Cassandra Osvatics, Hydrogen and Fuel Cell Technologies Office: Hello. Welcome to this month’s H2IQ Hour for an overview of record setting high throughput hydrogen fueling at NREL. My name is Cassie Osvatics with the Department of Energy’s Hydrogen and Fuel Cell Technologies Office or HFTO, supporting stakeholder engagement and other outreach activities. Please be aware that this WebEx webinar is being recorded and will be published online on our H2IQ Hour webinar archives. If you experience technical issues today please check your audio settings under the audio tab. If you continue experiencing issues, please send me a direct message. There will be a Q&A session at the end of the presentations and attendees have the opportunity to submit questions in the question box which is located near the chat function. I’d now like to turn it over to Mark Richards. Mark is our technology manager at HFTO and he will be monitoring today’s webinar.
Mark Richards, Hydrogen and Fuel Cell Technologies Office: Thanks Cassie and thanks everyone for joining. I’d like to introduce today’s two presenters from NREL. First is Shaun Onorato, hydrogen systems engineer and lead researcher. Shaun manages the heavy duty hydrogen research and development efforts at NREL’s energy systems integration facility which includes the new heavy duty hydrogen fast flow research station constructed by the NREL team. Second, is doctor Taichi Kuroki, a PhD in chemical engineering and lead researcher. Taichi conducts analysis and modeling work on hydrogen infrastructure and his focus is the development of numeric models, tools and computational fluid dynamics for hydrogen fueling stations and hydrogen powered vehicles. Both have been at NREL for about five years. Welcome to you both and I’d like to hand it over to Shaun.
Shaun Onorato, National Renewable Energy Laboratory: All right. Thank you Mark. So I’ll be starting the presentation and then Taichi will be jumping in to provide some updates on our modeling work. All right. All right. Again thank you for joining us. So this presentation is on high throughput hydrogen fueling and R&D at NREL. So this project was funded under an industry led cooperative research and development agreement which was a 50/50 split of DOE and industry funds. It was under the innovating hydrogen stations project and the primary project objectives were to create a heavy duty research capability, develop comprehensive high flow rate fueling models and generate publicly available tools and data to provide to hydrogen stakeholders. So you can see the station or our research facility, research station here in the top image as it was in 2020 and then the improvements that we’ve made and the new research capabilities as of now 2022 and that’s showing the expansion shown on the right side of the image.
So I’d like to briefly mention some information about NREL’s hydrogen infrastructure research. So we have sort of our larger at scale activities surrounding grid and renewables coupling. So this involves electrolyzers as dispatchable loads and power systems, dynamic operations and integrated – we integrate these activities with [Break in Audio]. We have shown on a smaller scale activities that are hydrogen production so that’s fuel stack scale electrolyzer and EOP performance, system optimization and then coupling that with grid and renewables for end uses, tying that back to our larger scale research, distribution and storage. So system scale distribution while we’re addressing things like storage challenges and vehicle and ground storage performance as well as some activities, a large amount of activities on modeling. We’re skipping over to the right side of the image CPN sensors so development and evaluation of safety sensors and systems. We do a lot of work in component failure in this area and as well as some characterization. And then the area that this work falls under is end use applications so that is transportation applications, industrial. We do do some activities around natural gas blending and renewable synthetic molecules.
So this is DOE’s H2@SCALE vision. So H2@SCALE is a Department of Energy initiative to advance affordable hydrogen production, transport storage and utilization to enable decarbonization and revenue opportunities across multiple sectors. So this graphic can be found on DOE EERE’s website and it shows the various pathways for hydrogen production and use. So on the left side of the graphic here we’re generating electricity from a variety of sources so renewables, nuclear and fossil energy, moving energy, electrical energy into the grid. Then utilizing the various hydrogen production methods, one being electrolysis to generate hydrogen and then using hydrogen in these variety of ways here so transportation, synthetic fuels, upgrading oil and biomass and then some industrial applications shows here in blue. We can also use hydrogen blended into the gas, natural gas infrastructure or heat or put that back into this hydrogen pathway and use it for power generation. So for this particular work that we’ll be discussing today we are very focused on this transportation subcategory here at the top.
So here at NREL this is NREL’s strategy for decarbonizing the transportation sector that mostly follows DOE’s H2@SCALE vision. So we are following the framework here. So with renewable at scale renewable production of electrons, using those electrons as I mentioned in electrolysis, producing hydrogen which we can use in a variety of ways within the transportation sector. So we are very – we had been previously focused on light duty vehicles and continue that research but we are now moving into the medium and heavy duty space and then looking forward into the future, looking at aviation, marine and rail. We can also use the electrons generated from the renewable sources to power buildings, EV charging which then translates back into transportation. We can also use hydrogen for fuels and chemicals production so liquid fuels, CO2 capture and other chemicals. So we are actively building out this capability at the laboratory and closely following DOE’s HD at Scale vision.
