Edward Byram and Talbot Chubb

DeVorkin:

We're now in Mr. Byram's laboratory, room 219A, in Building 209 at the [United States] Naval Research Laboratory. This is a continuation of the videohistory interviews we are doing to explore early years of X-ray astronomy at Naval Research Laboratory. The date is July 8, 1987, and I'd like now to turn to Mr. E[dward] T[aylor] Byram, and ask you, what is this large vacuum chamber? What was it used for when it came to the lab?

Byram:

It was bought initially to test the HEAO [High Energy Astronomy Observatory] counter. We had several jobs to do with it. We had to determine the gas fillings that we would use in the counter. We had two different operating pressures. We had to be able to operate at 2 p.s.i. absolute; we had to operate at 2 pounds per square inch over one atmosphere. So we had to have two different gas fillings, one for each of those conditions. So we used the chamber to determine what those mixtures should be, and we had to also use it to test the rise to altitude conditions. When the rocket takes off, it takes about a minute to go through the atmosphere. During that time, the pressure is continuously changing in the vicinity of the counters, so we installed relief valves to allow the excess gas to get out of the counters. The counters had very thin windows — 1/4 mil mylar — and we didn't want to burst the window on the way up, so the relief valves saved that. But we had to test the relief valves in here, and the pump on this system is accidentally just right. It takes about one minute to go from an atmosphere down to a few millimeters.

DeVorkin:

Which is the range that you want to go to, to 2 p.s.i.

Byram:

Yes.

DeVorkin:

This is quite a large chamber, and you did purchase this from a standard company or provider. Was this one of the first chambers that was purchased by you for calibrating and testing X-ray detectors?

Byram:

We have had — I never did any of the purchasing, but we did have some other smaller chambers that would accept a counter this size.

DeVorkin:

But since the HEAO counter is in another building — and we'll be looking at that in a later interview — we're going to be using this counter now to demonstrate how this vacuum chamber works. Okay. So I think what we'll do now is open up the door and start the process.

Byram:

Well, this is going to be a rather crude demonstration, because these inlet and outlet ports would have to be plumbed into the outside world, and high voltage and signal leads would have to be fed through these ports that the chamber has. But we would set the counter in here in such a fashion. It isn't going to set on that. But it would have to be set in there so we could observe it.

DeVorkin:

Okay. Now our next step would be to go to the filling station and begin the pump-down procedure?

Byram:

We would start the pump-down.

DeVorkin:

Okay. Fine.

DeVorkin:

Could you describe what that detector is representative of, when you flew detectors like that, and how it's put together?

Byram:

These detectors were flown, this particular model, in the late seventies — I mean, the early seventies and the late sixties. The honeycomb supports the very thin window. I don't know whether —

DeVorkin:

So the honeycomb is not a collimator, but it supports the window underneath?

Byram:

Its main purpose is just to support the window. The collimator was a much finer thing that now is over in TID [Technical Information Division]. The honeycomb itself is not terribly strong mechanically, and these bars are required to support the honeycomb.

DeVorkin:

And the resistance of the mylar to bursting, that's a test that you would be doing in this kind of a chamber.

Byram:

Yes.

DeVorkin:

That's right.

Byram:

But this counter is different from the HEAO counter, in that this counter always operated at over an atmosphere.

DeVorkin:

That was for a rocket sonde experiment.

Byram:

Yes.

DeVorkin:

Okay. You can take that out.

Byram:

All right. No, I don't really want to leave it in there.

DeVorkin:

Okay. So what we would do, then, is now reclose this. You're better at that than I am.

Byram:

The bottom one works better than the top.

DeVorkin:

Okay. Now we want to reset up in there.

DeVorkin:

Mr. Byram has now placed the X-ray detector into the vacuum chamber, and the next step will be to turn on the vacuum pump, pump down the chamber, and then start the testing process. So Mr. Byram, would you like to start that process?

Byram:

Okay. I think everything's all set. [Pause] Once the pressure gets down to 50 millimeters, then the mechanical gauge will start to indicate.

DeVorkin:

How long does this pump-down usually take?

Byram:

About a minute.

DeVorkin:

About a minute.

Byram:

It's gone a little over a minute now.

DeVorkin:

I can see the gauge has taken over now.

Byram:

Yes. That's low enough. We can switch to the other pump now.

DeVorkin:

Okay. What is this second pump? Is it a maintenance pump?

Byram:

It'll go to a little higher vacuum than the big pump, and it will maintain the vacuum. It won't handle the huge volume that we started with, but at this low pressure, it'll work quite well.

DeVorkin:

Okay. Now, normally, what else goes into the vacuum chamber along with the detector?

Byram:

A radioactive source, usually, to provide radiation, so that we can tell the counter is working. That's one of the tests that we make during this test, is to check for the energy resolution of the counter. The full width at half maximum should be about 20 percent of the full deflection. These counters are proportional counters. They're called proportional counters because the pulse amplitude is proportional to the energy that is detected by the counter. The range of sensitivity of the detector is like a few hundred electron volts up to 20 or 30 kilovolts.

