[Wilson, 1985]
Wilson, 1985 (Photo courtesy of NRAO/AUI/NSF)



NATIONAL RADIO ASTRONOMY OBSERVATORY ARCHIVES

Papers of Woodruff T. Sullivan III: Tapes Series

Interview with Robert W. Wilson
At the AAS Meeting in Haverford, Pennsylvania
June 23, 1976
Interview Time: 33 minutes
Transcribed for Sullivan by Bonnie Jacobs

Note: The interview listed below was either transcribed as part of Sullivan's research for his book, Cosmic Noise: A History or Early Radio Astronomy (Cambridge University Press, 2009) or was transcribed in the NRAO Archives by Sierra Smith in 2012-2013. The transcription may have been read and edited for clarity by Sullivan, and may have also been read and edited by the interviewee. Any notes added in the reading/editing process by Sullivan, the interviewee, or others who read the transcript have been included in brackets. If the interview was transcribed for Sullivan, the original typescript of the interview is available in the NRAO Archives. Sullivan's notes about each interview are available on the individual interviewee's Web page. During processing, full names of institutions and people were added in brackets and if especially long the interview was split into parts reflecting the sides of the original audio cassette tapes. We are grateful for the 2011 Herbert C. Pollock Award from Dudley Observatory which funded digitization of the original cassette tapes, and for a 2012 grant from American Institute of Physics, Center for the History of Physics, which funded the work of posting these interviews to the Web.

Sullivan

With Bob Wilson at the AAS [American Astronomical Society] Meeting in Haverford on 23 June ’76. Can you just tell me what your background was as an undergraduate and so forth?

Wilson

I went to Rice - started off to be an electrical engineering student and fairly quickly switched to physics, after I realized what the curriculum would be like. I went to Caltech actually as a physics student. When I started looking around for a thesis project- I can’t remember his name- a fellow at Cambridge.

Sullivan

Was visiting there for a year?

Wilson

Who was visiting there for a year. I live in the Athenaeum at Caltech the first year and he was there and there was a group of us who travelled around a lot. In fact I was looking around for a research project and he was telling me about masers and how masers would make a wonderful amplifier for radio astronomy. I went over and talked to John Bolton and they were all hot on doing something like that.

Sullivan

Did you know about [Charles H.] Townes' thing on the NRL [Naval Research Laboratory] dish at that time, I think it was about ‘58?

Wilson

No, this was- I wasn't very much aware of that.

Sullivan

What year is this that you're talking about?

Wilson

This is '57. I went there in the fall of '57 so we're talking about '57 - '58. I guess they thought they were interested can't in making a maser but the 90 foot dishes were just being completed at the time. And so the real effort was to get some sort of a receiver going on the 90 foot dishes. My first project there was to trying to build a - essentially a Ryle-Vonberg, I believe it is, radiometer, only switching between a cold load and a hot load. A time sharing sort of thing. Because at that time Gordon Stanley thought that diode switches were going to turn out to be essentially perfect devices. Well it turned out that diode switches didn't work nearly as well as we thought and that project got dropped.

Sullivan

These were going to be for the 90 foot...

Wilson

For the 90 foot dishes. I guess one of the 90 foot dishes was made to work first and with first a continuum receiver and then I worked on a local oscillator system which Gordon Stanley had designed. Sort of an abortion - it used GR oscillators locked in a strange way through some crystals. We eventually abandoned a good bit of that. Meanwhile, guess what happened was that my...

Sullivan

You mean there were no scientific results that came out of...

Wilson

Of that particular- in fact there was a very bad thing that happened as a result of my discontinuing that and went to a much more conventional thing with frequency multipliers. And the net result was that [Venkataraman "Rad"] Radhakrishnan later on wanted to - was chasing OH and the new system wouldn't tune up there. He undoubtedly would have found OH if that system had been made to work properly.

Sullivan

Where was it being designed for?

Wilson

1420.

Sullivan

And they couldn't get it to 16?

Wilson

It wouldn't go that far. It had too many fixed cavities. We ended up buying several commercial components.

Sullivan

What did he finally use for an OH receiver?

Wilson

Oh, it was a matter of changing the local oscillator. I don't think I was there by the time OH really was active.

Sullivan

You never did search for OH, you are saying? He only was thinking about it?