So DOE and the industry have identified a strong business case for medium duty and heavy duty vehicles in the fuel cell electric vehicle space. In comparison to light duty vehicles, heavy duty vehicles require extreme duty cycles, more range, significantly more on board storage. So using diesel as a metric industry and DOE have derived baseline targets for hydrogen fueling. So current aspects of light duty fueling infrastructure will need more innovations to meet the demands of heavy duty fueling. So typical fueling rate for a light duty fuel cell electric vehicle is around 3.6 kilograms per minute or 60 grams per second. So there are advancements that need to be made in terms of station design, fueling protocols and hardware that will be able to enable fueling these larger vehicles for the reasons that I mentioned. Hardware being things like nozzles, hoses, breakaways, receptacles and dispensers currently do not exist for these applications in a commercial form so they will need to be developed. So advancements made in this focus area for medium and heavy duty trucking will translate and enable other emerging applications such as marine import, mining and industrial, rail and aviation.
So this is NREL’s process for developing fast flow capabilities and generating our fast flow facility capabilities. So we are generating or gathering goals and requirements from industry, the federal government and from existing codes and standards. So one of the industry and federal metrics that we are using for the development of our fast flow capabilities is the DOE class long haul tractor trailer truck targets where we are targeting a hydrogen flow rate of eight kilograms per minute interim and ten kilograms per minute ultimate. And so these goals and requirements feed into our modeling and physical capabilities. So some of our modeling capabilities that Dr. Taichi Kuroki will touch on later in the presentation are computational fluid dynamics as well as our simplified fueling model H2 Fills. The modeling capabilities allow us to build out our fast flow facility, generate experimental data, leverage existing legacy data that NREL has been collecting over nearly a decade of research and development, feeding experimental data back to the modeling for validation activities.
These activities combined help us develop new standards for fueling protocols as well as updating existing codes and standards and evaluate some of the new hardware aspects that I mentioned on a previous slide which involves refueling hardware, nozzles, hoses, breakaways, dispensers and new fueling protocols. So it’s important in terms of the modeling aspects to understand from a physical aspects of fast flow and how these aspects influence station and vehicle systems. We also want to establish robust fueling models for validating protocol concepts and generating fueling tables for moving forward with those fueling protocol concepts.
So in order to understand some of the fast flow data that we’ll present later in this slide deck here, we’d like to go over the NREL heavy duty research capability design. So very excited to present to you a first of its kind experimental research capability for medium and heavy duty fueling. So this is the NREL fast flow facility. It’s located at the energy system’s integration facility in Golden, Colorado at NREL’s primary campus. We leverage the existing legacy light duty station capabilities and expanded on them. So we have a fueling capability which is gaseous up to 70 MPA, precooling capability down to negative 40 C, 10 kilogram per minute average fast flow rate with a target of 20 kilograms per minute peak and 80 plus kilograms of flow mass into a heavy duty vehicle simulator which I will dive into the details here shortly. The station has been or our research capability has been operational since July 2022. We have about 650 kilograms of bulk gas storage which I will also elaborate on here shortly and we have limited back to back fueling capability. So this fast flow facility enables evaluation of fueling protocols, components and hardware so things that I’ve identified as lacking in this area that need to be developed commercial scale and then developing publicly available modeling tools and data for these two UL hydrogen infrastructure stakeholders.
So this is a quick walk around the station. So we have a part of the station over here with our compression system that can be fed by our hydrogen that’s generated from renewable sources. Walking around the station here we have our low pressure storage. Hidden from view is our medium pressure storage and then high pressure storage in the back of the station. We have our heavy duty gas management panel that was built out for the addition of our high pressure storage tanks. We have a heavy duty dispenser which does not look like a dispenser in a traditional sense that you might see at a hydrogen fueling station. This is an open framework which allows us to mix and match components and test different hardware as we move into the research space. The dispenser then goes through a micro channel heat exchanger which is hidden from view in this image and we fuel into a heavy duty vehicle simulator which are the black tanks shown here. The system here is precooled or the hydrogen is precooled with a brine storage tank that is chilled with an R4 chiller system located across the street. The blue box in the center of the image is the device in the test area where we will be evaluating the hardware that I mentioned so nozzles, hoses, breakaways and dispensers with newly developed fueling protocols in this space.