DeVorkin:

What is the filling station used for in relation to this test?

Byram:

Well, initially, we had to determine what that filling should be, and we could put on a tank of xenon and a tank of carbon dioxide, and try various mixtures, and we could also try various operating pressures. We had to match the operating pressure [with] the high voltage to get pulse amplitudes that were within the range of our amplifiers and telemetries.

DeVorkin:

Could you show us the various parts of the filling station and identify them?

Byram:

Well, this small pipe here is the manifold, and we can connect up to six different tanks of gas or measuring devices. We have gauges up here to determine, crudely, what the pressures are. We have a high-vacuum gauge here to make sure that we have everything pumped out adequately before we start. Some of these mixtures are very sensitive to contaminations. For instance, water vapor is very bad. You really need to make sure the inside of the counter is clean.

DeVorkin:

How long did it take for the HEAO counter to out-gas, once you pumped it down?

Byram:

Well, the first time, it would take maybe an hour or two, but after that, once the counter's been cleaned up, it could all be done in 15 or 20 minutes.

DeVorkin:

Did you design this filling station on the side here?

Byram:

Yes. It's a very simple thing; there's not much to it — a manifold, six valves, copper tubing fittings. And we could couple those into an array of gases. Usually we were using xenon and methane for the low pressure operating condition. For the higher operating condition, we used a mixture of neon, helium, and methane.

DeVorkin:

How did you arrive at that mixture? Did you try different —

Byram:

I cut and tried. Just cut and try. Try a little bit of this, a little bit of that, and we wound up with a gas that we could use at atmospheric pressure, with the counter out in the room if we wanted to give a rough check. That mixture did not have good resolution, but it did simulate the operation of the detector so that we could test out the telemetry in our pulse amplifiers and the discriminators and the anticoincidence circuitry.

DeVorkin:

Did you put all six of the detector modules in here one at a time?

Byram:

No, we didn't. We had a prototype counter, and we did almost all of our testing on that one prototype. We did test for one week. We had the whole assembly in the large ThermoVac chamber that's over in A59, and there we had a full-up test that ran for a week. In addition to that, we went out to Goddard [Space Flight Center] and had tests there. Later on, we had a year of testing at TRW [Thompson-Ramo-Woolridge], in ThermoVac and — just general behavior.

DeVorkin:

This was an experiment that was largely a NASA [National Aeronautics and Space Administration] project; HEAO was a NASA project. Did they come in and set standards or evaluate your laboratory at all?

Byram:

They set standards for cleanliness, which we had no way of meeting, and we made no attempt to meet them. We felt that the operating conditions were adequate, and that we were not going to endanger our experiment at all. And they accepted it in the end.

DeVorkin:

How long was that process of convincing them that they had to accept it?

Byram:

A few months.

DeVorkin:

A few months. Did you continue testing at that time?

Byram:

Yes. They had no objection to what we did with the prototype, but eventually we had difficulty in the chamber over in A59, which required us to put at least one of those detectors in this chamber to determine where we were having a gas leak at low temperatures. We found the leak in just a day or two, and it was around the insulators that supported the wires at each end of the counter. These counters amounted to a multitude of proportional counters. Each counter was a 2-inch by 2-inch rectangular section and about a meter long. And the top layer had seven or eight of these chambers; the back layers each had six or seven. These various layers were used in anticoincidence to cut down on the amount of cosmic rays that we would see and would interfere with seeing faint sources.

DeVorkin:

During the testing process, what, in addition, were you looking for? I know you're talking about the maintenance of the mylar film, the window. What other characteristics that you haven't mentioned?

Byram:

Well, there was the rise to altitude. We had to dump roughly an atmosphere of gas out of the counter to prevent rupture of the window, and this pumping system simulated the rise of altitude for the Atlas Centaur combination that we were going to ride on.

DeVorkin:

You mean that rate of drop that we saw in the manometer was about the same as the rate of drop you would get with the rocket rising.

Byram:

Yes. And it turned out that we had an adequate scheme for getting rid of that with relief valves. Once we were in orbit, of course, we had to open all our valves, and we didn't try to do anything for a week, while the whole system outgassed. The experiment on the spacecraft had its own gas supply. There were two large tanks, 19 inches in diameter each one, filled to something like 450 pounds of our flight gas mixture, and then each individual module was filled to 400 or 500 pounds at liftoff with our test gas. That was the gas that had to be dumped. But we didn't want to use the expensive xenon mixture, which is like $1,000 a liter, but the test gas was very cheap, but we did have to get rid of it after we were in orbit. We had to outgas the tanks, the small tanks and the module itself

DeVorkin:

This is tape number two of our interview with Mr. E.T. Byram in his laboratory in Building 29 of the Naval Research Laboratory. We've just finished the testing procedure for the detector, and I would like to continue our discussion on the various NASA requirements that they had on you during the HEAO years. What were some of the other requirements that they had, in addition to cleanliness?

