 you're already familiar with this now that it's day four. We already had some fairly interesting talks about astrophysics, and if you would think you would like to try this yourself, then we have the right talk for you, which is doing astrophysics yourself. Hald Daumann had first contact with CCC in 1984, knows a lot about spectroscopy, a word that I can't say very well, and he is more of an amateur there, so he's the perfect person to tell us how you can do it yourself at home. All right, thanks. So we are talking about decoding stars. So what can we find out about stars? What properties do they have? What kind of mass? What are they composed of? What temperature they have and other things like that? And how do you do this? And the tool here is spectroscopy, and I sat down and built a spectrometer, developed some software for it, and I want to use an example. I want to show that it's not that difficult. So we will start in my garden. So what you can see here, that's my telescope, not self-made, of course. The red thing is the night guidance to balance out the earth rotation, because I have to have the star fixed in my telescope for a long time, and it's a 25 millimeter mineral telescope, and at the lower end you can see a camera, and this camera, when I do the spectroscopy, this camera is replaced by a fiber connector, and you can see that here. So there are two aluminum pipes here, just sticking into each other with some kind of device in there that allows me to take the star's light, which is being focused onto the opening there, and film that with an observer camera, which I've marked on the left side there, and see the light there. And you have to imagine this as... So the fiber has a diameter of about 50 micrometers, I think. That's roughly one of my hairs, and if I were to cut one of my hairs, that would be the diameter, and the star has a focus of 25 mu, I think that's micrometers. So, and those have to be fixed to each other for half an hour, even though the whole earth is moving. They have to stay matched. And this is made possible through this unit. Am I still there? I'm hearing other stuff. And now we first want to look at why are we looking at this particular star. So this is a photo of the Milky Way. Also from our garden with this XLR camera. And in the yellow box up there, I marked the Lyra asterism, and you can see the Vega very clearly, and then close to it is Beta Lyrae. And what's so interesting about it? Well, it's fairly weak for the naked eye. In Hamburg, we can just barely see it with the naked eye, but for 240 years already, we know that every 13 years it changes its brightness. So we have a periodical change in brightness, and this tells us that we have a double star system where one star is orbiting the other, and we are looking at the edge of it, and every now and then one of the stars covers the other, and then there's less brightness overall, and you can see this in the brightness curve fairly well. So each time it dips, the whole system is a bit darker, and if it's further up, then you see both components. And you could say, well, let's just look at that with a large telescope, go to Chile or Hawaii, and look at it, and well, it's still just one star. So what's going on there? And now let's see what we can do from our garden. I have to do a bit of theory for this, but not much. So what is a star? Anyways, a star is a ball of gas, which mainly consists of the components of the cloud, which formed it initially, and this is mainly hydrogen and helium, and a few or a lot of other elements, but those are just there in traces. And the star contracts and then starts fusing hydrogen into helium and producing heat. So the star is a kind of oven, and I've shown this on the left with the halogen lamp, which just has some continuous light across the whole spectrum. But the star is not just the oven or furnace, it also has an atmosphere around it. And so I showed this as a gas flame with a simple soldering flame, went into the kitchen, took some cooking salt, put it in there, and then the flame turns yellow. And this yellow is fairly precise. That's a very exact wavelength or color, or even two of them very close to each other, two wavelengths. Why are they so precise? I have an atomic model here, and it's because quantum mechanically the electrons, the electron energy levels are very precisely defined in the atom. And if through temperature or collisions, I move an atom to a higher level, that only changes by a very specific level. And then when it falls down to base state, it sends out the same wavelength each time at the same place. And this is how you get a very well-known color. And if we look to the lower left, we can see the Doppler effect or Doppler shift. You know that when the ambulance arrives, the sound is high and it passes you, it shifts downwards. And from this difference, you can directly drive the velocity of the ambulance. And it's the same with a star. So if we take this sodium for instance, this sodium for instance, which we assume is in the star atmosphere and the star moves towards us, then the light is blue shifted a bit. So those lines are not where you would expect them in the lab. And if it's moving away from us, then the wavelengths are a bit red shifted. And overall, we are talking about very small, orders of magnitude. So I wrote the numbers down here. So an atom has roughly one angstrom of diameter. So that's 10 to the minus 10 meters and the wavelengths are 100 times more precise than that. And that is almost as precise as I want to be able to measure. And so we're building that kind of device. One more thing. What do we need to build this device? We need, if we see this white light or it looks white, if we look at it with our eye, we have to split this into its color components with this kind of precision. So nature gives us for this the rainbow. And you can see that roughly corresponds