Wilson

He did search for OH a couple of times. I don't remember what he used for local oscillator. He must have done it without frequency switching, maybe. This was two oscillators to be pulsed on alternately and come right into lock when they came on for the frequency switch. Oh, I know. We had later done the interferometer but this was before the interferometer was working.

Sullivan

This is all a single dish still?

Wilson

All single dish stuff. Well, that’s jumping ahead a little bit. I guess at one point we had a continuum receiver on one of the antennas and a 21 cm line receiver on the other antenna. That survey of the galactic plane started all with John Bolton said, "Why don't we see what the planes like at 21 cm?" So we started taking cuts across and just sort of feeling along looking at [Gart] Westerhout's data at the same time. It seems to me that John sort of started me out with- we'd go across some of the hot regions but then after a while of doing that with the whole summer ahead and starting going at it very systematically.

Sullivan

And what came out of that survey?

Wilson

We looked at the spectral index of a number of sources using the Westerhout results and Mills Cross things. And then later for my thesis I looked at the more general disk component, derived densities, and the width of the disk.

Sullivan

When was your thesis finally?

Wilson

That was in '62, I guess.

Sullivan

So your thesis was concerned with the galactic plane also, just looking at the general background?

Wilson

Yes. The intention was I was going to do a hydrogen line but I had to do line work for my thesis when that local oscillator system didn't work out I went back to the other thing which I had sort of done in passing.

Sullivan

Was this all single dish?

Wilson

Yes, that was all single dish. Now about the time that was happening, while I was working on that, I guess John sort of decided he was going back to CSIRO and so there was an immediate push to get the interferometer working, which undoubtedly was an excellent thing. Of course everyone had something going on the single dish and didn't want to take it apart. John was quite right to say, "No, you're taking it apart and started making an interferometer."

Sullivan

So there was no radio source worked on until the interferometer got going then? You haven’t mentioned any anyway.

Wilson

Well, as part of - No. There were two things. There was- I guess Dan [Daniel E.] Harris, who preceded me by a year or so, did what's known as the CTA survey. He went through looking at 3C sources to determine spectra.

Sullivan

And that was a single dish thing also?

Wilson

That was a single dish. And then what later became my thesis produced what's known as the CTB list.

Sullivan

Oh, I didn't realize this.

Wilson

Which is a list of continuum sources in the plane. I guess out of that there were two or three - mostly Dan was doing sources that were previously known 3C sources. He wasn't trying to survey the sky. But seems to me CTA 102 and a couple of others came up sort of accidentally. We were doing drift measurements and if you got too busy eating or something and didn't get back to the telescope you might discover something new.

Sullivan

How do you remember those early days at Owens Valley and so forth?

Wilson

Looking back on it, it was fairly spartan. At the time I guess I felt we were spending- the money was readily available. From my present outlook I would say we did a lot of things ourselves. We built the IS strips. The sort of things that people just don't do anymore. It was a lot of fun because of that. John liked to survey so theodolite around and then laying out things John would be there acting as chief surveyor. It was an excellent experience in learning to do all sorts of things because the Observatory was being constructed.

Sullivan

Right, all the basics in astronomy, where north is and so forth.

Wilson

Yes.

Sullivan

Was the scientific program pretty much set by Bolton, the Director? Or was there a large range...

Wilson

A fairly large range. By the time it got - I guess Bolton's main interest was in the source work. Radhakrishnan was quite interested in the hydrogen absorption business. And that was sort of separate from Bolton's work on the sources.

Sullivan

And yet Bolton left just when the radio source work got really interesting. Of course he couldn't have known that.

Wilson

Yes. Well, he was there for the earliest of the work with the interferometer.

Sullivan

He was still there, I see.

Wilson

Yes. [Alan T.] Moffet and Maltby - they certainly had realized that there were double sources, large numbers of double sources from that. He wasn't there long enough to see a - well I guess it was many years before there really was an aperture synthesis.

Sullivan

Well, that's another question which maybe you can shed some light on. It seems to me that Caltech could have done aperture synthesis at a much earlier stage than they really got into the game.

Wilson

Yes.

Sullivan

Do you have any ideas as to why that didn't develop that way?