Byram:

Well, they required written procedures for everything we did, both here and at Goddard and at TRW. We conformed to their requirements on that respect pretty well.

DeVorkin:

But how detailed were these written procedures? Did they take a lot of your time?

Byram:

No, they didn't take too much time, but you had to include everything that you wanted to do, or you were in trouble. They would yell at you if you did something that wasn't on the procedure. You had to be sure of what you were going to do.

DeVorkin:

HEAO was a very large project, and I'm wondering, especially with NASA requirements for the written procedures and telling in advance everything that you were going to do, did this require more manpower? Did Dr. Friedman hire more people to work in your group during the HEAO years?

Byram:

No. We were really shorthanded the whole time. There was not enough money in the budget to hire more people. We could very easily have used more if we could have hired them.

DeVorkin:

What was the ratio of staff, physicists to technicians, people who would help you?

Byram:

They were almost all technicians. Dr. Chubb and I were the only ones who were not technicians, and Dr. Friedman.

DeVorkin:

How many people, then, were on the staff as technicians during this process?

Byram:

There were a half a dozen people in this building. There was a mechanical engineer, three electronic engineers and technicians, and in the ESD [Engineering Services Division], there were — at times there were a couple of dozen people involved — machinists, designers, draftsmen. They did our thermal design for us and the thermal checkout. That involved two or three people over there.

DeVorkin:

What was your relationship with the technicians when it came to designing various parts of the detector? Did you sit down together and map out your requirements, and they would come up with the design? Or did you give them a rough design?

Byram:

In general, as far as electronics go, I would tell them what I needed, and they would provide it. I did not have to give them any — the people in our branch were very good. They did not need much supervision. They knew what they had to do, and they knew how to do it.

DeVorkin:

Were things usually communicated, your needs communicated, in writing or in meetings, one-on-one conversations?

Byram:

Well, they did have to meet NASA's requirements — all the designs had to be in great detail and on paper.

DeVorkin:

Before the HEAO, when you were working on rocket sonde [and] smaller projects here at the Lab, did these procedures have to be written down?

Byram:

Oh, no. We did everything — the mechanical parts had to be drawn in detail. We did have to have good, precise mechanical devices that could be assembled and disassembled, and that was done in great detail. But the directions from me to the people who did these detailed designs was very simple — just word of mouth. Nothing had to be written, no memos, no sketches even.

DeVorkin:

Going back to the testing procedure, the one that you just took us through, were you primarily responsible for this, or did you design the various steps of the procedure that had to be done, and then did your technicians carry it through?

Byram:

No, I did all of it. I wrote the procedures and I performed them. Once in a while, I would get help to move a detector in and out of the chamber, but that was it. I maintained the chamber, I hunted around, got the test equipment together, assembled it.

DeVorkin:

When you first purchased this, or when it was purchased by others for the project, did it meet your requirements, or did you have to do any extensive modifications or tuning of the vacuum chamber? Did it reach the level of vacuum you wanted and then hold it easily?

Byram:

Not easily, but it did in the end. It was delivered in exactly the condition that we ordered, but it wasn't really complete. We had to add these fittings on the pipe. They just had pipes coming through the wall. We had to install all these fittings and leak-check them, and we had to supply the pumps, connect all that sort of thing. We had to build the manifold, we had to provide connectors to feed cables in and out, and all that required a lot of preparation, a lot of leak checking, but eventually it all went together.

DeVorkin:

Were these techniques that you had learned in your earlier vacuum test work?

Byram:

Yes.

DeVorkin:

How did you actually learn that procedure in the beginning, back in the fifties? Had you been trained somehow in vacuum technology, or was this something you acquired on the job?

Byram:

No, I acquired it after I got here. I came here as an electrical engineer. I haven't worked a day at it since. I've done everything else.

DeVorkin:

Everything else. How did you go about learning? Did you ask other people or —

Byram:

I watched them.

DeVorkin:

Who did you watch, primarily, when you first came?

Byram:

Oh, there were glass blowers, and the glass blowers in the branch were the vacuum technologists. I learned a great deal from them. The early projects were making and designing and filling and testing Geiger tubes to measure radioactive fallout from the atom bomb tests, primarily. And it was a good place to learn about vacuums and vacuum measuring equipment.

DeVorkin:

Would you say all of the procedures —

Byram:

Among other things, I not only learned vacuum technology, I learned a good deal of physics, a good deal of astronomy, a good deal of high-voltage techniques, electronics, design of amplifiers. The only thing I really came with that was helpful was that I was quite good at de-bugging, troubleshooting, and in the rocket shots, my big job was to make sure everything was working at liftoff.