to a prism. And on the right, I've just shown an old DVD. I've held it under the ceiling light. And you can see that this grid also splits colors into their components. Yes. And that's what such a spectrometer looks like. You can see that it's not very complicated. It needs a bit of aluminum and it should be a bit robust. And beyond that, we have six parts here that are important for the functioning of the spectrometer. At the lower right, you have the post where you have this gap here that brings the light down. The light that emanates from the fiber doesn't go out through a laser. But just like with a torch, it expands, the array expands. So we need something that collects the light and turns it into a parallel array so that it would hit the grid further down. And the grid corresponds to the DVD that we saw earlier. Here is where the light is actually split into its component frequencies. But they are still not completely separated. They overlap. So we have a prism behind that that separates it further and then gets the light into this camera. This could be a normal mirror-reflex camera, but it's easier to use a black and white camera so that these curves, the mirror-reflex camera comes with it. Don't come into play. The speeds you can measure, if we are to measure stellar Doppler effects, Doppler shifts are about plus or minus two kilometers per second regarding the component that's directed either towards us or away from us. This is how it looks from above. I have left something up that you can see an O-spectrometer here. And there is this silver box with the halogen label on it. Then there is a thorium argon light to calibrate the device. Calibrating means that every pixel in the camera corresponds to an individual wavelength so that I can say when there is light somewhere, this is this particular color or frequency or wavelength. And that we will look at. And we have this kind of scarily looking board here. You have this little weight there. All this is to reduce the noise that comes from a fiber when it doesn't move at certain frequencies. You have intensity, maxima or minima that kind of distorts the picture and degrades the quality. And you can take that out that way. So let's go towards the look at the calibration. These are the direct images that the device takes. We have the halogen picture where you can see where we call these orders as they are. And we call these Echelle orders. Echelle is a French word and means conductor. And we now realize why it means conductor. No, ladder, that is. And we see this ladder kind of pattern. And I've graphed this spectrum next to it so you can see it actually corresponds to colors. You have the red on top. It actually goes down into infrared. And you have the ultraviolet at the bottom. And the left side is always a bit more blue than the right. Now we have the position where the light actually arrives. If I use the thorium argon lamp, which is just like sodium in a way, we've seen that sodium has two lines. Thorium argon that mixed has about thousand lines. And these are all known from the laboratory. So that way with these patterns, I can say precisely where which wavelength ends up. And that gives me a direct way of calibrating the whole device. Now that's about everything that we need to actually image a star. The star is better lara. And at first this looks just like the image from the halogen lamp. But looking closer, we see certain patterns. These dark lines in the upper rectangle. Then the somewhat brighter rectangle. It looks a bit brighter in there. And in the yellow one, we have this kind of double dark line that repeats. And these orders actually overlap. And if you process it, you have to select the right sections of it, clip the image before turning this two-dimensional image of a spectrum into a linear one. I did all this with Excel VBA. The code is a bit long because quite a lot of image processing happens there. And the analysis, again, I do using Excel VBA. It is not really necessary. You can buy finished software, but writing it yourself is fun, of course. And you learn more that way. This image of better lara, these are actually two images copied on top of each other, both with half an hour exposure time. If I now run this processing in Excel, we get this one-dimensional spectrum. This is the continuous line that you can imagine that I arbitrarily put plotted at the one value. And then there are these outliers, the peaks towards the top and the bottom. And that is our star. There's a lot of information behind this. And that's what we need to get out. On top, I added a synthetic color spectrum. Making us see which colors these lines correspond to. On the very right side, you have oxygen and water. These are from the atmosphere of the earth, of course, not from the star. Because if we are on earth, the light arrives at us passing through everything that is in the way, whatever that is. And in any case, the earth's atmosphere is in the way. So you can see that very nicely portrayed here in this image. Then towards the top, towards emission, we have helium and H-alpha and H-better. These are hydrogen lines, very intensive lines, obviously. And then we have iron, the famous sodium from cooking salt and silicon. And the red lines are the areas that we are going to look at more closely. This whole line is about 100,000 pixels long. So in this image, the wealth of information that's in there cannot really be seen. We have to expand certain sections. So I chose these four. And with these, we will run our decoding to see what we can learn from that. These are these four extracts, about six nanometers or 60 angstrom wide. And the arrows mark the laboratory values where you would expect these wavelengths. And if we start from the left, we see iron and silicon two lines. And these are shifted towards blue. To the left, we have blue. And to the right, we have red. So something