So I’d like to walk through kind of each individual subsystem so you can understand more about the capability which will make the data make a little more sense here in the next few slides. So the system is – the pressurized ground storage for heavy duty fills is composed of low pressure, medium pressure and high pressure storage so you can see the associated pressures on how we fracture out this kind of cascade approach. We do not use low pressure storage for heavy duty fills although this system is integral to topping off the medium and high pressure storage system with a compressor. So we have ten, 18 kilograms per tank type ones for the low pressure storage. We have six type one tanks. So sorry. Low pressure is about 180 kilograms total. For medium pressure it is 6, 13.5 kilogram tanks. Those are type one tanks for a total of 81 kilograms. The high pressure system is fractured into type two and type one tanks so that is part of our kind of legacy system for the type twos and then moving to type ones is what we’ve added for our new heavy duty capabilities. So 8, 16 kilogram tanks, those are again type twos and a total of 128 kilograms. And then 8, 32 kilogram per tank type ones and that is a total of 256 for a grand total of 650 kilograms of storage. So we have demonstrated 80 kilograms of mass transfer using the high pressure storage into all of our heavy duty simulator tanks. I’ll elaborate more on what that means here shortly. And then 60 kilograms of transfer using only high pressure into seven of our type four tanks of our heavy duty vehicle simulator system.
So this is our brief walkthrough of what our heavy duty dispensary is comprised of. So it is mostly one inch and three quarter inch OD 20,000 psi rated stainless tubing with associated air operated valves so these are very large valves, very large lines. You can kind of see them in the picture here to the right. These are one of the largest commercially available hydrogen components when built. I’d like to point out this is an important aspect of when we walk through the data later. We utilize a parallel flow path for our control strategy. So that is comprised of a 9/16th mechanical flow control valve which is a regulation type as well as a 9/16th open close air operated valve. Because some of the refueling components that I mentioned earlier do not exist the current configuration for the station is a direct hard pipe connection to our vehicle system. So that is – we use a two and three quarter inch 20,000 psi tubing connection. This also acts as an isolation from that system which reduces our code setback distances from fire lanes. So the connection, there is also a connection that exists for mounting a breakaway hose nozzle assembly when those components are ready for testing. And that is shown in the lower right hand image here. That’s kind of up on the upper edge of the vehicle system. We have full – so its important to note that we have full hydrogen recirculation capabilities here at NREL. This helps us reduce cost and downtime and storage size of the station, the research station so the heavy duty vehicle simulator can also act as a tube tailored delivery back to our bulk hydrogen storage NFPA2 code.
Here's a quick very simple five dispenser graphic showing that that parallel flow path so with the mechanical flow controlled valve and then the air, the air operated valve through the heat exchanger and then into the vehicle system.
All right. So this is a brief overview of our heavy duty precooling system. So this system allows us for instantaneous precooling of hydrogen gas from ambient to negative 40 C and that is per SEAJ 2601 standards. So this is configurable to a variety of temperatures down to negative 40 C. This will be important for evaluating new protocol concepts in the future. So the system is chilled by twin 30 horsepower chilling circuits that chill our brine system so we have a 1,200 gallon brine storage tank. The brine in a potassium formate, water based heat transfer fluid that is capable of negative 50 C to 218 C. And then we have a custom designed micro channel diffusion bonded heat exchanger that is designed for flow rates up to a peak of 20 kilograms per minute. That’s kind of going back to those DOE targets and metrics that I mentioned earlier because low pressure rate drop and 500 kilowatts of heat load absorption. You can see the heat exchanger in its uninstalled state at the bottom of the image here and then installed in its thermal jacket on the station or at the research station.
Here is our heavy duty vehicle simulator. This is basically simulating one class eight semi-truck. There are nine tanks total on the system for approximately an 86 kilogram flow. It is configurable so we can select which tanks we want to flow into. So one tank all the way up to the total of nine tanks. So it’s composed of seven type four tanks, about a 60 kilogram fill for that total system. Each tank is rated to an internal temperature of 85 degrees C. We also have two type three tanks for 20 kilogram fill and those internal temperatures are rated to 125 degrees C. You can see the configuration of the heavy duty vehicle simulator here. So the blue tanks are type fours and then the type threes are the green. We utilize automotive style on tank valves with integrated bulk gas temperature sensors and thermal pressure relief devices on each of the tanks. We have two triple point sensors installed within each, with each type of tank. So one in a type four and one in a type three. And you can see the configuration of those sensors here. We are working to get a vertical measurement installed on the tank but that’s one of the engineering challenges that we’ve been working through to get that sensor data. We also plan to install a thermal chamber around this system so that we can condition each of the tanks in a different thermal environment. We have additional pressure transducers installed at the rear of the tanks and we have a tank manifold that’s instrumented and built for consistent inlet conditions.
So having taken a look at the design aspects of the research station I would like to go through some of our exciting fast flow test results. So the industry and DOE metrics for the fast flow fills were to complete mass transfer of around 60 to 100 kilograms. We are looking back to those truck targets of an average mass flow rate of 10 kilograms per minute and a peak of 20 kilograms per minute. We wanted to do this in under ten minutes to be on the same time metric as a diesel vehicle. We want to complete the fill, excuse me, at 70 MPA down to a precooled gas temperature of negative 40 C and we want to maintain the vehicle. We want to operate within the vehicle tank temperature and pressure limits.
