Tech Talk - Fuel Cell Turbo Compressors - Fuel Cell Technology Explained - Hyfindr Brandstätter
Hello. My name is Steven. Welcome to Hyfindr Tech Talks. Today, we are going to be talking about turbo compressors and with that is the person here, who has a Ph.D. from arguably Switzerland's most renowned technical institution. He currently works at severity that's the ETH actually what I meant. He currently works at Celeroton, where he is head of the R&D for fuel cell applications. Welcome Markus, Markus Brandstätter. Hello, Steven. Thank you for having me. It's great to be on your show.
Thank you, Markus. Yeah, you are connected to us by video link and you are in a special place. Can you tell us a little bit where you are please? Yes. I'm currently in one of our main working session rooms. Right next to the entire research and development department of our company, where all the concepts or the new concepts that we developed are being brainstormed or are being presented. Okay, wonderful. So, that is just next to all the R&D guys, I know you head off the team but the good news is in case we run into any serious technical question here you could quickly probably hop over to your colleagues there, right? Exactly, then I can grab my specialists and take them in it. Okay, so perfect let's go right into the topic. Markus, what is a turbo compressor?
Actually, I for that matter brought a slide with me or an image that I could show you to give you a little bit of an overview in the house that you could talk compressor looks like and how it functions. Okay, yes. We see it. So, you can see. Alright, great. So, on the left hand side you can see a simplified schematic overview of how such a turbo compressor is built. So, fundamentally it consists of three or in a simple form it consists of three parts, which is the impeller. So, the wheel that works on the fluid. The diffuser which is followed by our impeller and after the diffuser, we have a so-called Volute or spiral casing, which connects the compressor to the fuel cell stack for instance. So the impeller is number 2 in that image. Impeller, you can see it right here. Yes.
The diffuser is number 4 and the spiral casing is essentially number 6. Understood. So, maybe and now we can briefly go through step by step, what happens like from from a fluid perspective. On the right is inlet which is indicate the number 1. Going essentially directly flow is directed towards the eye of the impeller one is talking about the eye which is the center of the impeller. High rotational speed impeller is rotating we have a so-called flow passage which is generated by the blades of such an impeller, were we have basically an increase in or a change in angular momentum of the flow and the change of total pressure. So, across the flow passage which is created by the spiral case, which is the outer contour and input around 60% of the total pressure increase is taking place. Okay, so as that spins that increases the pleasure going out.
Correct. So, it's essentially first of all it changed the angular momentum and it increases the total compression. What is interesting to note after for instance the edge of these of this impeller we still have flow velocities of around 200 meter/second, which means we need to still decelerate this flow and to further increase the static pressure that is required for certain applications. This is done by the so-called diffuser, where we increase the volume thereby increasing the static pressure. After the diffuser, we have as I said spiral casing or the volute where we further reduce radial velocity and collect the flow towards the end application, which could for instance be a fuel cell stack.
Maybe on the right hand side you can see one of our more recent design and that we have already presented at numerous fairs, where we can see actually how it looks in real life, which is a cut through one of our CAD design. So, maybe I can also show you how typically that like what the variation of such impeller designs, which is from my perspective very interesting just to see that you can see there's a huge variety. Mixed flow designs, where we have also radial and axial components or the classical design, which one would see in fuel cell stack applications. So, what is the difference between you know one that narrows up and the one that's completely widening, what's the main difference? So, the main difference is directly it's a very technical question. Actually, the difference is directly to the flow coefficient. It's essentially a throughput, that's
such a compressor needs to be able to designed for. For instance, once we go towards axial flow designs, which is a classic for gas turbine designs for instance that high mass flows requirements. Whereas, when we go through really radial designs which is a purely radial design pretty even we have very low mass flows. Alright. Okay. So, the mass flor essentially makes a difference. Okay. Well, you have already mentioned some applications, where do we typically find
turbo compressors in the hydrogen economy. Yes. I mean there are three main applications actually and also actually for that the purpose I also brought you another image that I could show you, which is a very simplified schematic of a proton exchange membrane fuel cell (PEMFC). So, a piping diagrams essentially where we can see the air inlet. And here we have our air compressor where this is one of the main application areas for these turbo compressors compressing the air generating for the cathode side to produce the air or eventually that oxygen that is required by these fuel cell stacks. Yeah. So, we see that it pushes, the
air into the humidifier and then into the system and it needs also cooling as I see from that diagram. So, I know you're an expert. So, can you take us a little bit more deeply into the actual unit? So, we saw the first picture but please show us a little bit more about the components of it, please. Yes. So, let me briefly one more point to clarify so I mean turbo compressors can be used on the cathode side but in addition also on the anode side. For instance for a hydrogen recirculation with dedicated designs or for electrolyzer hydrogen recirculation. So, you also could use the turbo compressor for recirculating the hydrogen.