is moving towards us. If we look at the right side, we have hydrogen on top and helium. These are very wide lines. And they have a typical dip in the middle of them. And these two are not clearly shifted in the same way that the iron and silicon were. But they are not exactly symmetric either. We'll have to take a closer look there. And then we have the cooking salt, the sodium. And again, they are almost where you would expect them, not exactly where you would expect them. And if you look at the widths of these lines, you've noticed that sodium is a very narrow line and silicon is fairly wide. And the helium and hydrogen are extremely wide. What does that mean regarding the speed distribution of the atoms? It means that the atoms that form these lines are distributed fairly wide in terms of speed. Now, I took this image almost exactly two months ago on the 28th of October. And these are one-time images. And you can never quite be sure what you will see in one of those. That's why I've been imaging this star for a longer time over a year. And if we look at it, then we will see. Well, you have to first get used to it. You have these shadow images on top. Below you have exactly the same information. And in the shadow images, just portray it in a different way if you go to the details. And these are the values here are all converted to kilometers per second, giving us a speed distribution. And in these shadow images, starting at the top right, the silicon, I took one of the lines there. We see the period of 13 days that we've known for 240 years by now. But we also see that the silicon has peaks of plus or minus 200 kilometers per second if we put these scales on top of each other. You can see that in the graph below at a bit more detail. Now, looking at the sodium over all these 13 days, it barely moved at all. And that clearly is not connected to our star at all, but it's not connected to the Earth's atmosphere either. So we'll come to that later. And there on the left, we have a helium line. And here you see periodicity over these 13 days. But the peaks are not as marked as they were with the silicon. And of course, we know we have two components here in the system. It is a double star orbiting around each other. And we have one faster component that obviously generates the silicon signal. And then a slower one, which will be the heavier one, because they actually turn around their common barycenter, their center of gravity. And that emanates a lot of helium, it seems. And that emits a lot of helium. And if we look at the speeds here, looking at the graphs a bit more closely, if you analyze them by quantity, you see the speeds of the fast components. That is the larger amplitude. And the flatter curve is the heavier component. The blue items are our measurements. And the red ones for the fast component are the sinus curve, which I fitted to the measurements, assuming that we really are looking right at the edge of the system. In our view, they exactly orbit around each other and that the orbits are exactly circular. And you see that that is not a perfect fit to my measurements. And the reason for that is that the assumption isn't quite correct. So you could now derive the angles at which we're looking and things like that. But we're not going to go into that much detail. What I want to talk about are the speeds. We see a maximum speed of 188 kilometers per second for the fast component. And the slow component does 42 kilometers per second at most. Although the error is larger here. And these two curves are both shifted downwards. The green line is a bit centered at minus 19 kilometers an hour. And the other line is shifted the other way. So that means that the whole system is approaching us at a speed of 19 kilometers per second. If I now know all these speeds and the period of 13 days, then I can calculate what the radius of these two orbits is of both components around each other. The fast component at 41 million kilometers revolves around the center of gravity and the other at 7 million. So they're about 50 million kilometers apart. And we have to understand what that actually means. Is that much or is that not much? We'll come to that. If I now use Kepler's laws of motion, I can actually determine the masses. And the fast component we can see it has three solar masses. Well, that's not very little. It's quite something. And the second component, the heavier one, 13 solar masses. So altogether we have 16 solar masses in which every 13 days complete one revolution around each other. So how do we have to imagine this? This is the final result of the investigation with the equipment I have introduced. That's what it was used. And here's what you can learn from that. Looking towards the right first to the Earth and Sun system, the solar system, you see that the distance of the Earth to the Sun is 150 million kilometers. Comparing that to the 50 million kilometers in the better lyrae system, you see that 16 solar masses are concentrated in a fairly confined space. So there's hell breaking loose there. And if I am on the Earth observing the system, then first I see the hydrogen and the water, as I've already mentioned, because that is in our atmosphere. I also see the Earth revolving around the Sun. And depending on which day I am observing, I would find different speeds because the Earth, of course, moves at 30 kilometers per second itself. So that I have to correct for every time. All the values I gave are actually related to the center of the Sun. And now we also see where the sodium comes from. That comes from the interstellar cloud, which at 21 kilometers per second is moving towards us. It's not connected to the stellar system. And the stellar system, what have we learned about that one? We've seen these large masses. We were able to... From the lines that we saw. I only showed you silicon and sodium, but I have spectral lines of