So these are our major fast flow tests that are completed to date. We have a variety of tests that have been completed. These are just the two major fast flow results that we’d like to share with you today. So the first was completed in August. That was a 61.5 kilogram fill. We achieved that in 4.7 minutes. So much faster than that ten minute metric there. Average mass flow rate of 13.2 kilograms per minute, a peak mass flow rate of 18.7. We have an ending state of charge for the tanks at 94 percent and so this test was only completed into the type four tanks. This is important for Dr. Kuroki’s work on the validating fueling models. We completed a second test in October and this was a benchmark test to try to see how much mass transfer we could achieve into the complete heavy duty vehicle simulator system. So we’re targeting over that 80 kilogram fill there. So that was 82.3 kilograms in 6.6 minutes, ran fast, an average mass flow rate of 4.6 kilograms per minute and a peak of 23. So happy to report that we meet and exceeded all industry and DOE targets for fast flow tests in the heavy duty vehicle simulator system.
So I’d like to jump into the data here and take a quick look at some of the results. So this is a quick summary here, kind of summarizing this would be specific to the 82.3 kilogram fill flow test. So that was using all nine tanks of the heavy duty vehicle simulator system. We started around 1.5 MPA and ended at 83.4 so it’s roughly 220, a little high there as the type threes can accept a little bit higher state of charge to align with the SOC of the type four tanks. So that translated to an average pressure ramp rate of 12.3 MPA per minute. We had an ambient temperature during the day or around 19 degrees C. So you can see in the chart here this is the heavy duty dispenser pressure. And so that is – we’re accessing the medium pressure system here and that is around that 6,000 – sorry for the mismatch in units here but 6,000 psi starting pressure moving through the medium pressure system, then accessing the high pressure system starting with the type two tanks where we pair those together. And so we generate this saw tooth pattern here of moving through each pair of high pressure storage tanks and then moving to the type one high pressure storage tanks where we get the small saw tooth pattern here and then moving on through that system to complete the fill. At the bottom of the chart we have the flow control valve inlet and outlet pressure as well as the hose pressure. So as you recall we do not have a hose installed on the system in its hard pipe but this is where the pressure indicator is located for where that device would be located would be located across the system. So this is just pressure loss that you’re seeing.
This is a quick image for flow control valve actuation so same pressure chart from the previous slide overlaid with value accusation. So for the medium pressure system we are opening that air operated valve to its full open state, accessing that medium pressure system and then slowly opening the mechanical flow control valve to the desired state to achieve a fairly I guess consistent pressure ramp rater across the system. And the actuation time of that flow control valve is pretty slow due to its size. So as we’re getting that flow control valve open, closing the air operated valve and then continuing with the fill slowly opening up that flow control valve until we reach the end of the fill.
So this slide shows the heavy duty dispenser temperatures. So this is the Joule-Thomson effect heating across the flow control valve showing on the top here. This is just an interesting aspect of hydrogen heating as it moved across that restricting orifice. The lines shown at the bottom of the chart here are the heat exchanger performance. So this is the bottom line showing the hydrogen entering the, exiting the heat exchanger system and then moving to the hose position so that that kind of theoretical hose position and then on to the heavy duty vehicle simulator. So this is the heat loss across that system there so relatively good performance.
This is the heavy duty vehicle simulator pressures overlaid with mass flow. So just a quick kind of recap to seeing that medium pressure system as we’re accessing that mass flow and then mirroring the image that you saw before for each individual spike of the high pressure system relatively good. Average pressure ramp rate up into ______. This shows a heavy duty vehicle simulator internal tank temperatures and so these are the temperatures taken from the on tank valves. Important to note here so the type four tanks are shown in solid lines. The type three are in dotted. And so for the type fours not getting anywhere near that 85 degree C threshold temperature for the type fours. And then for the type threes nowhere near the 125 degree C max internal temperature. There is some difference in these temperatures and that’s due to the heating, due to solar heating and so some of the tanks do receive direct sunlight currently at their location on the pad. And so you will see some fluctuations in those temperatures due to tanks starting a little warmer than the others that are shaded.
So moving on to the heavy duty vehicle simulator, these are the triple point thermal couple temperature readings inside the tank so that’s kind of across that horizontal plane, the heavy duty vehicle tanks. So the type threes again shown in the dotted lines again nowhere near that 125 degree C max and then type fours getting around maximum of a little over 70 degrees C so nowhere near the 85 degree C internal for this particular 82.3 kilogram fill. I’d like to briefly mention the station tank temperatures. So we are – because we’re depressurizing the station, research station side tanks so quickly we are seeing a significant drop in temperature of the station side tanks. So this is the medium pressure system here and getting down to nearly negative 10 C on one of the tanks. And then for our high pressure system nearly negative 30 for this particular test. So this tank does alarm at negative 40 C. And so the type ones getting down to negative 29. So there is a concern that on very cold days when we might perform these fast flow fills that the station side tanks might get into an alarm state. And so something that we can use to mitigate that would be pairing more high pressure storage tanks together but it is something we’ll be looking into as we move into the winter months.
