles. As far as the costs for hydrogen, there was a target of $3 per kilogram for the hydrogen and then we did a survey of the experts and found out that based on the early market stations, the demonstration stations, we could get down to $7.70 to $10.30 per kilogram for the on-site natural gas reformation, and $10 to $12.90 per kilogram for electrolysis on site. And for the average fueling rate for these different structures, we saw an average of 0.77 kilograms per minute with the target being 1 kilogram per minute. So there were fuelings above 1 kilogram per minute, but the average ended up at 0.77. And then after some of the stations shut down, we saw an average of those stations being 0.65 kilograms per minute. So—and outside of this project—DOE independent panels concluded at 500 replicate stations per year and for the distributed natural gas reformation, at 1,500 kilograms per day stations got down to $2.75 to $3.50 per kilogram, and then for the distributed electrolysis it went down to $4.90 to $5.70 per kilogram. And that was outside of our work here.

[Next slide]

This is an example of how infrastructure reliability analysis supports additional R&D projects. So we've found that a lot of the sites were showing a mean time between failure of 25 days or less but still consistently delivering 250 to 350 bar. So that means they're either fixing stuff quickly or they had redundant compressors on site to take care of any failures there. So we can see in the pie chart that compressors were 51% of the issues—of the 1,330 maintenance events—and then control electronics came in second, with dispensers and air systems next. So we've started projects on compressors and hoses, as examples for R&D, to test their reliability here at the lab.

[Next slide]

We talked about the comparison with incumbent technologies for forklifts and we said that the cost advantage is approximately $2,000 per year, so all of the costs get broken down into annualized costs, and you can see each of the categories there. There's the maintenance of the battery or the fuel cell, there's the lift truck maintenance, cost of infrastructure, and then basically if you look at the cost—per-lift cost of charge or fuel infrastructure, you can see a big difference there. So the green item—for battery lift, the infrastructure is $1,400 per lift per year and for the fuel cell forklift infrastructure is $3,700 per year per lift. So you can see that the infrastructure for hydrogen is a little bit more expensive than for battery but there's other advantages like less warehouse space, less maintenance on the fuel cell, and items such as that. There's also many other conclusions in the report that can be found by a simple search on our website.

[Next slide]

Now we also work on fuel cell bus evaluations here—an international effort, with many different stakeholders. On the right, top plot, you can see that the total hours for the Zero Emission Bay Area bus fleet, which has 12 buses, total hours on all those buses are almost to 80,000 hours. And the leader of those is at 16,216 hours, and the second one at 11,908 hours. On the bottom left you can see maintenance by category. So labor hours attributed to each of the components, so you've got on the top the green one is the hybrid system maintenance, you've got fuel system, Brusa charging, cooling and air intake, compressor, and then the actual fuel cell system is the blue one on the bottom. You can see by months, there's a certain number of labor hours associated with each component, and that's from the British Columbia Transit propulsion project, and there were 20 buses that were included with that. Also our partners for the bus program were the National Fuel Cell Bus Program, the Department of Energy, Department of Transportation, the Federal Transit Administration, the California Air Resources Board, other public transit agencies, and developers and integrators.

[Next slide]

Switching from the real world operation data out in the field, we also focused on lab data and we've collected data voluntarily from 15 U.S. and international fuel cell developers. You can see all the different applications that these are actually tested in the labs. You've got backup power, automotive, bus, forklift, and prime power. Along there you can see the DOE targets for hours, projected to 10% degradation again, and backup power has about a 10,000 target, automotive has 5,000 hour target, which is about 150,000 miles, and you can see buses about 25,000, forklifts at 20,000 and then prime power an interim target of 40,000 hours. And you can see where the lab data comes in, they're all a little lower than the targets on average but progressing well.

[Next slide]

For the material handling project, we have a lot of high use facilities with almost 2,000,000 operation hours, over 290,000 hydrogen fills, a lot of hydrogen dispensed, and some of the key metrics per fueling times and amounts, you know 0.6 kilograms per fill and about 2.3 minutes fill time, which is a huge increase over the battery forklifts when you have to charge them or swap out a battery. About 490 units in operation and about—maybe average about 4.4 hours between fills. So a little bit more than half a shift.