silicon and iron, of course. So I can determine the temperature at the surface. And we realized it's 13,000 degrees. And of course, we know that helium is gathering in the middle and the zone within the star where the fusion occurs is migrating towards the outer layers. And the lighter component behaves like a cannibal towards the faster component as the faster components mass moves to its outer layers. It swallows material and forms it into a disk. And that disk is actually so strong that we don't actually get to see the inner component. And we've seen only hydrogen and helium, but no other elements that would then show us the speed of the second component again. You don't see those because the star is completely covered. That disk is not a narrow disk. It's at least as thick as the star. That is called an accretion disk. And that is the gainer, the star that takes. And we don't know a lot about this star. We only know that it has 13 solar masses. And also the other, we've seen that it has three solar masses. We know the speeds towards and away from us. And I didn't really talk about this. The hydrogen or helium distribution looking at that, the emission lines. You see that no matter what the constellation is, whether the user is at the 0.5 position, where it should cover up the star behind it, or whether it's at the one position. It doesn't make a lot of difference regarding the actual energy output, which means that most of the hydrogen and helium is either above or below. And I would call those jets. It could be different structures, but there is quite a widespread speed distribution there. I didn't talk about this in much detail. It's about up to plus minus 400 kilometers per second, very high speeds involved here. And we know the whole system is moving towards us. It's 19 kilometers per second. We know the period of 13 days. And in fact, we know that every 100 years, the system loses 19 seconds of it. The reason is that the light component loses mass, passing it on to the heavy components. So the mechanical relationships are changing. And the distance from beta iray to the sun is 960 light years, which we cannot see this in my data. I got this data from Gaia. It's relatively fresh. But this system is still observed not only by me as an amateur, so a lot of people look at this, but it's still a subject of current research. And the exact details of how this mass transfer happens and which exact structures or fine structures you can find in this disk, what do the jets look like? So this is still current research. And what I wanted to show is how you can, with relatively simple means, so six components and some aluminum or aluminum, and a telescope you need, of course, but how you can use those to make fairly detailed observations about a star that's a thousand light years away, which exceeds what you can get for the photo from a 10 meter mirror. So they can't observe that, but I in my garden can observe that. So thank you very much. I am done with that. Thank you to everyone who listened and also to the Chaos Computer Club that I've been able to do this here at all, especially thank you to the Jean-Marthe Lübeck and Knut Henke who thought that I could participate here and Ulrich Preitschberger for some design hints for this spectrometer and a lot more people, of course. So that's it for me. Thank you. Thank you very much from us as well. And I think not just chat. And we noticed that this is a great basic talk for people who are interested but who didn't study physics or it's been a while. So thank you very much for explaining so wonderfully what it's about at the beginning and how everything works. We have several questions from the audience or from the chat. And I think there will be some more. So we have one question at the beginning. I have a five inch telescope and a manual spectrometer. Can you do that with this? Yep, you can still do it like that. So the nice thing about spectroscopy is that you can use simple means to get good results already. But you have to combine the things in a clever way, of course. So the spectrometer, I have to connect to my telescope somehow. And I can do this. I did it with this fiber. But I could also, if it's a regular split spectrometer and it doesn't have fiber, then I can also in some way mount it directly on my telescope. And if it's a five inch one, that is a fairly powerful device. And if you don't drive the resolution as much as I did it here, then you can still even do spectroscopy of celestial objects which are even weaker in their light than the star. But I also have a question here if there's more question on the optical components as well as the camera, so the noise and light sensitivity. Yes, of course. There is some information there. Maybe it would go too far here. I can go back to the slide where I listed the elements. So the kilometers of one inch lens, 125 millimeters, then the grating is 97 lines per millimeter, 63 degree grid, which you can buy as such, costs about 200 euros. The prism is an F2 prism, 60 by 60 also costs roughly 200 euros. The camera is an attic which is cooled. So the noise is due to this fairly well, the noise behavior, because I cool it down to 25 or minus 30 degrees below the environmental temperature. And so I didn't go into detail on this, but of course I do some dark field recordings to eliminate the noise. And then the photos we see here, the noise has been removed there. There is some processing in there as already. But it would, as I said, also be possible with an SLR camera with a lens. So the objective, the Samyang here is an F2. You could also use a regular one inch camera, one 35 millimeters. It should have a lot of light sensitivity, but I think with simple means you can still get fairly far. What should you say, how much money do you need for this basic setup? Approximately, I think I did it a bit more precise than approximately. I think there's