DeVorkin:

Were there sources for your learning more about astronomy, more about physics, high-voltage technology, again, were there people who were at the lab who already knew something, or were there textbooks, or classes that you took?

Byram:

A lot of it came from field trips, where I mixed with other people. I spent a lot of time at Mt. Wilson Observatory, one whole summer. Another whole fall I spent with the rocket sonde branch at Fort Churchill. There I was on my own. I had to do everything that was connected with our little experiments that we launched there. They were mainly experiments to look at solar radiation.

DeVorkin:

Who did you work with at Mt. Wilson, and what did you do there?

Byram:

We had a series of rockets launched at Point Mugu [California], to look at the sun during solar flares, and I went to Mt. Wilson and used their solar telescope with an H alpha spectroheliograph. In fact, I watched the sun in H alpha from morning until night, watched the sunspots, kept Point Mugu informed about what was going on. When I could see the activity was increasing, I would tell them, "Watch out, there's going to be a flare pretty soon." And pretty soon, there would be a flare. But they only caught one flare that I recall; that was a limb flare that settled one of the big problems that they wanted to solve. They wanted to determine where the X-rays were coming from. We knew there were X-rays connected with solar flares, but nobody was real sure whether it was the sunspots or material ejected during the flare or what.

DeVorkin:

It sounds like you learned these techniques and subject matter pretty much on the job, talking with people, being involved with other people. Is that a fair statement?

Byram:

Yes, that's right. Another thing I learned was spherical geometry from these spinning, precessing rockets.

DeVorkin:

Is that something that Dr. Friedman asked you to learn, or that you assumed was a problem?

Byram:

No, it was — like many of the jobs I've had. It was a job that nobody else wanted to do, so I just picked it up.

DeVorkin:

What about the younger people at the lab? As they were hired and worked with you or worked in this building, did people seek out your advice or work with you to learn the techniques that you knew?

Byram:

No, not really. We were more or less a compact, complete group, and we just stayed together, and we didn't grow, we didn't shrink for many years.

DeVorkin:

Let me ask, what was this vacuum chamber used for after HEAO?

Byram:

It's only been used once, and that was to support a prototype for the OSSE [Oriented Scintellation Spectrograph Experiment] experiment, which is a high-energy satellite that's going to be launched whenever they get around to launching the shuttle. That experiment is complete now, and it's very soon going to be put in mothballs, but they have a prototype device that's going to be further tested here. It was tested several years ago.

DeVorkin:

Is that what this diagram is?

Byram:

This diagram shows what their procedure was. They would cycle it up and down through temperature, and determine the characteristics of their scintillation detectors to determine the temperature coefficients of the detectors themselves, I think was the principal aim. They're going to repeat that procedure in a month or two.

DeVorkin:

What are they going to do with the prototype?

Byram:

They're going to launch it on a balloon, and its main mission will be to look at the supernova in the Large Magellanic Cloud.

DeVorkin:

Are you working at all with the people who are developing OSSE?

Byram:

I've done some work for them. One of my other odd jobs that I've picked up was to do the optical surveying, so I did all their optical alignment.

DeVorkin:

Of the instrument.

Byram:

And I'm going to do it on the balloon, too.

DeVorkin:

This is a balloon to be launched in the southern hemisphere, I would imagine.

Byram:

I'm not sure whether they're going to launch it in Australia or from where they usually do, in southern Texas. The balloon, you know, will go up quite high, and it'll be able to see the Magellanic Cloud from there, from Texas.

DeVorkin:

Did you help them with the vacuum procedures as we went through them today?

Byram:

Yes, I helped. I helped them. I did whatever maintenance was necessary on the vacuum system, maintained it, made sure it was adequate for their purpose.

DeVorkin:

Well, you've carried us through the whole vacuum procedure, and we've talked about the techniques for HEAO and beyond HEAO. Is there anything else you feel we should talk about in this room to help us better understand how your techniques evolved and better understand your contributions to the X-ray group at this time?

Byram:

No, I think we've covered it pretty well.

DeVorkin:

Okay. Thank you very much. Of course, we'll meet again and look at the instruments themselves when we have another session in a number of weeks. So thank you very much.

Byram:

Thank you.

DeVorkin:

This is a videohistory interview with Dr. Talbot [Albert] Chubb at the Naval Research Laboratory. The date is July 8, 1987. The sponsor of the interview is the Alfred P. Sloan Foundation, and this interview is part of the Smithsonian Videohistory Program. We're in room 231A, which is Dr. George [Robert] Carruthers' laboratory in Building 209 at NRL. And I'd like to ask you, Dr. Chubb, what are we seeing in this room?

