About science; About astroelectronics

This is about science produced by the California Institute of Technology and originally broadcast by station KPCC in Pasadena California. The programs are made available to this day sion by national educational radio. This program is about Astro electronics with host Dr. Albert Hibbs and his guest Dr. Edward Dennison. Here now is Dr. Hitz. The largest telescope in the world is the 200 inch telescope located on Palomar Mountain in Southern California. The 200 inches is the diameter of the main mirror in this instrument. And the reason it's so large is to detect the light from very faint and very distant objects in fact that has for many years been the object of a telescope making bigger and bigger instruments to see fainter and fainter objects. But now there's a new approach to this idea of measuring faint objects or even getting a better look at the ones we can see closer by the use of electronics and new instrumentation. One person who's been working on that most heavily at the

Palomar Mountain observatory is Dr. Edwin Dennison our guest tonight who is an astronomer of some years experience and training he studied at Swarthmore in the University of Michigan also in Holland. He spent nine years at the observatory at Sacramento peak and the last three years at the Mt. Wilson and problem are observatories. Ed how did it begin to look at the instrumentation approach rather than just making a bigger telescope as a something that has just opened up recently or has it been an idea it's kicking around for some time. Well I think it was something we got into. We first tried to solve small problems at Sacramento and when I came on here we decided to try and extend the capability of these instruments. What was the what was a sort of a small problem that Sacramento would fall under this. Well the very simplest and very first

was making a market town area that would work accurately for measuring densities and photographic films and plates when the object you're looking at are extremely small. Yes and also when you simply have an exposure whether it's large or small how do you make a quantitative measurement of what is on the plate and more than just a qualitative judgment of what is there. So just an instrument to measure light transmitted through a plate was a right and this was not new as an instrument. But we brought some new thoughts and new ideas into the electronic circuits for Mike towers. In fact when I left Sacramento peak we were hoping the micro tower directly to the computer. So you didn't have to have a graduate student there writing down little readings in his notebook. That's that's exactly right. Well what happened how is this progress then in the subsequent years from the micro for Comet or type of

operation. Well I think we're working. Always towards the goal of being able to make our observations at the telescope with electronic instruments. Electronic devices data collection systems and get the results of these observations directly into the computers and then onto the astronomers desk for further reduction. With essentially no intervention of a human being as to say automated tell us it's the automated telescope approach that's right we want to make sure that at all times the observer the astronomer knows what is happening and has complete control of his data system. But he at the same time must not be part of this data system because that's where systematic errors can begin to creep in. Leave that to the machinery. That's right. Well is this simply a

question of saving work for the astronomers or does it amount to some increase in the capability of the telescope itself. It's a matter of increasing the effectiveness and capability of the instruments that we have. We're not likely to have many new larger instruments and especially not ones of larger diameter in the next few years. So we're trying to make the best use of the telescopes we have now and try and get the most from them and from the clear hours. Exist at the observatory. Well can you actually build instruments that can see dimmer objects than a good photographic plate. Yes with modern pulse counting techniques and photo multipliers actually stars can be measured which are simply beyond the capability of photographic plate to detect.

To some you wouldn't even be able to see you can nevertheless detect with instruments. Yes that's right this may take many hours of integration and of course it's a problem of knowing where they are to find them in this in the sky. One has to take variable stars with it and look at the peak brightness and then as they go on down and go below this threshold. But this doesn't extend the actual light gathering power which of course was the big object in building a huge telescope. Original Yes. Is there any estimate of how much the light gathering power can be extended. Well it was it was I with a photo multiplier type of instrument for example. I think that. I'm more inclined to think of it in terms of the increasing effectiveness of the observer or the astronomer at the telescope. How do you measure that. The amount of astronomy that he is able to do in any one night at the telescope. So he has a program laid out of what he wants to look at and it's going to take him so many nights and.

If you can shorten this program with the instruments but this is this is what you mean. Yes. In fact most of the observers in the Mount Wilson Observatory staff have many more programs and plans that need many more hours of telescope time than are possibly available. So we are able to show that with the data systems we make now they're able to observe say twice as many stars in a night as they were previously able to do. This is simply due to the introduction of the automatic data collection systems. Then we further feel we can go on and increase this with more photo multipliers operating simultaneously or by the use of image tubes or various devices of that kind. An image too was that similar to television to what is an image tube.