Wilson

Data handling was fairly primitive at the time. Most of the early interferometry work was done with data coming out on short records. In fact one of the things I helped build at the observatory was a lobe predictor. It was an analog machine. It's ridiculous to think of now but it was a device with a synchronous motor and ball and disk integrators, which I think eventually got connected into the elevation drive or the declination of the antenna so that it would automatically come out and predict the lobe rate. And there was 1 hertz or something added to the slowed down lobe. But at that time- oh, and eventually there were integrators for that so that you actually integrate over some period of time, sine-cosine of the lobes.

Sullivan

But at Cambridge also they must have had similarly primitive methods in the late ‘50s and yet they were doing aperture synthesis to some degree.

Wilson

I remember, let's see 1960 maybe I visited Cambridge as a result of going to a summer school in galactic structure in Holland and I remember at least a paper tape punch there punching data out. Sometime along in there at Caltech there was a paper tape punch bought and Fritz Bartlett spent quite a while with the whole system getting a paper tape to work and trying to get the data into the computer and get some results. But up until I left in '63 I think very little had come out of all that.

Sullivan

A little bit surprising is that a place like Caltech you'd think would be right on top of the latest.

Wilson

Yes, I don't understand it looking back on it. My subsequent experience- well, there weren't many computers at the time and I know that those standalone data collection devices are hard things. I think what happened, what should have been done, I'm sure was to go out and purchase one from Hewlett-Packard or someplace- a company that would put it all together and give you a working device. Instead several different things were bought and put together and the students left to do the job.

Sullivan

It sounds like the modus operandi though was not to buy packaged things.

Wilson

Yes.

Sullivan

That style was set by Bolton I suppose who built things from scratch?

Wilson

And by the amount of money available.

Sullivan

Yeah. This was all ONR [Office of Naval Research] funding was it not?

Wilson

Yes.

Sullivan

So you got your degree in '63?

Wilson

Ah, '62 and I stayed on for another year.

Sullivan

What would you say were the main results of your thesis?

Wilson

Oh, a few more points on the flux of some of those sources. I keep seeing references to them occasionally. I haven't seen much reference to the separation of the thermal and free-free and non-thermal components of the disk, the different widths and so forth. So I think probably the additional measurements on the sources. We went and looked at a number of the sources with the interferometer and about all we could say out of that initial work was that there were fine structure components within them. It was only considerably later that people really looked at them with higher resolution instruments.

Sullivan

What did you do with that extra year you stayed on?

Wilson

Well, by then I'd gotten involved with Rad and the hydrogen line interferometry. And so I mostly - Radhakrishnan - I've forgotten the name of author- the first of the interferometry 21 cm...

Sullivan

Was Barry Clark on that?

Wilson

Yes. Clark, Radhakrishnan and Wilson later became Barry Clark's thesis - sort of an extension of that.

Sullivan

It was primarily Radhakrishnan that was pushing this?

Wilson

Yes.

Sullivan

So after this year you went to Bell Labs, is that right?

Wilson

Yes.

Sullivan

And then what did you do there?

Wilson

At that point the 20 foot horn had been built, used for a while for project Echo at 2.3 GHz and then it had been fitted up with a new travelling wave maser to receive a beacon from Tellstar at 4.08 GHz. And the excitement of Tellstar was sort of dying down. And fairly shortly after I got there they decided to quit monitoring the beacon from Tellstar. So it became available to us as a radio astronomy instrument.

Sullivan

You were not hired as a radio astronomer, I take it?

Wilson

Well, sort of part-time radio astronomy. I think that there was some promise of being able to do radio astronomy but certainly not full-time occupation as a radio astronomer. At that point you could point at any given azimuth and elevation you wanted by velocity controls. There was a Western Electric engineer who was building a coordinate converter and Arno [A. Penzias] had started a helium cooled load absorber at the time I got there. So I proceeded to work on a low-loss switch so that we could switch between the antenna and the absorber using ideas that had mostly been floating around that place. A thing called an ohm coupler which couples a rectangular waveguide into one of the polarizations of a circular waveguide with very little loss. So I took two of those and took the circular waveguide- let me back up, the output of the horn reflector fairly naturally can come out of the circular waveguide. So you have two polarizations in that waveguide. The switch I built took the signal from the cold load put it in one polarization and then we took a piece of circular waveguide and bent it so that it had an elliptical cross section. Therefore it became, well bent at the round amount it became a half wave plate. Half wave difference in the two polarizations. Then when you rotate that it rotates the output polarizations with respect to the input. The net result is a switch which has approximately the loss of equivalent amount of waveguide. It was rather symmetric from one port to the other. At that time I guess our push was to measure the flux density of some sources. I worked with Dave Hogg. They had previously had a helicopter to try and measure the gain of the horn reflector. We made some modifications of original procedure. Hired another helicopter for a day and put a source box on the helicopter and compared the signal from the horn reflector to the signal from a standard horn which [?] measured in the lab. Made two identical horns to measure their gain. From this we - and we had to go through some mess to measure a directional coupler to get any actual difference. So we carried through a fairly accurate calibration of Cas A 4.08 GHz.