I wanted to briefly touch on our engineering challenges before turning the presentation over to Dr. Kuroki. So we are working on switching strategies from the bulk gas storage system. We do recognize that this is somewhat unique to the NREL research facility but is something that we’ll be working very hard on over the next few months with research projects. Implementation of multiple sensors are required within the vehicle system to accurately observe internal tank conditions. So we have faced some engineering challenges with integrating those sensors. We do have pressure and multiple temperature measurements within each tank and we recognize that we need to integrate more. Reliability of OTV readings has been a little spotty as some of you in the industry may know and so it will be really important to make sure that we integrate those additional pressure and temperature measurements within the tanks so that we do not have to particularly rely on OTV readings. Flow control valve actuation times as I mentioned result in a major lag during dynamic processes. There is a potential – this could be a potential limiting factor to implementing new fueling protocols due to the speed of current technology. And then accuracy of mass flow reading for fast flow testing. So this will be required for a limitation of fueling protocols. NREL, we do have a mass flow reader out on order for the system but currently are waiting on that due to supply chain constraints. All right. So I will turn the presentation over to Dr. Kuroki for a summary of our modeling and computational fluid dynamics work.
Taichi Kuroki, National Renewable Energy Laboratory: Thank you Shaun. Hello everyone. My name is Taichi Kuroki. Thanks so much for having us. So from that I’d like to introduce our heavy duty fueling models. So this slide shows an overview of our heavy duty fueling models. Our DOE heavy duty project we are developing two models. One of them is a one dimensional hydrogen fueling simulation which is called H2FillS. The other model is a three dimensional computational fluid dynamics model which is known as CFD. These models are physics based thermal fluid models and these models are used to evaluate the changes in the gas temperature pressure in the mass and fueling system components during the filling process. From that I’d like to talk about the details of each model. Next slide please.
First I’d like to talk about our H2FillS model. Like I mentioned it’s the one dimensional thermodynamic hydrogen fueling model and this model is openly available to the public and you can download this model from the link in the first bullet point. But the current advisory was developed specifically for light duty fueling. But good news is that we upgraded that H2FillS model for heavy duty fueling on the DOE heavy duty project. And we called that graded model HD-H2FillS. And thought we are going to release the upgraded HD-H2FillS model to the public early next year. Next slide please.
These slide shows what we can do with the HD-H2FillS model. HD-H2FillS is capable of simulating the real work with hydrogen fueling process from the high pressure storage system through the vehicle during the simulation of our model calculation, the gas temperature, pressure and mass at all components from the high pressure storage system through the vehicle tanks. So this capability allows station developers and vehicle OEMs to understand how individual components affect fueling performance such as temperature and pressure rise in the vehicle tanks. So our HD- H2FillS models should be valuable for station developers and the vehicle OEMs to optimize their station and he vehicle designs. But around HD-H2FillS model can simulate the temperature distributions within fueling components such as temperature gradients in the vehicle tanks. That’s why we use CFD models. Before I talk about our CFD models I’d like to show you the reliability of our HD-H2FillS model that were compared with heavy duty fueling data collected at our heavy duty station. Next slide please.
And when we validated our HD-H2FillS model against heavy duty fueling data collected at our heavy duty station for us we set the specifications of that NREL’s heavy duty dispenser and heavy duty vehicle simulator. And then we set the fueling conditions to the model. The right table shows that fueling conditions we’ve set for the model validation. So those fueling conditions are based on fueling high flow fueling experiment. In that experiment the ambient temperature was 23 degrees C and the heavy duty simulator system size was 86.8 kilograms. And the initial gas pressure in the heavy duty vehicle simulator was 1.7 MPA. The pressure increased up to 76.2 MPA within 360 seconds. That corresponding pressure ramp rate was 12.4 MPA per minute and the time averaged mass flow rate into the heavy duty vehicle simulator system was 13.1 kilogram per minute and the peak mass flow rate was 23.2 kilograms per minute. And under all these fueling conditions and specification we run the HD-H2FillS model and we compared the modeling results with corresponding experimental data.
Next slide shows the comparison. So this time I’d like to show you on the comparison of the simulated and the measured heavy duty vehicles simulator gas temperatures because I believe many of you are curious how quickly the gas temperatures in heavy duty truck tanks increase during the high flow fueling process. And in the graph the solid lines show that measured gas temperatures and dashed lines show the simulated gas temperatures. In the dark, in the light blue lines show the gas temperatures measured and simulated in the type three tanks. And the other lines show the gas temperatures measured in simulated in the high flow tanks. As you can see there is a difference between the simulated and measured gas temperatures is large at the beginning of the fuel. But the difference decreases by the end of that fuel and at the end of that fuel the difference between the model and the experiment is almost negligible. We compare our model with other fueling tested data and then we ______. The accuracy of the model is always consistent so this is the reliability of our HD-H2FillS model. Next slide please.