Chubb:

Well, we are, as you say, in Dr. Carruthers' laboratory. Dr. Carruthers is in charge of the ultraviolet astronomy work that's being carried out at the Naval Research Laboratory. George Carruthers has been in charge of this work for some fifteen years, I guess. He brought to NRL a technology of image converters, or, really, the germ, the idea of a technology, which he has subsequently developed. At that point the ultraviolet astronomy program, which had been previously a star mapping program, became an imaging program. Now, what we see here in this local area is the place where George examines photographic films that have had images recorded on them by his image converters. The rest of this laboratory has largely been used for the development and the testing of such image converters. The bench that's now seen is just a work area, and in the background is some large vacuum equipment in which these convertors are tested against ultraviolet lights to record spectra.

DeVorkin:

Those vacuum chambers are just visible now on the right, is that correct?

Chubb:

Yes. George has grafted his program onto the earlier work on photometers, and still uses ion chambers to calibrate the absolute energy of some of the stellar fluxes or cometary fluxes that are recorded. Here we see gas tanks that are used for filling, as a source of gas for filling the ion chambers. And at this point, we see a filling station itself. This is something we will be talking about subsequently. It's where the specialized photometers used in the program have been made. Here we see an ultraviolet spectrometer, a Seyea-Namioka spectrometer, which is used to measure the spectral response characteristics of ion chambers and photon counters.

DeVorkin:

Why don't we then move into the area where the filling station is situated, and we'll continue our discussion there.

Chubb:

Okay.

DeVorkin:

You had to have a good number of filling stations?

Chubb:

I can't remember exactly. We mostly used one, you know, but there were others, one or two others. I'm not quite sure why that was.

DeVorkin:

There we go. Was this one built by Carruthers?

Chubb:

Oh, yes. Well, he may have had the shop, you know, do the welding. I'm sure that he did. But this is typical Carruthers construction, with the angle irons. Something he brought here from University of Illinois.

DeVorkin:

We're now at the filling station, and I just want to mention that this is Dr. Talbot Chubb, who is one of the long-time members of Herbert Friedman's NRL X-ray astronomy group. And Dr. Chubb is with us today to talk about how his group in the 1950s used filling stations such as the one in Dr. Carruthers' lab to fill and test their photon counters and Geiger tubes. Dr. Chubb, if you would take us through a step-by-step procedure, [a] reconnaissance of some of these old lab techniques, we'd find it very interesting.

Chubb:

Okay. Well, this is the Carruthers version of the filling systems that we used to use, the heart of the filling system being a manifold to which sensors are connected and which provide inputs for gases and vapors. Now, for example, here we have tanks of an argon-methane mixture and nitric oxide, which are two gases used by Dr. Carruthers, and we have organic chemicals, methyl iodide and ethyl acetate, which, again, are vapors which are used in proportional counters and Geiger counters.

DeVorkin:

Why are they in little glass flasks like that?

Chubb:

Okay. These vapors — these liquids have typically vapor pressures between 10 millimeters, 10 torr and 100 torr; in other words, less than an atmosphere of pressure, so that they don't boil if you just put them in normal atmospheric pressure. I didn't mention the essential part of the filling station is a pumping arrangement for exhausting the detectors and the lines which are used for filling. When, for example, one is using something like ethyl acetate, one puts it in some kind of container below a valve, and then one, after the system is well pumped out, one opens the valve and exhausts the air above the liquid, and the liquid will boil at that point, a little bit, and displace the air, and remove all the air. You want to be very careful in both Geiger counters and proportional counters to have no air in them, because that doesn't work very well as the counter gas.

DeVorkin:

What is the larger container below?

Chubb:

As I say, this is a Carruthers filling system. He uses a somewhat different technique than we used in the old days for filling gases. He prepares a pre-mix of gas. What he does is pump down this container and then introduce a certain proportion of vapor into the container of vapor A, say, closes the valve, pumps out, closes this system off and pumps out the line, then introduces a larger component, say, gas B, [and] opens this valve. He wants 10 percent A and 90 percent B, he adjusts [by] reading the pressure until he gets that proportion, closes the valve, and then pumps out the system. Then he closes off the pumps again, and then fills the detectors from this container. That is not the system that we used primarily in the old days. In the old days we had individual valves on each of the lines that ran to the detectors. We pumped them out, introduced vapor A, closed off the valves to each of the detectors, closed off the vapor, the source of vapor, pumped out the line, then introduced the main gas to the desired pressure, which might be 600 millimeters of mercury, 600 torr. Then we'd open the valves just briefly, let the gas rush in, displacing the vapor, then we'd close the valves off, and the mixing occurred in the tubes themselves, and that way we got all the tubes to behave electrically the same. We got a pretty good uniform mixture. So that's a little bit different, a different technique.

DeVorkin:

Do you know why he changed it to this design?

Chubb:

Not really. No, I'm not quite sure. But this is a very good way of doing it. You pre-mix your gases.

DeVorkin:

I understand there are larger differences as well in the construction of this filling station that Dr. Carruthers made, and the ones that you used back in Building 30, where you had a number of these stations. What are the major differences?