Sullivan

So your goal was to make absolute measurements of the flux of the radio sources?

Wilson

Right. Now this helium cooled load that Arno was building was as a reference temperature for determining scale. But it was much too good for that.

Sullivan

That's what I was going to say. It didn't need to be that good.

Wilson

Didn't need to be that good. We had several other- we had a total of four techniques. I don't think I can enumerate them right now but they're in the paper on the flux. That was certainly an over-kill on that part. He started out and said, "The price of helium is coming down, the cryogenics be damned," and built the best termination he could build which had a piece of brass waveguide going straight down into the helium. We ended up putting temperature sensors along the waveguide so we could in fact measure the temperature of the wall as a function of distance and calculate the loss. When the thing had just been filled it had something like 1°K above the helium bath temperature at the flange. And after the helium level went down it gained another 1° or 1.5°. We could calculate it and in the end agreed pretty well. We had assumed - rather the sky let's say - stayed constant.

Sullivan

Well then how did he measure at Cas A?

Wilson

We transferred that to four weaker sources. The next step was going to be to measure the halo component of the galactic radiation.

Sullivan

Which was a great controversy and still is.

Wilson

Yes. Is there a halo? Still don't know whether there is a halo or not. We had available to us a hydrogen line, almost hydrogen line travelling wave maser that was some sort of military system. I don't know what it did but fortunately and unfortunately the guys who made the maser set it up so that it wouldn't quite tune to the hydrogen line, because they didn't want that to be wrecked by this military system. There's one part of the story I left out. That is that during, I guess the second year before I left, Radhakrishnan spent almost a year at Bell Labs getting a pair of travelling wave masers with 90 foot antennas. So there had been...

Sullivan

As phase references or...

Wilson

No, these would be amplifiers. The idea was to swing these things from the focus of the 90 foot antennas. A lot of effort went into the cryogenics and plumbing to make that work. Now the two had to work together as phase stable as part of an interferometer. That was the scheme.

Sullivan

And did that work?

Wilson

In fact I think almost nothing was ever done with it. The problems were too great. The horn reflector having a focal position which only moves in azimuth and which has a big room around it. It was a much easier place to use travelling wave maser.

Sullivan

And so you got these from Caltech then?

Wilson

No. Those had been made pretty much to order for Caltech. Rad went there for a year, learned how to do it and had the extra parts made and had those. But we got some of the military prototypes.

Sullivan

I didn't quite follow the business why there weren't right on the hydrogen line.

Wilson

Because [Stovell, Derick Stovell?] who was head of the travelling wave maser work was aware of the hydrogen line and he didn't want military system to move up into the hydrogen line. It was just his little protection for astronomy to build those things so they wouldn't go there. But that then gave us problems with matching into the thing later. But anyway we were going to get - we had a couple of those - and we were going to make one of those work on the horn reflector. So we decided the first thing we should do is understand what we had at 4 GHz where we thought it couldn't have more than half a degree and probably much less than that from any halo.

Sullivan

This was much higher frequency than anyone had looked for the halo?

Wilson

Yes. I think sort of 400 MHz was the previous effort. And in the course of that we rediscovered that things were too hot. And we spent a good bit of time trying to get the extra noise out of the system. The story about pigeons.

Sullivan

Tell it to me, please.

Wilson

Ok, you don't know that story. There were a pair of pigeons that lived in the horn reflector. The horn reflector was a sugar scoop and for wind reasons it's normally stowed looking down. The end of the horn, there is a cab which was heated, and a pair of pigeons discovered they could fly right in and go right up to the Mylar window and it was sort of a warm place. And they had taken to living in there. Of course, coated the walls with pigeon droppings. We had sort of a hard time getting the pigeons to leave. And we had to clean the thing out.