[Next slide]

Similarly we have different sites for backup power, and they're operating reliably in 23 states. And some of the reasons for unsuccessful starts—you see 99.7% successful starts, so there are a few that didn't start when they should have. One of the problems was the e-stop signal, one was no fuel, and there are other system failures also. So about 842 systems are in operation with about 4 to 6 kilowatts of power available at each site.

[Next slide]

We also do the infrastructure evaluation for vehicles. This shows we track basically where the stations are, where the future stations are coming in. Plotted down on the left, Los Angeles area, so California is strategically putting a lot of money towards infrastructure build out right now as far as spacing out the stations, locating them in certain cluster areas. So you can see the circles here just show a 6-mile radius and a 3-mile radius, so you can see how the stations are kind of closing in into a nice network. Top left shows the iPhone application where you can actually view where the stations are through the Alternative Fuels Data Center. You can Google that and find that. And then on the top right, it's just a trend analysis of what day of the week do people fill. So you can see, the y-axis on the left side is dispensed hydrogen percent of total, so you can see that the least amount is dispensed on Sundays and the most amount is dispensed on Wednesdays. And then the white lines are individual stations, you can see the averages for those stations on those days. We try to look at a lot of trends for these stations as they're coming up into the more commercial market. There's a lot of other CDPs with fueling rates and maintenance and capacity utilization, all of that you can find on our website.

[Next slide]

All of our data supports the DOE updates, records, and status. Examples of that are an Early Market Fuel Cells for Material Handling fact sheet, Fuel Cell Bus Targets record, and a Fuel Cell Backup Power Deployments record. These can all be found on the DOE site.

[Next slide]

And from our Annual Merit Review, we've gotten peer review feedback, so you can read through these while I leave it up, but if you focus in on the bottom two, I'll read them. "This is a great way to put all the data together and get information back to the industry and potential customers." and "This project is essential to benchmarking the progress of fuel cell systems over time and across industries." So with that, I'll turn it back over to Jennifer.

[Next slide]

Jennifer Kurtz:
Thanks, Sam. So Sam went through just a few of the results and at the very end of the slide deck, we'll have the website again. And I would just note that all of the results, even the legacy results, are available on our website, they can be broken down by different categories, reports, different times. So I encourage you to check out the website and get more information there on these different applications. With that, I want to go into—take a little bit of a jump into one new project that we're working on right now. This is the fuel cell cost and price aggregation project.

[Next slide]

I'll start with the DOE record. This is looking at fuel cell system cost and it's important to note that this is a model projection for high volume, so 500,000 units per year. The source for this record is actually on the bottom of the slide. And it shows where the status is at high volume for cost, looking at $55 per kilowatt and the ultimate target being $30 per kilowatt. I'll note that it says updated analysis and previous analysis. The updated analysis included adjustment for platinum, higher platinum cost. So this is so far what DOE is using for cost reporting. What we're looking to do is make it more relevant to what the industry and the market is right now for fuel cell costs.

[Next slide]

So we are looking for fuel cell developers to participate in this project by supplying data to us for cost and price of fuel cell products. The one important note is that we would treat this information like we do all of the other NFCTEC data, that there is a review process with all the project partners. I forgot to mention that when we publish CDPs we go through at least a two-stage review of the report or the publication in advance of releasing it. The people who provide data have the opportunity to provide feedback that ranges from everything looks good, go ahead and publish, to you know I'm really confused about this particular result, or this format for the results, or I'd like more information. It ranges a whole window of potential feedback. So we take that and go through two iterations before we publish any of the CDPs. This particular project would follow that same process.

So we want to support DOE with identifying what current costs are for fuel cell projects. In the lower left hand corner is an example result that would look at the dollars per kilowatt for different applications against the number of units. And note that we're looking at one to a thousand, maybe ten thousand units produced in a year. So something much lower than the 500,000 unit production used for the model base, high volume cost numbers. On the lower right hand corner would be some of the cost numbers that we've collected so far. Backup power, forklift, and prime power and this is dollars, thousands of dollars per kilowatt in the range. So this is another way we can publish information without revealing the commercially sensitive individual information.

