Episode Transcript
[00:00:10] Speaker A: Welcome to the EU Energy Projects Podcast, a podcast series from Enlida and France focusing on the clean energy transition for the European Union and the EU Commission funded energy projects that will help us achieve it.
[00:00:29] Speaker B: Hello and welcome to today's episode of the EU Energy Project podcast And I'm joined by Bjorn Rannestadt, Innovation manager of the SickFagrid project, who's going to give some insights and how it is enhancing the integration of renewables to the grid.
Welcome Bjorn.
[00:00:48] Speaker C: Thank you.
[00:00:49] Speaker B: Can you please start with a brief background introduction to the aims of SIC4Grid and I assume I'm pronouncing it correctly, abbreviated form.
[00:01:01] Speaker C: Yeah, I would say you pronounce it correct. So the abbreviation of course is playing on the words silicon carbide and then for the grid it's a Horizon Europe project which has been running now for almost four years and it's revolving around developing 3.3 kilovolt silicon carbide technology and modular converters for the future renewable energy and the medium voltage DC grids.
I would say a little different to many such projects. This project take more or less the whole value chain into account, all the way from the sick silicon carbide materials themselves on semiconductor level, on wafer level, to power modules to power converters, but also taking the grid into account and also handling of data, load data, also the whole design phase of the converters systems and so on, and taking all of that into account, not only focusing on the semiconductor itself but also the whole chain so that we can get, or the aim is to get more efficient, compact, reliable and sustainable power conversion at the same time strengthening the European sili carbide value chain.
So that's the background. And the parties in the project or the partners in the project are also from very different areas, which reflects what I just mentioned. There are some universities, three universities University in Brussels, Free University of Brussels in Belgium, Aalborg University in Denmark and then Mondragon University in Spain.
And then there are many, let me count, there are nine industry partners.
Also taking all the way from the low level, let's say value chain and all the way up to the large system. So starting from the low level, let's say SOI Tech and csem, Soitec in France and CSEM in Switzerland. They are working on the semiconductor materials, the substrates for what is called the smart SIC substrates and the whole semiconductor technology.
Then a little higher up in the hierarchy of the products, we have Hitachi Energy which takes the chips and the semiconductors are developed or the substrates, sorry, developed by ZOYTIK they then develop it further to get the real semiconductor chips and then putting them or packaging them into power modules.
In order to get such systems or the power modules up and running, we need a gate driver. So we have a Mantis from UK who makes gate drivers, smart gate drivers with some monitoring features.
Then we have powercon and KK Wind Solutions or KK Group. I represent KK Wind Solutions and we change the name during the projects. That's why it's a little misleading. But KKWIN Solutions or KK Group, we are both, our companies are making the converters both for wind and for other applications industries.
Then we have ITML from Greece who is doing the digitalization. Because also a big part of this project is not only the device itself or the system itself, let's say the hardware, but also getting data out. Because when you can get stream live data or data which you can evaluate later so that based on that you can look at lifetime issues and remaining lifetime of the components, load conditions and things like that so that you can optimize on the load not only in the design phase, but also during the operation phase. So that's ITML from Greece doing all this digital stuff. Then we have Euro Quality from France who is more doing the project support, dissemination and exploitation. And lastly EDF from France who have been involved predominantly in the specification of the project. They are the utility and the end user of all of this. So you can see going from substrates all, all the way to EDF who owns wind farms and solar farms and so on.
[00:05:16] Speaker B: Okay, we'll go into some of the more details of that later in the conversation. But to just go back, why silicon carbide and what are the particular properties that are being benefited from in this project?
[00:05:30] Speaker C: In the last, let's say few or many years there has been a lot of research on what is called wide band gap devices and silicon carbide is one of such devices. And for the power levels and voltage levels which are relevant for this project, the silicon carbide technology is the one. Silicon carbide, let's say in short, is a wide band gap material which has some different properties compared to the conventional or today's technology based on silicon.
By having a higher bandwidth voltage bandwidth, it allows for faster switching and also lower conduction losses at the same, let's say within the same building size, so on. And as part of that it also opens up the possibility to increase the voltage both in absolute level but also relative to the loss. So in essence we are looking into increasing the Voltage compared to, let's say the conventional, which is typically around 1700 volt devices. Now we move up to the 3.3 kilovolts, so a little more than, or let's say doubling of the voltage level at the same time reducing the losses.
And the reduction of losses, as I said, is due to lower conduction losses, meaning whenever the switch is conducting some current, the losses are lower, but also when the switch is turning on or turning off.
And this allows for a higher switching frequency. And all power converters, whether they are for car chargers or mobile phones or wind turbine converters or whatever, relies on some relatively fast switching. So some thousands of switchings per second is needed. The faster you can switch or the more often or the higher switching frequency you can run with, you can reduce passive components around it that could be capacitors, inductors and other components.
So by utilizing a higher switching frequency at the same time reducing the losses, we can also reduce the size of some of the components around the semiconductors. So that is predominantly inductors and capacitors.
And this have a high or is beneficial when at system level, when we want to exploit the possibility of going from, let's say, the more conventional AC distribution to DC distribution.