Chubb:

Okay. In the early fifties, the filling stations were a much more important part of our business. All our sensors really were specially tailored for our purposes. We manufactured these very specialized detectors which we flew in the rockets in those days. And all the old systems were glass. Instead of a stainless steel manifold, we had a glass manifold. All the valves were stop cocks, just glass stop cocks which were used. The diffusion pump was a glass diffusion pump with little pots in which vacuum pump oil boiled, rather than the metal pots. So those were two of the major differences, I mean, some of the major differences. We had a different support structure. We had clamps that held the glass lines. But the main technology, tubes were put on with glass, generally, rather than with copper tubing. For example, this is a corona regulator which would have been used in, I don't know, about 1960. You can see there's a glass tubulation on this, just like in many vacuum tubes. This glass tubulation, that's soft glass, which seals to a chrome iron, but we worked with the glass blower. The glass blower would very gently, with a little special not-too-hot flame of natural gas, would seal this little tubulation to a glass tube. So again, that was the connection that was made. Then when the tube was filled, then he would just heat the glass tube, and the tube was always filled with a little bit less than one atmosphere of pressure, so that the vacuum in the tube would then pull in the glass and make a vacuum-tight seal. Then we would apply these protective covers, and that would make a tube that could be handled without fear of breakage.

DeVorkin:

Can we get a close-up shot, do you think, of this tubulation?

Chubb:

Sure.

DeVorkin:

So this is the part that would fold in.

Chubb:

Yes. That would start off as a long glass tube. We would buy these shells from a tube fabricator, and that tube, then, would be joined by glass blowing techniques to another tube of the same material.

DeVorkin:

Why did you work in glass, everything in glass, and Dr. Carruthers in stainless steel?

Chubb:

Well, I think — well, for several — one reason, you did not have available, in those days, the good stainless steel fittings and valves — high-vacuum valves — that are now available. Valves like that, for example, or high pumping-speed valves like that. Whereas you did, at that time, have good stop cocks carefully ground, using high-quality vacuum grease to make a complete seal against the outside air. So that was just the technology that was available at the time. Also we had glass blowers. The man I worked with, whose name was Robert Moore, he wasn't the first glass blower to work in the group. And there were many specialized tubes that were made by glass blowing techniques. We weren't just dependent on commercially manufactured tubes. As a matter of fact, even some of these specialized tubes that came in use in the seventies, these have a silver stamping, silver chloride seal to a crystal window. This technology was pretty much perfected by our glass shop. At that time we had at the Naval Research Laboratory a very substantial glass shop in a different building that was capable of manufacturing special detectors, and this is a very satisfactory high-temperature seal, the silver chloride seal. There's another type of a seal that the glass shop could make. You notice that this is not a ceramic construction; it's glass, but for this outer insulation. —

DeVorkin:

This is the back? Okay.

Chubb:

Okay. You have Kovar, glass, Kovar, glass, Kovar. And that kind of a double seal — there's some special name for it — could be manufactured in our own glass shop, so we could make experimental tubes like this in those old days.

DeVorkin:

That's the back end of the tube.

Chubb:

That's the back end of the tube. The front end of the tube is just like the other one I showed. It has a silver chloride stamping, a crystal window, and the silver stamping, silver chloride seal, crystal window, and often the silver chloride is overcoated with a little epoxy or paint of some sort to protect it from damage by light, actually.

DeVorkin:

I see.

Chubb:

So there were different techniques. In those old days, we worked, as I say, with a lot of technician support. Now, I worked very closely with the glass blower in developing the special ultraviolet-sensitive photon counters which we used and the soft X-ray counters. We also had people who were trained [in] the [United States]. Two key electronics people were trained in the [United States] Marines during World War II, and they actually designed and built most of the circuits that we used. Of course, in those old days, we're talking about vacuum tubes. There were no small, lightweight, high-voltage power supplies that we used; in the very first rockets, we used battery packs, where you could buy 300-volt batteries and stack them, actually, to get 900 volts, 1,200 volts, or whatever the detectors required. And we had shop people who became design technicians, who actually designed the structures of our rockets and the housings that held our detectors. And we worked very much as a team, with a great deal of delegation of authority, and without the scientists looking in great detail over every last item that the other people did, because everybody was responsible for his own job and took it very seriously. We all worked very closely, as I say, as a team.

DeVorkin:

Weren't you in a position, however, of having, as well, to determine what the technical people would do, so you had to know something of that they were able to do?

Chubb:

Oh, yes, yes, and we worked with their circuits, too. We set the objectives, you know, of what the next experiment would be and how we would go about doing it. But the detailed design of subsystems — the physicists didn't do all that detailed work themselves; they depended on these very competent other technical people.

DeVorkin:

As an example, trained as a physicist, would you know what was required to built a filling station in practice?