[Next slide]

This is an example, or this is the cost/price data template that we're looking at. Looking at the current price, we understand that there is a lot of variability that could be included in price information. So we request some more feedback. Is it based on a certain number of units produced or a certain number of units in that sale, what type of application are you looking at? These pieces of information help us correctly bin the data so that we're comparing apples and apples, if you will, when we're looking at price information. There's a few other pieces of information. If there's a—to show perhaps a decreasing cost trend, we've requested some of the legacy cost information here, or price information. And I am using cost and price in this discussion because we understand that it may be a little bit easier to get price information to start, so we are open to both cost and price and we would correctly identify whether we're looking at cost or price when we publish these results.

[Next slide]

This slide talks about the different benefits of participation. And we normally look at benefits—from what Sam went through, some of the technical benefits, the benchmarking capabilities when you get your individual CDPs, for instance. And also the records that are being published through DOE. So specific to this particular project would be a breakdown of external—so this would be—external benefits would be to the larger community, for instance, going through with the realistic cost of fuel cell, or price of fuel cell technologies right now and that would get wrapped up into DOE status and record keeping. We also think that reporting on cost and price information right now would help identify the trend that we're seeing, or that we expect, in some of the decrease of cost as some of the sales of fuel cell products increase. So going through this project would help us track that a little bit more completely and in an independent and objective manner. We also want to acknowledge that the cost of high volume projection is not what customers see right now, it's not what the systems are being sold for right now. So we want to fill in that gap between what's going on right now versus the high volume cost. Internally, one of the pieces is the benchmarking, looking at where your system cost and price are with relation to the rest of the people who are supplying the data. And we have had a lot of collaboration with industry where we find, we start looking at something and maybe there's a new direction that we would go with a type of analysis. I would say that this applies a little bit more with some of our more in-depth analyses like the durability work, but we do very much enjoy, and think it's critical to the work that we do, collaborating with industry to understand the status of technology, both performance-wise and cost right now.

[Next slide]

And with that, I think we're going to open it up to questions. This is how—and Alli or Jim, please correct me if I'm wrong here, but you submit your questions through the chat feature of the webinar.

[Next slide]

And then I want to just leave this slide up for a little bit. Again, this is our website, all of the applications that we talked about today, including the lab durability status project, all of the published reports, are available on our website. The snapshot is on the left hand side there. And then I want to take a quick moment to thank our team. The image on the right, you've got Sam and Mike Peters, Matt Post, Keith Wipke, Chris Ainscough, Genevieve Saur, Leslie Eudy, and myself. So email is at the bottom and we enjoy this work, we enjoy working with industry, reporting on this status and just a fun set of projects to work on—and also just make a little plug for ESIF that it's a fun building to work in and be a part of the next wave of efficient energy research.

Jim Alkire:
Okay, thank you Jen. We will open the floor to questions. We already have a few coming in over the GoTo webinar. We'll get on those in just a second. I invite everyone to continue to submit questions, we'll try to answer them as best we can throughout the rest of this webinar and if there are any questions that we either don't have time to get to or we would prefer to take back and think about a little bit before coming up with an answer, I think we're also going to log all of the questions and answers and also post those online with the webinar. So please go ahead and shoot us some questions and in the meantime, we'll get started on answering a few of the ones we already have. First question, can you please reiterate the definition of prime power?

Jennifer Kurtz:
Certainly. We consider prime power to be stationary power. The simple definition is that it's not backup power. So it operates more than in an intermittent grid-down situation. But it's not necessarily a system that is operating 24/7. So the stationary fuel cell power ranges could be the small residential sizes up to the larger commercial-size stationary fuel cell systems.

Jim Alkire:
Okay, thanks. The next question here, how do you explain the huge cost difference between automotive and other applications?

Jennifer Kurtz:
Sorry, Jim—with the cost information we don't have the vehicle cost data so do you think the question is related to something, another report? Or is it more the durability comparison?

Jim Alkire:
Yeah, it's tough to tell. It could just be outside knowledge here.

Jennifer Kurtz:
Okay. With the cost information we haven't reported on automotive costs. There are certainly press releases out there in terms of cost to lease fuel cell cars coming up, but we don't have that particular information that we've published right now.

Jim Alkire:
Another question here might help out. The question is, the fuel cell cost in one slide was said to be $55 per kilowatt and then another slide was some $1,000 per kilowatt. What's the difference?