From that we’ll talk about our CFD modeling work. CFD is thermal fluid ______ software that allows us to simulate thermal and flow fields inside, outside components. So this capability makes up for HD-H2FillS disadvantage. But running 3D CFD simulations are computationally very expensive. So our CFD modeling work leverages NREL’s high performance computing system or HPC. And we use it at this moment so as the figure in the slide shows we access our HPC with ANSYS Fluent our HPC runs 3D CFD simulation and the simulation results. Next slide please.
At this moment our CFD modeling work focuses only on heavy duty truck tanks. The other heavy duty vehicle simulator we have seven type four tanks. Some of them have straight injector. The other type four tanks have an angled injector. This is we modeled those straight and angled injector tank models and ANSYS Fluent and then we ran a CFD simulation based on one of our high flow fueling experiments. Next slide please.
This slide shows the thermal fields in the straight and angled injector tank model. The left graph shows the tank in the conditions that were used to run the CFD simulations. The movie above shows the thermal fuel in the straight injector tank model and the movie below shows the thermal field and the angled injector tank model. In the straight injector tank model there are less temperature gradients at the beginning of that fuel. And also you can see some hot spots on the upper surface of the liner at the beginning of that fuel. On the other hand in the angled injector tank model the thermal field is always uniform throughout the fueling process. So through this movies you can see the impact of the injector shapes on the thermal fields in heavy duty truck tanks.
So next we evaluated the reliability of our CFD models with the corresponding experimental data. In this graph the solid lines show that measured gas temperatures. Those gas temperatures were measured with on tank _____ sensors, OTB sensors. The right figure shows the OTB sensor locations. In the CFD models we evaluated the gas temperatures at the same locations where the OTB sensors are located. The CFD data are as shown with the dashed lines in the graph. As you can see both CFD simulation results are matching the experimental data but we can’t see any inference of the hot spot in the experimental simulation data of the straight injector. But at least we can conform our CFD models are reliability. So we evaluated more details of the thermal fields for our temperature field in the CFD tank models.
Next slide shows the maximum and the bulk average gas temperatures in the CFD tank models. In the graph the solid lines show the maximum gas temperatures and the dashed lines show that both average gas temperatures in the straight injector tank model. So the straight whenever you focus on the simulation results dealing with a straight injector tank model, that straight injector causes a large difference in the maximum and the bulk average gas temperatures because of the hot spots that can be found on the upper surface of the liner. On the other end in the angled injector tank model the difference in the maximum and bulk average gas temperature is very small. So the angled injector – so it’s mixing the hydrogen in the tank very well. So this is evaluation results of the CFD simulation results. So this also this comparison shows why running CFD simulations is so important. We can’t update these results only with the model HD-H2FillS. So this is the reason why we are liberating 3D CFD models from this DOE capability project. So that’s all I have for my presentation so I’d like to pass this presentation back to Shaun. Thanks so much.
Shaun Onorato: Thank you Dr. Kuroki. All right. So we’d like to briefly touch on our current and future heavy duty projects so this is sort of how we’re leveraging the results that you have seen today and moving forward with – so this is our heavy duty dispenser and nozzle assembly project. This is a partnership with Electricore, Bennet Pump Company, WEH Technologies and Quong and Associates. So this project is developing commercial level heavy duty dispenser from Bennet Pump Company. It’s a retail focused dispenser and then a commercially focused nozzle assembly which includes nozzle, receptacle, hose and breakaway and there’s also – so these are capable of fueling heavy duty vehicles so we’ll be testing and demonstrating these systems at NREL’s heavy duty research and development facility under real world conditions so that you can see that the dispenser is installed now on the back pad in front of the heavy duty vehicle simulator and that system was piped up with hydrogen and we’re doing some commissioning activities now. And later we’ll be connecting the WEH hardware for evaluation. So we’re targeting the same metrics that we’ve talked about today so 100 kilogram fill in ten minutes with 70 MPA pressures. This project is specifically tasked with performing SEAJ 2601category D fill and then moving forward with advanced protocols as they are available. And then it uses IRDA communications and you can see these components here. Those were also shown in the annual merit, DOE annual merit review hydrogen slides.
The second project is the heavy duty fueling methods research project. This is another cooperative research and development agreement that’s industry led. And so this will be doing the assessment of heavy duty fueling protocols. So we’ve done that baseline work of evaluating can we do the fast fill, how fast can we do it and maintain conditions within the vehicle storage tank. So this is taking a look at applying fueling protocols, kind of dialing back that fast flow to more of a controlled state, looking at heavy duty station architectures, functional safety requirements and implications of these things on the total cost of station ownership. So there’s a very heavy modeling component to this project so we’re looking at techno economic assessments and TCO as well as evaluating a second set of industry provided hardware. This is from _____. This is a heavy duty nozzle, receptacle, breakaway and hose set that we’ll be evaluating on the NREL station. This project has very close collaboration with the EU pride initiative, the protocol for heavy duty hydrogen refueling under which new advanced fueling protocols were being developed for heavy duty trucks. This projects also explores advanced communication strategies beyond IR. So you can see some of the partners here so the primary being _____ Energy. That representing a larger group of industry folks that are closely working with Pride. We also have Chevron and Argon National Lab.