Absolutely. Yes. Dedicated designs though but my business unit is really focusing on the air paths on the cathode side but generally speaking turbo compressors can also be used for hydrogen recirculation. Understood. So, back to the original question of yours up into the more detailed view of how the turbo compressor system looks like and what I brought here is actually a cut through of one of our compressor systems. So, first I'm going to zoom out a bit. In the beginning, I only talked about the aerodynamics which is only a portion Obviously, it's an important part but it's only a portion of what the compressor system is consisting of. So, on the left hand side we can see the compressor
and on the right hand side we can see the converter. So, maybe let's go from left to right. So, I already explained to you the impeller the aerodynamics and the impeller is usually connected to a rotor which rotates at a high rotational speed. The rotor itself is suspended by so-called gas bearings which makes this entire thing special. Why does it make it special, because it's generating compressed air which is oil-free and particle-free. And virtually no wear. That's the second bit. Then obviously the torque needs to be generated to compress this air right and the torque is generated by an electric motor. In this
particular case, we can see it's mounted around or in between the two radial bearings which is our winding and in the middle we have our magnet or a permanent magnet arrangement. Yes. So, we are talking usually about permanent magnet synchronous machines that we use here, which are especially designed for high rotational speeds, which where we have to take into account a couple of intricate things in the design process.
Yeah. So, actually you mentioned high rotational speed so I'm just you know trying to imagine being at the impeller, so that gets up to a very high RPM stage and you also mentioned some really fast speeds there. I mean don't you reach the sound barrier there? And things like that don't you get into other kind of physical problems because of that? So, maybe to give you a couple of values here, which might be interesting for the viewer as well. So, we range between rotational speeds of so for the larger compressors around 120 kRPMs or 120,000 RPM up to 300,000 RPM for the smaller systems. Usually
which means when we convert that into into velocities we're talking about 300 to 400 meter per second (m/s). Wow. So, generally speaking of sound barrier is something we try to avoid because supersonic operation always results in shocks. So, compressibility effects shocks always are associated with irreversible processes and ultimately also in a reduction in efficiency. Okay. So, that's we are trying not to operate in the supersonic. And I see there was cooling I saw from diagram. So, what is actually being called is the electric motor that that's what's being cooled? So, obviously within that system, we have a couple of of sources for losses and one is for instance the gas bearings, they generate losses but mainly it is the electric motor as you indicated and well we either can cool it with water or with a cooling medium or it is air cooled. Okay. I'm very curious
about the gas bearings, Markus. Maybe you can tell us a little bit more about that because I know you also have a bit of a specialty around that. Firstly, how does it work and also I mean if it's a compressor it's actually has one chamber where there's air. Isn't that the interaction with the actual medium that is being pumped? So, we have the same medium in there obviously. Yeah. So I have to briefly have to check, whether I have a slide that I could show you. Yes, here we go. So let's dive into that the details of gas bearings. I mean gas bearings are
very interesting. Generally, speaking one can differentiate between hydrostatic and hydrodynamic bearings for mobility applications which is the most current case for instance for road applications or aerospace applications. Hydrodynamic bearings is the usually that the one uses the reason being simplify design few external parts and so on. How does it work? One has essentially a shaft and then we have one has a bushing and through rotation one generates a pressure through the rotation within the small gaps between the journal and the bushing, which generates a cushion where the journal floats on top. That's essentially
how it works fundamentally. So, by spinning the fluid around or in in the gap of these two you create a cushion that is strong enough for that to carry it. Correct, exactly. I mean I can maybe briefly show you. I mean there are different types of gas bearings one could for instance use the simplest one. One can think of is a plane bearing where essentially just as the
cylinder and a bushing which is a a pipe so to speak and just by rotation and the Couette flow generates a cushion, where the rotor starts floating. Strongly and narrow regime where one can operate these plain bearings. And for instance the technology that we mostly use is Herringbone grooved bearings, where we can engrave structures to increase the regime where these systems can operate. Okay and