Maybe I should explain that briefly because it's embedded in some of the use cases for the projects, which I can explain a little later maybe.
But typically in today's grid, most of the distribution is done by ac.
By putting converters into the system, we can allow DC distribution, which has further some improvements of the transmission losses and the utilization of the cables and so on.
[00:08:31] Speaker B: And what then are the particular use cases that you're looking at testing these in or are being tested in?
[00:08:39] Speaker C: Yeah, talking about testing, I think there are some of these use cases are actually very large. So not all of the use cases are tested. We're going to test some blocks of the use cases. But the use cases defined were three different use cases, all in the renewable space. One use case in solar distribution or distribution grid around the solar panels, and two use cases related to wind.
All the three use cases have the same in common. What I just mentioned, taking into account the possibility of having efficient converters, let's say higher up in the voltage chain, so that you can, let's say, push the DC to a higher voltage. And let me explain that in a simple way. Typically the, the distribution grid is built up in a way so that to all the equipment, let's say all the wind farms or the PV farms, you have an AC distribution grid. Then from of course, from the very high voltage level and then down downstream in and, and down to the low voltage where you have the wind converter or the PV converter typically operating up to 1500 volt on the DC link, 690 to 1000 volt AC.
So from the low voltage side of the wind or the PV you have a lot of 50 or 60 Hz transformers and then you have AC distribution.
When you have wind farms far out, let's say offshore, you need to go from AC to DC and that is typically done with high voltage DC stations. But this is by itself a very expensive and complicated thing.
So what we are exploiting here is to say, okay, can we either completely get rid of the HVDC station and make a more simple setup by going straight into D.C. from the wind turbine itself.
And in this case the DC will be somewhere from plus minus 40 kilovolts all the way to plus minus 160 kilovolts depending on, let's say, the details of and which of the use cases. In the same way, on distribution for PV, typically the, let's say local distribution is done on 33 kilovolt ac.
We are now looking into the possibility of going from the 33 kilovolt AC into plus minus 40 volt kilovolt DC.
And why is that beneficial? It's beneficial because let's say from the cable point of view, from the loss point of view in the distribution, by having a DC on +/40, you can reduce the cabling compared to an AC system and you also get a pure power distribution. You don't have blind power or reactive power which is just circulating in the AC grid.
So the three use cases are one for an mv, so medium voltage DC PV farm, a solar cell farm, and then two different use cases for the wind. The difference between the wind demonstrators or the use cases is that in one of the use case we go directly into high voltage or medium voltage DC from the wind turbine itself by means of the ANC converter and then a transformer which with a rectifier on the secondary side. The other case is with the use of a solid state transformer, meaning we have active switches on both sides of the transformer operating in several kilohertz compared to a 50 hertz transformer. So by that we can reduce the size of that transformer, we can get directly into DC distribution and we don't. Yeah, so you can make the cabling more effective and potentially get rid of the HVDC power station or substation in the wind farm.
[00:12:40] Speaker B: So what findings have emerged so far?
[00:12:47] Speaker C: Yes, the project has as I said it has been all the way from the substrate level all the way up to these use cases. And as I said, the use cases themselves have not been built or demonstrated. That's huge projects by themselves. So on modeling, a big part of this project is actually modeling but not in a simple ways to say okay, we just make a. Let's say a simulation model and see how it goes. It's very much on the optimization making models on different level. We call from high fidelity models to low fidelity so that you can very efficiently run through all sorts of scenarios in terms of how not only the operation the control schemes, but also the dimensioning of the silicon carbide switches, the passives around it. The findings we have done is that it seems that this. This hypothesis we've had, let's say that this is possible from an. Also from an efficiency point of view seems to work. We have. We have a building block called the solid state transformer which includes it's a transformer itself but also the switches around it and so on. And these building block of around 2.5 megawatts. And we've run large optimization schemes on these and found that we can in a very wide operational range and so on get to more than 98% efficiency of these solid state transformers and these modules.
So that by itself let's say shows that the. The fundamental idea of the use cases and so on seem to be valid useful. In addition to that the on the semiconductor level SOI tech had or has this fundamental let's say technology or idea that they could they what they call smart sick. So typically when you make a silicon carbide substrate that's a very expensive part of process and due to the handling and so on of it, it becomes. It's very time consuming energy consum and very delicate in terms of let's say failures and so on.
And SoyTech has another way of doing this or the idea, let's say was to have another material as a base material and then just a thin layer of the silicon carbide itself. And by doing that you can get up to 10 times more utilization of the. Let's say the fine crystal material. And they have shown it's still something which is going on. They have made a first, let's say batch or a version and that now we're working into the second generation of it and so on. There were of course some. Some challenges in the first. There were some. Some slightly higher let's say failure and so on, but they have really got up to speed on that. And then Hitachi as well have taken that and developed further on the semiconductor itself and so on. So this has proven to be a very good, let's say product or technology and shows that there is a high potential for these silicon carbide devices with which at the end of the day will be much, will be cheaper, but also in terms of energy consumption, not during the operation in the wind turbine or the pv but during the production phase is going to be reduced compared to let's say today's technology, if you like.