Well there are various kinds of image tubes. One of the very sensitive types is indeed a television too but unfortunately it is not of sufficient stability and of the proper characteristics to be good for photometry. The most successful types to date have been image tubes and which one has a photo cathode and a photographic plate so the image that falls on the photographic plate is then. Intensified into an electronic image and then this is recorded on the photographic plate. I see so instead of the light just hitting a plate it hits a photo cathode first releases electrons which then go on and have photographic plate and yes so you see the sensor the same thing but you've gotten some multiplication of effectiveness. That's right and you can take is increasing effectiveness in two ways One is to decrease your exposure time which means that you get more

objects in a shorter period of time and the other way is to make a more accurate fuller metric measurement of your object in the same time as you now require with a photograph. Doesn't that increase also the background noise which would disturb the photo metric measurements and doesn't it. When you get the same amount of general disturbance to photometry with the intensifier that you would without it. No. Because you're concerned with gathering a certain number of information bits during a particular time and storing them and the more bits of information you get the more accurate your measurement. You don't just multiply the noise going from Asia while you multiply the noise and the information but since the information has a slight edge on the noise then you continue to improve this edge. I see I see. In this using this kind of an arrangement.

How much time can you save if you can. If you want to get the same type quality of photograph out of it can you save half your time. The measurements the dative indicated that the best of the image intensifiers of the simple type and I talked about are approximately 5 to 10 times faster than the best of the photographic materials that we've had in the past. That's quite an improvement. And yes this is like tripling the diameter of the telescope. One way of looking at it yes. What about the resolution that you get to is there any smearing added into the image when you put it through the image intensifier. There is and so since the image intensifiers are themselves a fairly small area it means that you're only able to examine a small part of the night sky or a small piece of spectrum and intensify this.

So you still don't gather as much information perhaps per hour. As you would with a photographic plate because you're only looking in a very small part of the spectrum. So you would use this when you had a particular spot a particular source you wanted to look at or a very special region of the spectrum and that's when this would be valuable. That's right to say for programs where you're looking at particular spectral lines for rotation and stars or programs of this kind. Then the image intensifier looks like it's going to be quite successful. I hadn't you had this the first time you've mentioned using it in spectral analysis. Are there other devices also that enhanced the use of a spectrograph in a telescope automatic devices that make it more effective as well as just picture taking operation. Well we have some devices were building where we wish to make a series of

measurements. In the in the spectrum simultaneously as to say you take a single photomultiplier tube and put it at one point in the spectrum then the light or the information from the remaining part of the spectrum is lost. You are simply throwing this away. Now I told her about having to photo multipliers work simultaneously. We're in the process of building a device that will have 33 photo multipliers all working at once simultaneously throughout the entire spectrum and all gathering data into Pulse count counters and then into an automatic. Summary punch. So you get the data automatically from thirty three devices operating at once and covering thirty three spots across the spectrum and can you sweep these across the spectrum so you can overlap the regions that they look in and get end up with the full spectrum out of this essentially so that is to say we can in our first case divide this

spectrum into thirty three broad bands. For a rough look or a general impression of what the stars like or these may be reduced to thirty three narrower bands each of which has a special significance for the star. Or it is possible to sweep the spectrum past the slits and thereby collect all the information from the from the star. Thirty three times as fast as a single photon model that's right that's right. And all of this you say then goes into a punch. What comes out a piece of tape a card a stack of summary punch cards which then can be brought back to Kel-Tec can fit into the large computer there and again literally overnight the data reduced and be ready the next morning for the observers. This must put a lot of graduate students out of work. Well yes although what we do is we put the graduate students to work building

the instruments so they are entirely automated out of jobs. What do you think is the biggest problem you had in this type of instrumentation. Well it's it's very hard to say what the largest single problem has been. I think that one of our biggest struggles is to keep our instruments of high reliability and so that the observer will feel some trust in them and will use them with confidence. I can imagine that this might be difficult. Astronomy is a very traditional subject. I can imagine many observers who don't trust your new fangled gadgets a bit. That's right in fact. Many of the people will look at the instruments that exist now at Mount Wilson which were built something over 50 years ago and they will say now these instruments which were then the state of the art mechanically in optically