Jennifer Kurtz:
The difference with the $55 per kilowatt is that is—that's at 500,000 units a year, so high volume. The cost numbers that we've started collecting, or the price numbers, are representative of what is available right now. And so at much lower volumes costs and prices are higher, and that's part of why we want to fill in the gap because there is, for instance, customers who are looking at the record expect a much lower dollars per kilowatt but may not understand or get the difference that that's at a high volume units per year as opposed to the hundreds or thousands of units per year.

Jim Alkire:
Okay, thanks. I think I can field the next one which is just administrative. It says will we be able to download the slide deck? The answer to that is yes, we will have this webinar and I believe a recording of the webinar posted online as well as some questions and answers. I'm just scrolling through here to get another question. So there seems to be a lot of focus on vehicles and correspondingly PEM fuel cells. Do you analyze other fuel cell types as well?

Jennifer Kurtz:
We do. We include—for instance, with the lab durability we have some solid oxide fuel cells. Some of the other stationary prime systems are with other fuel cell technologies. But most, the majority of the fuel cell systems that we've analyzed to date have been PEM fuel cells systems in the motive categories and the lower—the backup power, lower-durability requirements and low-cost-requirement systems.

Jim Alkire:
Okay, thanks. The next question is, in one of the early slides it says that the vehicle project was renewing. Can you explain that a little more about what it means, the project renewing in one of the early slides.

Jennifer Kurtz:
Certainly. The project renewing is from our perspective. What was called the Learning Demonstration, the DOE Learning Demonstration—which included the 183 vehicles and 25 hydrogen stations—wrapped up. We completed the final report on the performance, the status, and additional needs in 2012. We are getting started with our next steps, or next generation of the evaluation project, and we're getting started on that right now. So the project renewing is from our perspective. We're going to get new vehicles that we're going to start taking a look at. We're going to be able to compare the cars that are operating on the road right now with the performance status reporting on some of the 2015, 2020 targets, and we also will be able to show some of what we expect, but we'll go through to verify some of the gains that are continued to be seen in performance. Range, fuel economy, efficiency, durability, things like that. So the project renewing is from NREL's perspective that we're going to start another round of fuel cell vehicle analysis and technology validation.

Jim Alkire:
Okay, thanks. A related question was about the cars. How do you define gen one, gen two, et cetera?

Jennifer Kurtz:
It's primarily based on the fuel cell system power plant. Generation one in the Learning Demonstration would've been fuel cell technology from 2003 time frame that started getting rolled out onto the road in the 2005 to 2008 time frame. Generation two would be that next set of improvements to the fuel cell power plant system that we took a look at. So generation one and generation two are relative to our validation work. It's not necessarily the first generation of an individual fuel cell vehicle OEM. They may have gone through a number of generations or technology advancements up until that point. So generation one and two is very specific to the Learning Demonstration project evaluation and it is primarily a time window split around 2008 when we started looking at the generation two vehicles. Generation two vehicles we saw freeze capable vehicles, we saw an increase in efficiency, we saw an increase in durability, we saw an increase in range. So those were all features that were implemented by the developers and OEMs to improve the performance of those cars.

Jim Alkire:
Okay, thank you. We have a question about the cost analysis. What is the best way to interface with you to provide data on costs for items such as bipolar plates?

Jennifer Kurtz:
I suggest contacting one of the emails, techval@nrel.gov or jennifer.kurtz@nrel.gov, and starting the conversation that way.

Jim Alkire:
Okay thanks. Has NFCTEC collaborated with or will consider collaborating with academic entities?

Jennifer Kurtz:
Yes. We do now and we are—one of the features in, for instance, a press release with NFCTEC was actually identifying the different stakeholders that we are interested in collaborating with, and academics were one of those groups. So universities, government, and industry were the three primary categories that we're looking at. And again, the emails are probably the most efficient or effective way to start that communication. On our website is the—if you're kind of looking at our website, looking at the different Web pages under Technology Validation, this techval@nrel.gov is the primary email address and certainly my email address is just as effective as well.