Then we have also worked a little or some on circularity or let's call them a little more soft if you like values they're not soft, they're just as hard as the others. But compared to the physical products and so on it's a little but.
And there has been different work packages topics on that as well and we've seen that let's say this is positive also in the circularity perspective, although there are some practicalities in the final stage of the project, which is still a bit challenging.
The aim of the project is predominantly in the design itself and the lifetime. Very efficient design, efficient manufacturing, long lifetime and then possibility of actually prolonging the lifetime due to the digital feedback of the system, let's say, so we can monitor what's going on and so on. All of that is very strong. But then the final part of when you at some point you have to disassemble and so on, then that is still a little weak I would say not necessarily different to any other power, electronic devices and so on, but it's around the same, let's say. Yeah.
[00:17:13] Speaker B: Can you add anything further about the circularity aspect, how that's being achieved?
[00:17:20] Speaker C: Yeah, we could put in this way that it has been achieved in the sense that we use less material predominantly use it for longer time and recover more. That these could be the three sort of things we want to do and the two first ones. So use less material and use it for a longer time seems to be achievable. When I say achievable, of course it must be understood that until now it has been very much modeling. So it's, let's say a digital model and we will soon move into some measurements as well. So far everything builds up to the same story, but it's not tested even for 20 years of operation and so on. So still, of course, as always with research we assume a couple of things, but under these assumptions we use Less material, we can use it for a longer time. But on the recovery part, recover more is still something which I think we need to work on. This is a general topic in the industry today.
[00:18:14] Speaker B: I think then are the longer term plans both for further developing the technologies but also potentially commercializing them.
[00:18:25] Speaker C: In general there is a general goal of working towards higher TRL levels. So we have built a laboratory demonstrator model which is going to be tested in roughly two or three weeks. We're going to start up tests. We have gained a lot of experience on the module level and so on. And now of course we have the whole value chain going from, from the chip manufacturer and all the way to the power converters.
So depending on where you are in that let's say chain there is also different time span. Let's say the expectation is that on the semiconductor level and power module level this is moving towards a full something which could be mass produced within. I will not put up numbers here then I think Hitachi and so on, this is part of their own internal planning and so on. But in general we could say it's moving in that direction.
On the higher level, system level it has longer time spans because then it's first of all the projects are larger, this is grid scale and so on.
So the specific plans have not been put out let's say. But we are as companies and also the universities are continuing the let's say the research on this and interest in this topic. And I am let's say personally quite sure that the sili silicon carbide will enter the converters. Whether we will go all the way to taking all these advantages into account, let's say on the short term it's going to be difficult to say but at least starting out from, from the lower level semiconductor level and then taking it into the power converters, I believe that is is going to happen let's say relatively soon.
And as a company now I will take the company hat on. I can tell you that this is going on and I know also from the universities there is no, not an upcoming project in this consortium planned at least for now.
But I am also convinced that we will now in the final phase of the project we will continue to discuss what are the next steps and so on and I'm quite sure there will be something coming.
[00:20:40] Speaker B: But there are a number of European silicon carbide projects in the energy sector and how does sic for grid complement these and support the development of a sovereign supply chain?
Which I think's the idea.
[00:20:57] Speaker C: Yeah, there are Some projects directly, let's say in the same package as this project under the Horizon umbrella. These there are three projects called the sigsters. So you have the SIG for grid project, but we have a couple of others advanced which develops high voltage sigmosfets for MVDC applications. I think this project is a little more say down to the component itself, a little not so holistic if I may say, as the SIGVA grid. So it's more detailed on the. On the component level. And then there is another project, to be honest, I don't know how to say that, but It's a forensic or 4:2N6 project which aims at developing even higher voltage silicon carbide devices up to 15 kilovolt, both silicon carbide MOSFETs, but also IGBTs.
And definitely these two projects are complementing each other because the SIGFIB grid is more on it. It's basically covering the whole chain. But of course in each topic let's say we cannot go as deep. It's very much on. On the grid integration, on the distribution, taking that as a starting point and then from that look down the chain and see how the. The silicon carbide switches is part of that. I think the other two projects are more detailed on the switches themselves and I think that's complementing other.
There are of course other projects around in Europe which are not necessarily funded by the Horizon, but other funded by other founders. There are some in Aalborg where I happen to be quite often in Denmark and there has been several. We have also joined in on project which ended last year Mvault which was also revolving around medium voltage silicon carbide. We were focusing on on more detailed on the power converter itself in a wind turbine application. So I think there are several of these projects around.
[00:22:53] Speaker B: Okay, good. Thank you very much Bjorn for talking about this important project which will really support the integration of more renewables into Europe's grid.
So thank you very much again.
[00:23:05] Speaker C: Yep, you're welcome.
[00:23:10] Speaker A: You've been listening to the EU Energy Projects podcast, a podcast brought to you by Enlit and friends. You can find us on Spotify, Apple and the Enlit World website.
Just hit subscribe and you can access our other episodes too. I'm Aretita Radimo, thank you for joining us.