Chubb:

Oh, I think I would. Yes, I would know that, but in the electronics area, I wouldn't know as much about how to design a rate meter or a DC amplifier as the technicians who did that. So as far as building a filling station, no, I probably wouldn't know as well as the glass blower how to maintain the station, change the pump oil, and I wouldn't know the details of what was available in all kinds of graded glass seals, glass to quartz, Pyrex glass to soft glass. I mean, he was a specialist, you know, in that kind of thing. But as far as general design of a pumping system or working out the procedures for filling the counters, and, more specifically, in the choice of treatment of the tubes and the choice of gaseous and vapor fillings which were used, then that was the physicist's — well, my particular responsibility in many cases, well, in basically all the cases.

DeVorkin:

Okay.

DeVorkin:

This is videotape number two of our interview with Dr. Talbot Chubb, July 8, 1987. Dr. Chubb, we have now covered the basic characteristics of the filling station, but I'd like to know what are the various characteristics of the functioning of the tubes that you're testing, that you're looking for. What are you testing, and what do you look for, such as thresholds, plateaus, that sort of thing?

Chubb:

Right. Okay. Our problem in studying, for example, the X-rays from the sun, which was one of the main areas of interest in the early days, there we needed special thin-windowed, X-ray-sensitive counters, Geiger counters. We had to manufacture those with certain specifications, as far as window thickness and material, and fill those and make them operate at the voltages that were provided by the engineers, the technicians and engineers, in the rocket. And we had to ensure that they would operate stably at the proper voltage. Now when I say that, if the tubes in any way leaked and air got in them, then they would either not operate at all, or the operating voltage would rise. Or many tubes had a very narrow operating range, and that was a serious problem. So one of the main things we were looking for in the development of Geiger counters was tubes which would operate over 100 or 150 volts of operating range, and yet otherwise have the high efficiency counting characteristics that we needed.

DeVorkin:

How would you then set up this whole testing station to foster that kind of characteristic?

Chubb:

Well, the main thing was to be able to see how the tube behaved as you changed the voltage on it, and how it responded to a radioactive source. And at each — well, I worked mostly on one of the vacuum systems in room 209 with Bob Moore, Robert Moore, and next to this system, we had an oscilloscope, a little more primitive type than this, and a laboratory high-voltage power supply. And we would run wires with cutaways very much like you see here, which would bring high voltage to the anode or central wire of the Geiger counter, through a resistor, and then we would couple through a capacitor to the oscilloscope. And this would permit us to see the little electrical discharges which occurred in the tube when a radioactive particle excited it.

DeVorkin:

Now, this particular tube that you provided us is a BS-1 tube from the early period, at least an early design. Could you say something about what it's wrapped in and why it's connected in the way it is?

Chubb:

Well, this is just a means of making electrical contact to the tube. The outside shell of the tube is run negative relative to the central wire. The central wire is positive to collect electrons, and the electrons, as they are accelerated towards the central wire, create an avalanche of ionization, which amplifies the electrical signal. And as a matter of fact, in the Geiger tube, the avalanche then spreads along the wire, creating a short-lived corona discharge, which chokes itself off in a properly operating Geiger counter and creates a large enough pulse that was able to be handled by the primitive electronics that was available at that time. As a matter of fact, the very first measurements from rockets with Geiger counters did not even run through rate meters; they ran through DC amplifiers, and the current through the tube was recorded, rather than the actual counting rate. So we had to measure pulse amplitude versus voltage for those tubes. So stability of operating voltage was an important characteristic.

DeVorkin:

Was not?

Chubb:

Was. Was an important characteristic, because it determined the photometric calibration until we converted, two or three years into the rocket program, into rate meters.

DeVorkin:

Would you like to go through the test here?

Chubb:

Okay. All right. Well, this tube, which was probably filled some 20 years or so ago, is what we call a halogen-filled counter. This is a BS-1 type tube. This tube was developed not for the exploration of the upper atmosphere, but for the important problem of measuring radioactivity levels in the neighborhood of the atomic bomb test sites in the Pacific, for example. And it became the heart of the PDR 27 instrument, if I remember my numbers right (laughs), which was a Geiger counter-based instrument with four decades of radioactivity range, which was the mainstay of monitoring the low-level activities that were important for personnel safety. Other instruments were used at the very high ranges, where people really didn't go into very much. Fallout ranges of activity were monitored by these BS-1 counter tubes.

DeVorkin:

So it was a low-level radiation device that, while useful for testing and for reconnaissance of low radiation-level fields here on earth, was also quite well designed for solar work. Is that correct?

Chubb:

The study of the properties of these counter tubes led to another family of tubes, [which] were the ones that were actually applied to the study of radiation of the sun, initially, and then eventually of the stars. Okay?

DeVorkin:

Yes.

Chubb:

Would you like me to see whether this tube will operate? (Laughs)

DeVorkin:

Yes. Absolutely.

Chubb:

Okay.

DeVorkin:

Let's give it a try.