have lasted for 50 years. But your electronic instruments that you build today will they be around 50 years from now and working well and I can see why reliability becomes a problem. That's right. If you're up for a 50 year life to have this is pretty rough on a piece of electronic equipment that by today's standards. Where do you find the greatest interest in your astronomer customers I don't know whether that's a proper thing to call them and what do they seem to want you to do the most. I think the the fuels that will cover the photometry automatic data handling. Also they will come to us with new ideas new things they want to try out very quickly and rapidly and of course then we have to scurry around and find all the pieces and parts to put together. And sometimes a instrument to fit their needs. That's right. And then of course is the danger if you put something together quickly that it won't be reliable and yet will be somehow patched up and used for a number of years and this can cause

problems too. Yes and you come back under the. Conviction of many ahead of time that you're not building your instruments to last. You spoke before of having the instruments as automatic as possible so that the observer has doesn't have to worry about them. Suppose he wants to worry about them. What kind of controls do you provide him to operate the instrument to change a setting to do something different. During the course of an observing period how much freedom does a astronomer have to who are around with the black boxes you deliver to him. Well we try to provide him with controls so that he can and outside of the area doing damage to the equipment has complete freedom to control the instrument. Our instruments are always designed so that any one time the observer can suspend the complete operation of the data gathering.

He is a little dubious about the cloud cover or whether he set a slit or a filter wheel is in the wrong position or something just doesn't seem correct. He can simply press a button that suspends or holds the operation. And then after he's made the checks if he feels everything is all right he can go ahead and restart the operation and take on where he left off. Or if he dislikes what he finds he's able to essentially dump the operation stop. He did a collection cycle and then started over again. So that his control then is the same sort of control that he has over the telescope itself. This is the kind of instrument control you're giving for this. That's right we make sure that the astronomer is all was completely dominant over his instruments the instruments must in no way dictate to the observer how he has to make these observations or the kind of observations he as to make. And so when we start the design or

concept of a new instrument go around and talk to all the various people that have used or may use this instrument and put together what seems a collection of very low. Quite a variety of approaches to a problem and some of them seem a little strange but if we can make an instrument that can accommodate all these specialized demands then we finally have a versatile instrument that is generally more capable of handling the future. On the face of it sounds like an almost impossible task. Well do all the various different requirements must be laid on for these things. We don't always succeed. What about the maintenance problems do you find it necessary to have a man on hand throughout the observing period who can make and undertake maintenance work if its necessary. Or have you saw the reliability problem well enough so that that isn't necessary.

We followed maintenance from two points of view. The first is a regular preventive maintenance program in which we have. The instruments examined essentially once a month to make sure that everything is in good operating condition. And then at the beginning of each observer is run since the observers have in general assignments of anywhere from two to three nights on the 200 inch to perhaps as much as a weeks during ascent and 100 injured. WILSON The first night we provide a man who will actually set up the equipment for him and help him check it out so that he so much from the air with it but also has the backup there. If something is wrong at that time of course it's fixed and checked out or some other alternative piece of equipment put in its place. Once the observing run starts we

find that piece of equipment seems to go through the entire run with little or no change. So this has been so far fairly successful that the maintenance problem. One thing that I know astronomers are worried about in Mount Wilson and the mountain instruments is the heat inside the dome particularly in some of the special rooms where spectrographic equipment is located. Does your electronic equipment introduced in these places add a heat load which has makes it difficult which disturbs the optics or causes other problems in the viewing causes turbulence of the air and I think this way it can and we're very much on the guard for anything that will introduce any more heat than necessary. Fortunately nowadays we have a great deal of transistorized and miniaturised equipment to use wherever possible we put our power supplies away from any area where it might get in the path of the telescope.