Jim Alkire:
Okay, we've got another question about different methods of generating hydrogen. It says hydrogen has generally been produced by using natural gas, does your cost analysis include other means of producing hydrogen, either on-site or centrally?

Jennifer Kurtz:
Okay. For the cost of ownership analysis—as an example, for material handling, we had most of the sites were delivered hydrogen and that's what we used for the cost. We also had on-site generation from natural gas that was considered for cost. So I would say the short answer is yes, but I always like to give a long answer. It depends on who we have data from in terms of what systems we're including in those costs. So natural gas, hydrogen—or natural gas reforming to hydrogen is certainly a part of that. With the cost table that Sam was talking about we looked electrolysis, we looked at natural gas on-site reforming; a lot of the current infrastructure stations are delivered. So I would say we've got the spectrum of hydrogen production options that we're analyzing.

Jim Alkire:
Okay. Would you be able to calculate the impact of bipolar plate costs to the overall cost of the PEM system?

Jennifer Kurtz:
I think that would be a good question to take offline and perhaps discuss in more detail with the person who's asking that question. The short answer is that would be part of the consideration—for instance, with the material handling cost of ownership we look at—we broke that cost of ownership into different segments of the ownership cost, right? So the infrastructure costs, both upfront and operating, the fill time labor costs, the capital equipment costs, maintenance costs. So I would anticipate that if data is available, we would do similar breakdowns, but the big caveat there is if we have enough data available to do that.

Jim Alkire:
Okay, thanks. Just give me a second here. Okay. Which fuel cell component has the highest contribution to the dollars per kilowatt?

Jennifer Kurtz:
I would encourage looking at the record with the different breakdowns of the fuel cell components that are based on the modeling results right now to answer that question. We don't have—we haven't published any of that information. We're just trying to get started on the price data right now.

Jim Alkire:
Okay. Just one second here. Alright. For prime power, are there calculations to determine the cost analysis that include the total co-generation ability for using waste heat from PEM or solid oxide to operate a rankine cycle generator?

Jennifer Kurtz:
We don't have that breakdown right now. We didn't talk too much about prime power applications, that is, we do have all of our results on the website right now. We are looking at breakdowns of installation costs of the different systems, and I would say in the prime power, that is an area in general that we are actively pursuing getting more data so that we can break down some of those numbers a little bit better. What has the combined heat and power potential, what's the performance of that, and how is that impacting the life cycle cost, the operation cost of the system. So we don't have that—I will just throw out that it would be good if there are some stationary developers on this webinar or end users that would be interested in working with us to better identify what those cost breakdowns are, please contact us.

Jim Alkire:
Okay. Do the reports break down cost of the components?

Jennifer Kurtz:
With the cost of the components—the material handling, we broke it down more by operation categories. What do you need to do to operate it as opposed to components. With respect to a fuel cell stack or system, we don't have that information right now. There is modeling available that would help us break that information down, but as of right now we don't have that data published under NFCTEC.

Jim Alkire:
Okay, thanks. Will you be releasing any information related to replacing or reducing platinum use in fuel cells or other advancements?

Jennifer Kurtz:
That could certainly be a part of the breakdown when we're looking at trends over time, but it is probably better suited within the fuel cell program under Dmitrios, for instance, to look at the breakdown of the platinum loading, for instance, on fuel cells. But for instance, when we were looking at durability from lab projections, there's a trade-off, right? You can load it up with platinum and then it's very high cost but it has good durability. Decrease catalyst loading, it has lower cost but it may not have as good of durability. So they are all connected and I think there's ability for us to do some of that reporting and I also think that within the fuel cell program, the research going on, that's also another very good resource to identify where the research is in terms of platinum reduction and the impact on some of the performance metrics.

Jim Alkire:
Okay, thanks. We have a question here about station reliability. It seems the failure was defined on a component basis. What was the motivation for that choice? And was similar analysis done for system-wide failure?

Sam Sprik:
I guess the system-wide failures would include stuff like power outages and electrical issues and sensor issues shutting down stations and stuff like that. So we do have that break out, they don't happen as often. It's easy to identify when a component breaks down or its particular failure mode. So we do track all of it and we just basically concentrate on the major issues that come up.