Sullivan

It made no difference?

Wilson

No.

Sullivan

What other things were...

Wilson

We were worried about the cracks in the thing. We calculated the loss in the waveguide parts of things and worried that maybe the places where the aluminum sheets were riveted together wasn't right. So we got some aluminum tape that had conducting adhesive on the back and taped over those joints. And that didn't make much noticeable difference. Things may have slightly lowered in temperature during all this process but wasn't more than 1°.

Sullivan

And how much excess were you looking at?

Wilson

We were seeing 3.5, that’s what we thought. It was only after - we had planned to check on the backloads of the antenna. One reflector was invented at Crawford Hill. I think probably Harald Friis did it. And it had been used as a radio relay antenna for quite a few years. People had measured several of them on the antenna range and had calculated the backloads and thought they pretty much understood horn reflectors. And here we were getting too much. And one of the things we wanted to check was the response in the back direction.

Sullivan

I gather that no one had checked this accurately though?

Wilson

Yes. So we in fact set up a transmitter and signal generator and horn. I think it was probably the same horn we'd used before. And took that around the top of our hill to several places and tried reorienting horn reflector. And one other complication, the horn reflector has two principal planes of polarization. In one of those planes of polarization the energy is kept well away from the edges where spill-over might occur. And the other plane it's strong right up to the edge. You think of a mode in a waveguide, in the e plane it's cosine shaped and so if that expands out into this horn, the electrical field will be 0 at the edges and you'll get almost no spill-over. And in the other plane it will be high at the edges and the two discontinuities, both the end of the horn and the edge of the reflector, will be illuminated and will cause a diffraction from various parts of the sky. We knew what we thought the difference was supposed to be and we measured it again with our transmitter, about 10 decibels difference in general backload level and that agreed pretty well with the difference of temperature we saw. That gave some considerable confidence that the temperature that we thought was coming in from the backloads was probably about right. That was the order of 1° for the bad plane of polarization and 1/10 or so for the other plane of polarization.

Sullivan

Anything else that was looked into?

Wilson

Oh, I guess we cross-checked much of the receiver. The helium cooled load and the switch- well we built a new throat section for the antenna is one thing we did, just in case there was something wrong with the other one. The other one had come down to an over-sized waveguide and then had a tapered down to the size we were using. So we built another one which came down more directly. That didn't make any difference. That made the loss worse because we had a little more of a smaller waveguide in it. But it all fit together pretty well. I think it was about the time we were making, starting to make the backload measurements that Arno talked...

Sullivan

[Robert H.] Dicke?

Wilson

No, well, he talked to Dicke as a secondary thing. Essentially it got to Dicke but was talking to...

Sullivan

Oh, I think I know what you are referring to. Was it [Bernard F.] Burke?

Wilson

Burke, yes. Bernie Burke. And Burke had seen a preprint from Dicke's group and suggested we should talk to Dicke. And at that point we didn't know they had in mind. So we called up Dicke and their group came over and we showed them our measurements, showed them the antenna, and how we'd made the measurements. They shook their heads and said, "Well I guess you've done it - it looks like that's real." At that time they though, they of course had done a completely separate calculation of the early stages of the big bang on the assumption at that time was a multiple big bang.

Sullivan

In what sense?

Wilson

That the universe was oscillating, an oscillating universe. And they had come up with a number I think of about 10°K for the background temperature and were setting up experiments to measure that. So after some dickering about where it should go we were preparing papers. Our was written - you'll see if you read it very much as an experimental result, they presented the theory that suggested the explanation.

Sullivan

You meant as to what journal it should go to?

Wilson

Yes.

Sullivan

Dicke wanted a physics journal and you wanted an astronomy?

Wilson

Yes. I thought that Astrophysical Journal a little bit more suitable place but [?] the end.

Sullivan

This was '63 or '64?

Wilson

That was '65 I guess.

Sullivan

Well, as I said that is where I’m sort of cutting things off. Do you have any other comments about before that era that you think would be relevant to the history? Thought of that guy's name from Cambridge?

Wilson

No.

Sullivan

Well, thank you very much. Ok, that ends the interview with Bob Wilson on 23 June ’76.


Modified on Tuesday, 05-Feb-2013 15:23:24 EST by Ellen Bouton, Archivist (Questions or feedback)