So I’d also like the briefly mention our advanced research and integrated energy systems which is the ARIES project. This is a kind of by grid level large scale project that I was talking to you about at the beginning with the NREL infrastructure research. So there is a large hydrogen component to this. So it’s a research platform that can match the complexity of modern energy systems and conduct integrated research to support groundbreaking new energy technologies. So we have a 1.25 megawatt PEM electrolysis, over 600 kilograms of ground storage, high throughput compression and a one megawatt PEM fuel cell on this. This is located at NREL’s second campus on the wind site. And so a component of this project will be that high throughput compression that will be leveraged for some of our heavy duty activities moving forward with plans for some other heavy duty activities at this campus later on.
So this is a future work summary so in terms of hardware we’ll be performing additional fast flow tests to generate data for model validation. We’ll be evaluating advanced protocols with industry supplied heavy duty component sets that you just saw on the previous slides. So these include nozzle assemblies and the commercial dispenser. The protocols will be from Pride, SAE and other partners and we’ll inform SAE and ISO working groups for fuel and protocol standardization. On the modeling side we will be validating that heavy duty version of HD-H2FillS against heavy duty fueling data and releasing that sophisticated model to the public for use, assisting SAE and ISO in development of heavy duty protocols. We’ll be expanding our modeling capabilities to include liquid hydrogen as well as marine and aviation and rail, perform the technoeconomic assessments and total cost of ownership that we talked about under the HD fueling methods project to assess HD infrastructure needs based on hardware work, modeling and industry feedback.
This is a quick NREL test matrix just showing kind of where we’re at so that baseline performance evaluations are complete and then moving forward with implementation of fueling protocols and then testing those two different hardware sets into the winter of this year and spring on into summer. And with that I would like to conclude the presentation. And so it takes a team to perform this work and so I would like to thank the team shown here in the image so Sarah Knowles, Owen Smith, Matt Ruple, Taichi. Jeff Moore, Josh Martin. Katie Hurst, Daniel Leighton, Spencer Gilleon. And those not listed here, a huge effort on our team’s part to bring you the research, the exciting research that you see here today. So if you have any questions feel free to reach out to me, Shaun Onorato for the heavy duty technologies area, Dr. Taichi Kuroki for system modeling and CFD and our group manager Katherine Hurst. So with that I’ll conclude. Thank you.
Mark Richards: Thanks. There’s a lot to absorb there. We’ve got a fair number of questions in the chat, in the Q&A. I don’t know if we’ll get to them all but we’re going to have a go. So first one I’m going to pick up is were there any concerns raised by the projects industry partners related to the temperature and pressure profiles seen by the HPBS?
Shaun Onorato: Not that I’m aware of. So you’re asking if there were any concerns with the temperatures seen in the tests?
Mark Richards: Yeah.
Shaun Onorato: Yeah. I think the no concerns I think more excited that during the fast flow tests that performance of the system was quite good.
Mark Richards: Ok. And does the test system have communications between the storage system and the station?
Shaun Onorato: There’s currently direct line mod bus communication. So we are working on integrating IR at the moment and then moving to those advanced protocols when they’re available from the other industry partners.
Mark Richards: Ok. Thanks. And the equipment used in the facility was it off the shelf largely or were there any custom made components?
Shaun Onorato: Mostly off the shelf.
Mark Richards: Like the storage system.
Shaun Onorato: Yeah. So mostly off the shelf in terms of the storage system. You have commercial grade automotive components for the OTVs. Yeah. And then as I mentioned the heavy duty dispensers those were the largest components for in terms of valves available at the time. We did have the custom micro channel heat exchanger was custom.
Mark Richards: All right. And have existing vehicle manufacturers indicated a willingness to standardize parts needed for heavy duty refueling? You may not know the answer but.
Shaun Onorato: Yeah. I unfortunately probably can’t provide too much insight into that other than to say that we are participating in the codes and standards efforts and that testing that we do here at NREL will play a role and will feed data into those entities for standardization activities.
Mark Richards: Ok. And is the angled injector approach used anywhere in industry that you’re aware of right now in the field? Yeah.
Shaun Onorato: I’ll let Taichi take that one.
Taichi Kuroki: Yeah. Actually I’m not sure but I heard some of the vehicle manufacturers use angled injector to enhance the vehicle tanks.
Mark Richards: Ok. Let’s see. We’ve got a whole bunch more here. So do you do any monitoring for fugitive emissions given the size of some of the valves and things like that at the moment?