fairly exactly 1,700 euros without camera, but with the lens. If that's not bad, you can consider that at least. Yeah, if that's up for you. And I would say other hobbies can also be expensive, definitely. Even very expensive. I have one more question about calibrating the spectrometer. How well do you have to calibrate it and how do you do this and how often? All right, yeah, calibrating, we can maybe go into a bit more detail here. I do this with a halogen lamp because these positions can move it. It depends on the temperature and on the humidity. And also on the telescope, I think. So I have to do some kind of calibration. But a basic calibration is not a fundamental calibration. You don't have to do that often, every few months maybe. The fine calibration I do practically every time. So each day with each tournament. And the precision that you can achieve here, I also have on this image, is roughly 400s of an angstrom. So we saw an atom is one angstrom. And the calibration here is, no, I think it's 1,400, not 4, 100. Of an angstrom. Are there publications about this? So they're coming to not on the page because, but in a kind of journal, because it would show that this research can be done with cheap means. Yes, there are, of course, a lot of publications about this. There are also books about this. I myself, so this is all, I didn't come up with all of this myself. I profited from the ideas of others who've done this for a long time already. There are books, let's see one I have here about spectroscopy or which someone I know in the spectrum of these circles wrote of this book. I can hold it up here. It's called Astrophysical Instrumentation and Measurement Technology for Spectroscopy, Theory, Practice, Technology and Observation. Daniel Tsoblowski and Jota, something, I'm sorry. I looked a lot into this book, but there are also other things. There are a lot of publications here, because this tool of spectroscopy is everything we know about our universe, what we do about exoplanets, everything is based on spectroscopy. So this would be my next question if you can see exoplanets like this. Well, see, I think you showed something in this presentation. They also had this pull of the planet on the star. And using this, this is similar to the double star system, or binary system I showed now. And in this system, the velocity was around 200 kilometers per second. And with a planet, even a heavy planet has around 50 meters per second. So I need a lot more resolution for this. And this requires larger telescopes and larger spectrometers. So I couldn't do this with this equipment. So this would require a factor of 100 or more. So yeah, roughly a factor of 100. So would that be possible with your setup? No, no. This high accuracy I will not achieve. I would have to do it all a bit larger. That would be possible. I'm sure there are amateurs who could do this. But in addition, the appropriate stars would need to be found. Because the longer I stretch out the spectrum, the less light remains at each point. So at some point I will need a larger telescope. Doesn't have to be that much larger. But the size I have is a bit too little. And many amateurs have larger devices. Yeah, it would have been my next question, how large the telescope is. I think I mentioned it. It's a 10 inch telescope with a 20 centimeter opening. The focal length I actually shortened a bit, because the star is always a point light source. Even in Chile, with the large telescope, this star is still a single dot. It's not resolved into a disk. So all it takes is all that matters is that I don't lose any light. And therefore, I reduced the focal length to about 1 meter, 30 centimeters. And you built a lot of yourself. What was the hardest thing for self-building? The hardest part to build was the adjustment. You don't see that very well. You see if you observe the lines, they are not actually straight. The whole thing is a bit slanted. And that's not optimal. It would be better if the lines could be made perfectly parallel and not have, and if they would not bend as they do. And that is the most important thing for self-building. And that is something that I haven't quite managed to do yet, and I could readjust things. And that's what takes the longest time as well. But if that is finished and you don't touch it, then there is no longer any problem. What took me the longest time was actually the software. I couldn't count the hours. But that was also because I had to learn, as it is all the time with a good hobby, right? Okay, I have maybe a bit of a joke question. Could you calculate horoscopes with it? No, actually not. But you can calculate all kinds of things. I have this laser, for example, a nanosecond laser for measuring distances. If I point that to a coin, and look at the light that is reflected off the coin, then I can measure the coin and its composition. And nanoscopes, but maybe that's not so important either. But last question, what is your next big project? Well, this is actually just an example. For example, last year I did spectroscopy on Neowise, the comet. It looks very different. You can see carbon bands here. I can't go into the details what that all physically means. But there's a very different chemistry and physics in there, in the comet's coma. I can observe it and analyze it. And also other stars. The advantage that I have as an amateur is that, as I've just shown with Betel Leire, I can concentrate on a single star and keep looking at it. Professionals cannot do this because they don't have the observation time. They have to apply for it in advance, a long time advance, and there is a competition for time. So with this kind of equipment, I can actually, in terms of long-term oppositions, I can actually still gain valuable information, even for the professional. Oh, that's beautiful. I think there is no other field where that is so well possible.