when you have this kind of like I said if you have a pressure chamber on one side where it's spinning essentially. Do you see any effect of the air? In this case, you're moving air but do you see any effect of that in the bearing when the pressure rises in the spinning. No, there is no direct interaction. Okay, yeah and so essentially this floats then and just using the opportunity here, what happens when like you know the compressor shakes or gets shocks and so on? You know because I mean obviously, if it's a physical bearing let me just call it that you know a ball bearing or so everyone knows okay there is something but here now we are floating on the cushion. How does that work? Yeah, one is when actually when I talk to other engineers, it's always surprising to them that these gas bearing type of bearing. I mean I can briefly show you another image. They actually have stiffnesses, which are comparable to ball bearings which is for some people very surprising. Which means how does it work. So, by rotation we have this gas film where the journal is floating and once
we have a vibration shock. So, it's essential displacement. On one side we decrease the gap of this gas film which increases the stiffness of this gas film. Thereby, exerting a force on the journal in the opposite direction which means there is always a force acting into the centering of the journal when we have an external force acting and it's such as vibration or shock. Here, on the right hand side you can see actually one of our compressors being tested on a vibrational bench, where we for instance tested systems up to 20G under real conditions. So, this is clear that it's the shock perspective. Okay. That's very interesting that is that kind of you know resistance. What about load bearing? Sorry, if I ask but like you know
so what if it's a very heavy shaft or something. These air bearings fine for that as well. I mean you have to be obviously for it. Yes. The size increases and there needs to be obviously the attention to how you design such a bearing but in terms of forces, you can definitely design also for larger shafts. I mean for instance here we tested up to
20G acceleration which means the forces of this rotor under acceleration is quite large. Okay, last question for the air bearings. what speed does it have to have to actually start working? You know because I was just going through my head like it could be using another application but it doesn't need to have a certain speed right before it starts. Yes, I can
briefly elaborate on that as well and actually that's a very common question, which is why I also brought an image that I could show. Oh wow. Typically speaking when we look at curves when we actually accelerate and increase the speed of such gas bearing type bearings. We have a mixed regime, where essentially the rotor is still not yet floating and we have a sort of liftoff speed, which is at a very low rotational speed compared to maximum and after that essentially the rotor floats and has virtually no contact anymore. Okay wow. Well, no contact that sounds like easy maintenance and these kind of things, which brings me to another question Markus. I know that the compressors are often like when we say one of the cost drivers in the whole especially fuel cell systems. Can you elaborate a little bit on that like why or what is it that's so expensive on a turbo compressor like this? Just for fun. Maybe I can
give you a little bit of an insight into how much value is also given by such a compressor. Let me briefly show you one of the images here. So, for an entire fuel cell system component share value for the air compressor is up to 20% when we talk about the entire fuel cell system which also and furthermore not only the value that is created by the compressor but also the power consumption is also 10% to 20% of the entire system which also is a strong argument why these compressor systems are worth to further develop and improve. Which is why when they are a substantial part of the fuel cell system which is what I'm trying to say. Okay so that means you guys are working on getting them even better hopefully by bringing down the cost as well. Can you give us a bit of an outlook on what you're working on? Yes, I mean there are a couple of things in the market and in the development of what is happening these days, one thing is obviously that the market is moving to larger and larger stack size for instance that's one of the development paths in the industry and in the market. The second bit
is for instance a continuous integration of for instance electronics into compressors for a decreasing or downsizing of everything to save space and I think that the last important bit is industrialization. So, further being able to produce in a process stable and cost efficient manner products for the hydrogen market. Wow, that sounds like a lot of work still for you to do and the guys around you. I know that in the area of industrialization Celeroton has actually I've personally seen you move from generation to generation you know integrating more stuff into one unit and making it even easier to fit in. So, good on you and also good on you for sharing your time and your knowledge during the session. We've unfortunately already reached the end of the time that we have Markus, but I guess this was very insightful. So, thank you and
not only thank you to you but thank you to the viewers for watching. I hope that you enjoyed this video. If you did, please drop us a line or subscribe to the channel here. I can tell you that when you go on Hyifndr.com, you can find many components just like this one that we've talked about. So, you can get into contact with people like Markus or his colleagues at Celeroton. It
would be a pleasure to have you there. Please enjoy working on stuff that makes the hydrogen economy and everything else work. We look forward to seeing you in another video. Thank you again Markus. Thank you for having me. And thank you for watching. Have a wonderful day. Thank you. Bye. Bye-bye.