We have now at Palomar and soon will have close and special data rooms that are insulated and separately ventilated from the main telescope Rossini. The other the heat generating portion of your equipment you can isolate enough to avoid the problem and use solid state equipment in the sensitive areas. Yes and wherever we can we watch very carefully for this in the prime focus and the 200 inch. For example we try to keep the amount of heat we put in as less than the amount of heat that's introduced simply by the man being there. This is the this is the point right up at the top front end of the telescope. That's right rather scary looking spot up there in that little box. It's right it's really a six foot diameter tube. You know but it looks so small that it's a monster that it's a long ways off the floor. Has anybody given your instruments a sort of run for distance and tried to find the most the faintest galaxies

observable with an image intensifier for example. And if so what's the record. I'm afraid I did. I don't have time for they just don't have the information to my tongue what I know that our scanners and pulse counting equipment have been used on many of the quasars by Dr oke. Celtic and I are measuring both their distances and their brightness and so on and so I think the very new and exciting objects that I thought to be the very distant ones have been measured in this way. These have red shifts equivalent to a distance of about three or four billion light years up to numbers like this done by. Yes that's right. Times of light travel which are comparable to the age of the solar system. Yes yes looking perhaps into the origin of the universe and we look at

these things. Do you have any contemplation of what the future of this business is going to be. Do you see new types of instrumentation coming along too that are going to go much further than where you've gone now. Well it's always very difficult to project. The I think that we can now see fairly clearly how we can handle the problems of single channel and dual channel 33 channel multi-channel sort of instruments with a single photo multipliers. We have a pretty fair idea of how we're going to handle some of the simple image intensifier projects. But I think the next step will be the all electronic. Two dimensional for talent or that is to say devices that will work like photographic plates but have the sensitivity and low noise characteristics that we find in our photo multipliers and that

ultimately these systems will somehow be directly connected to our computers. And by that time we help you understand a little bit how to use the computers to reduce and simplify this data and bring it down to what information is needed by the observer. So you would replace a photographic plate with an electronic light gather or light reacting business. And then instead of reproducing a picture for somebody to look at you would actually take a set of measurements across the electronic information that's turned out from this put them into the computer and come out with a table of answers rather than a rather than a picture. That's right. Now the data that's fed into the computer for example could be turned into an image or a picture to be looked at. This is the GPL and there's no reason why. The astronomer should be so far behind but they are so that here again

at the. As far as the entire data collection process is concerned the observer has control and can see exactly can monitor exactly what is happening at all steps in his data collection process. You can do as he wishes but also it is possible for him to get his answers without himself having entered the data chain. So it sounds like the future that you might look forward to is an almost completely automatic system of the from the telescope right out of the answer with everything in between handled. Mechanically Yes too much human interference at all this. When you say the answer is no there the. Be a little too strong when I say answer the data that the astronomer seeks times out. Read that as you go through several intervening steps. That's right because of course the astronomer is trying to get

information about the nature of these objects. And if you can be simply reduced to an answer in which you can have a computer then write a paper and send it off and then it no longer becomes of any excitement or interest. It is just simply giving more capability. I suppose eventually the astronomers will be able to sit in their offices here in Pasadena and punch a computer that will send a signal to Palomar Mountain and the telescope for them to set up the instrumentation they need and as soon as that observing night is over it will have on their desk in the morning the data all properly reduced and ready for analysis. Well I've been accused of such subversive thought. In a way this is this is the direction in which I think we have to work in order to be more effective. On the other hand it is also true that there are many many times in which the

the presence of the astronomers in the dome and at the telescope at night and he will seem to make decisions about how the data should be collected. What you can't program in advance. The intuitive feeling of the scientist with his instrument in his experiment. That's right because he knows the kind of reduction he's going to have he knows the kind of consistency or non consistency of what he's getting. Gets the little little ideas as the data begins to come in and he thinks of new experiments and me. Now once these have been done there are some observations for which you simply are making repetitive observations of things of this kind. When I was sitting there and waiting for data to accumulate. Yes and under these circumstances then there is little value in having the observer directly to do so with any luck the automatic instruments will take the drudgery out of astronomy and leave only the fun behind.

Well one way of putting it that writer put the drudgery in a different area probably. Well Ed thank you very much for talking with us tonight about the new types of instrumentation for astronomy. This was about science with host Dr Albert Higgs and his guest Dr. Edward Dennison of the California Institute of Technology. Join us again for our next program when two more members of the Cal Tech faculty will discuss a subject of interest about science is produced by the California Institute of Technology. That was originally broadcast by station KPCC in Pasadena California. The programs are made available to the station by the National Education already oh this is the national educational radio network.