Jennifer Kurtz:
And with the compressor example, that was one example of a leading cause for infrastructure maintenance downtime and led to some additional focused research and testing—for instance, the accelerated life testing of compressors to identify failure modes, which will have an impact on improving overall infrastructure reliability. And as Sam mentioned, there are a number of reliability maintenance-focused results specific to infrastructure that are available on our website to get into more detail of the different categories, the different failure modes.

Jim Alkire:
Okay. I think that's a good segue into the next question about compressor technology. Can you elaborate on the needs for improvements in air and hydrogen compressors?

Jennifer Kurtz:
For improvements, the biggest thing that we are focusing in on is the failure modes and working on those failure modes to improve reliability. So that's where our focus is. One particular example is collaborations where we're doing accelerated life testing on the compressor. So running at higher loads or higher operation time than what's being seen in the real world right now and then doing deep dives on materials, mechanical failures, if there are control strategies that can be changed in terms of how you operate inputs and output pressures—things like that that we're focusing in on. So I would say that the biggest push right now for compressors is that reliability aspect and understanding why we see so many unscheduled maintenance events for compressors—compressing hydrogen, I should say. It is hydrogen compression specific.

Jim Alkire:
Okay, thanks. We have a few more minutes here. Where can I look to find out what type of fuel cell systems automobile OEMs are using today?

Jennifer Kurtz:
I'm sorry, Jim. What type of fuel cell systems?

Jim Alkire:
Yes.

Jennifer Kurtz:
Okay. So, the PEM fuel cell system is used in automotive applications right now.

Jim Alkire:
Great, thanks. Can we have access to your technical and cost analysis models?

Jennifer Kurtz:
If the question is referring to the modeling work for the high volume, I would suggest going to the record and identifying the best contact person for that. Most of our work that we are intending to do is more aggregation-based as opposed to modeling work. If the question is focused in on the cost of ownership analysis that we've done, for example, on the material handling, I would suggest perhaps a conversation offline and looking at the material handling cost of ownership report that identifies the assumptions and how we looked at the cost of ownership.

Jim Alkire:
Okay. Do you have any idea of the cost breakdown between the balance of plant and the stack for prime power and backup?

Jennifer Kurtz:
We don't have that information right now for current systems. I think that our template is not aimed at doing the breakdown of BOP versus stack versus system right now. The first wave that we want to go through is actually just getting some of the system costs. And what we've found when we get started on some of these new projects is we take the first step and then we expand on how we can look at reporting on those numbers a different way. For instance, with the fuel cell vehicles when we started in 2005, we had intended on releasing I think it was 15 CDPs or published results. We ended up with 99 and that was just going through, identifying different ways to look at it, different things that were important to the audiences that we were targeting. So—and then with the modeling, the high volume modeling, there is some information on the breakdowns of the BOP versus the stack, and repeat versus non-repeat parts of the stack.

Jim Alkire:
Okay, thanks. We're going to take just one more question if you don't mind. I do appreciate all the questions, we are running out of time. But Jen, one last one. What is the difference between durability and reliability on your CDPs?

Jennifer Kurtz:
Durability is looking at the fuel cell stack performance. So it's not necessarily an end of life, it doesn't identify when you would need to do maintenance or need to do unscheduled maintenance. So the durability is looking at specifically fuel cell stack voltage at specific current levels over time and how long it takes, excuse me, to either get to a 10 percent voltage drop or project out to a 10 percent voltage drop. The reliability analysis is focused in on maintenance events, unscheduled maintenance events. The frequency of them—on the bus side, the mean time between roadcalls, for instance, or the mean time between failures calendar time. Also the categorizations of unscheduled and scheduled maintenance. So from our perspective, durability is really focused in on the fuel cell stack, and the reliability would be more system maintenance, frequencies trending over time for unscheduled maintenance.

Jim Alkire:
Okay. Thank you very much. With that I'll turn it back over to Alli for the wrap up.

Alli Aman:
First of all, thank you so much Jen and Sam for taking time to present in our webinar series. Just as a reminder, we did record today's webinar so a recording along with slides will be posted to our website in roughly about 10 days. I will send out an email as soon as those post. Thanks again and we hope you enjoyed our webinar today and we hope you plan to join a future one coming up. So thanks so much everybody.

Jennifer Kurtz:
Thank you.