Shaun Onorato: We do. So we do pretty frequent leak checks at the research facility. We have the research station is very highly instrumented to detect gas. So there are local CGs on each of the individual subsystems as well as Fire Eyes, acoustic detectors, leak detection so highly instrumented on the back pad for detection of leaks. And happy to report that we don’t have anything to report.
Mark Richards: All right. There’s a handful I’m going to group together that kind of have to do with the test setup. Anything besides temperature and pressure that you’re monitoring? I think the question mentioned weight but I don’t believe you’re actually weighing any of those tanks on the simulator?
Shaun Onorato: No. Yeah. That is just, yeah, temperature, pressure, max flow.
Mark Richards: Ok. You had plans to put temperature sensors in the simulators in various locations. One question related to well, it would be nice to have one where the CFD is showing hot spot. Do you want to comment on that?
Shaun Onorato: Sure. And I think Taichi feel free to jump in. I think one of the engineering challenges that we have faced as well as with other groups that are doing similar work is implementing the thermal couple trees into the vehicle storage tanks. And so just from a physical engineering aspect it’s been difficult to try to get those sensors located within there, trying to get the sensors inserted into the tanks, not damaging liners, coming up with a robust design that won’t leak with a significant number of sensor locations being able to implement that has been a challenge.
Mark Richards: Ok. And Taichi I guess for you in the CFD model can you comment about what kind of turbulence model you use? How many elements are there in the model and what’s the actual run time for the simulations?
Taichi Kuroki: Ok. Yeah. We use ______ turbulence model, number of measures around 800,000 nodes. So I didn’t mention but I was going to mention it takes around 20 to 30 days to receive the data.
Mark Richards: And this is one high performance computing platform.
Taichi Kuroki: Right.
Mark Richards: Not something a lot of folks have access to. Correct? Ok. Of course we always get one of these. Where can I get the experimental data?
Shaun Onorato: Sure. So we are working to try to get some of the data on a public access point and so it’s something we’re still working through. But hopefully we’ll have something up on the website soon as we work to publish this information.
Mark Richards: Ok. I think we have time to squeeze in a couple of more. There were a few related to back to back fueling but I’m pretty sure you mentioned that the system is not really set up to do that. This isn’t a commercial fueling facility.
Shaun Onorato: Yeah. So for full heavy duty fills we can only perform one fill a day. Recovery time for one fill is significant. So that takes quite a while to get all the high pressure systems back up to full pressure. We could perform multiple fills a day if they were smaller. So you maybe factored into like 30 kilogram fills, maybe one of the 60 kilogram fills, we might be able to get two or three in per day. But the full 83 kilogram fill is one per day and then a recovery time.
Mark Richards: And kind of related to that we had a question about making sure that you don’t have over temperature of tanks. Obviously the chiller can get new gas at minus 40 C but were there other challenges that you faced in doing these fast fills to maintain that temperature or ensure that temperature limit wasn’t crossed?
Shaun Onorato: Yeah. So a significant amount of controls were each fill is Dr. Kuroki runs a custom simulation to check all the various parameters that we want to achieve because we’re doing this without a fueling protocol. So we’re basically basing things on SEJ2601 but it’s a loose correlation. And so Dr. Kuroki is running a custom simulation the more perform the fill and then feed the data back to the model. You’ve talked about some of the engineering challenges. So very slow flow control valves at these sizes, having to perform that fill and just do that slow actuation versus more of a dynamic process, looking at things like station side storage tanks getting too cold and then the other side of making sure that the heavy duty vehicle simulator tanks aren’t getting too hot. So a variety of things that come into play for each fill.
Mark Richards: Ok. I think we have time for one last one. Can you speak a bit about why you hard plumbed the station to the vehicle simulator?
Shaun Onorato: Yeah. So as I mentioned the components for refueling at these sizes for these mass flow rates are not available. So they are not available in a commercial scale. And so the first few tests that we completed needed to be completed at a hard pipe scenario. But as we talked about in the latter slides we do have those components that we’ll be evaluating moving forward.
Mark Richards: Ok. Well I know there’s probably a few questions I didn’t get to. I apologize. I’m glad there’s so much interest in this topic. I’d like to thank Shaun and __ gain for all the great information and for your time today and I’m going to hand it back over to Cassie to wrap things up. Thanks.
Cassandra Osvatics: Thank you Mark. That concludes our H2IQ Hour today. Once again I want to thank the presenters from NREL. A recording of today’s webinar and presentation materials will be available within the coming weeks. Be sure to subscribe to HFTO news to stay up to date on that and thank you all for attending. Have a wonderful holiday season. We’ll see you in the new year for more H2IQ Hours.
Mark Richards: Easy for you to say.
Cassandra Osvatics: Yeah.
Mark Richards: Happy holidays.
Shaun Onorato: Thank you.
Taichi Kuroki: Thank you.
[End of Audio]