Chubb:

Well, we have here a weak radioactive source, actually somewhat weaker than — I'm adjusting this scope so that it triggers just from the line frequency, and I'm turning up the gain. You can see an occasional pulse. Well, it's now working at 560 volts. I'll go down to 540 volts, and you can see there's just a little pick-up on the trace and no evidences of counts. Now I'm going to turn the voltage up in steps of 10 volts. Now you can see it's beginning to respond occasionally to counter pulses. That's not still proper Geiger counting operation. This is quite a sensitive instrument, so we're not seeing what we would call regular Geiger counter pulses. But now, you see, I'm going to turn down the gain of the instrument, and you can see the pulses are now much larger, and they are more or less uniform in size. I have to stop down 100 volts, and we can see that the pulses grow rapidly. When you first see those uniform pulses, that's what we call Geiger counter threshold, and as we go up in voltage across the tube, those pulses do grow in size.

DeVorkin:

You want to determine the range over which the pulses are measurable, the voltage range over which the voltages are measurable?

Chubb:

Well, once we went to rate meters, all we were concerned with, really, [was] at what range did they give a uniform rate of response, the number of pulses per second. We didn't worry too much about the amplitude, as long as it was above a certain value. But when a counter is over-volted, if too much voltage is applied, then you begin to get double pulses. I'm going to switch to internal, to source trigger, and now it's just triggering on the pulses. And as we turn the voltage up to 680, 700, drop back to 800, this tube has got a very wide range of operating characteristics, as a 20-year-old tube, and it was this long life and stability of this tube that was very important to the operational requirements of the Navy to measure this radioactivity that occurred from fallout on ships in the tests.

DeVorkin:

As you would be testing the tube in the filling station, would you have a similar diagnostic system set up, and you'd come here periodically to run it through the voltages?

Chubb:

Well, no, we didn't usually leave the tubes on the system. We would test them, see that they were operating satisfactorily, weren't going into discharge, which is a phenomena that is more prevalent with non-halogen tubes. No, we would take them off the system, and then we would periodically test them in this kind of form, with a sealed-off tube. And by measuring the threshold voltage, the voltage which you first see those uniform pulses, measuring that as a function of time, we could tell whether the tube was leaking. If the tube had a small degree of porosity and was leaking, then that voltage would gradually drift up a volt a day or 2 volts a week or something like that, or maybe 10 volts a day, you know. It depends on how bad the leak was.

DeVorkin:

Was there a way that you could repair that tube?

Chubb:

Well, the X-ray counters generally were — the windows were sealed on with a baked epoxy resin, an alkyd resin, in those days, but we often didn't really know where the tubes leaked. I mean, you could paint the insulators and hope that you had sealed the leak off, but we didn't really use helium leak detectors in those days to actually locate where the leaks were. As a matter of fact, many of the windows we used — well, that isn't quite true. I would say that generally we would try to seal up a tube and refill it, but if it didn't work the second time or so, we had other shells, and it was better just to fill a new one and hope you got one that was tight. (Laughs)

DeVorkin:

How does this testing procedure —

Chubb:

Let me show you, as we go up in voltage.

DeVorkin:

Sure.

Chubb:

We're near the upper operating voltage of this tube, which is about 800 volts, and as we turn up, these pulses get — that's as small a gain as I can put on it. You can see that you're beginning to get erratic pulses and more pulses sort of following the main pulse. This is called multiple counting. You get strings of pulses. And now the tube has gone into a form of discharge in which one pulse triggers another pulse, and if it's pulsing continuously, I can take the radioactive source away, and it continues to pulse. Now, in a normal organic quenched tube, that would destroy the tube very quickly, but with a halogen tube, you can turn that tube back down a couple of hundred volts, and there it's operating perfectly normally. And it's this long, essentially infinite life which made the halogen tubes very important in the radioactive fallout monitoring kind of a problem.

DeVorkin:

Go ahead.

Chubb:

Well, what I wanted to say may be a little bit about the history of how these were developed.

DeVorkin:

That would be fine.

Chubb:

Okay.

DeVorkin:

Yes.

Chubb:

These tubes, the halogen tubes, were really developed before my era at NRL. They were developed by Dr. Friedman, Herbert Friedman. He originally was involved in X-ray spectroscopy in the laboratory, and was using counters to measure the beams scattered off crystals. And as part of this, his work, he studied the properties of counters, and he used various vapors and chemicals to make counters. The old standby was just alcohol and argon, I believe, in the old days, and he tried other chemicals, including methyl bromide, which made a very long-lived counter — I mean a long plateau, not long-lived counter, but a counter with a long plateau. And he found that methyl bromide and methylene bromide, when those tubes went into discharge, the type that you showed here, which would happen occasionally, would actually get better, and this led him to try this bromine and chlorine and iodine, which are basically the halogens. And so that's how these halogen tubes were discovered. And it was found that you needed only a very small amount of quenching agent in these tubes, and they would operate at unusually low voltage, such as this is 700 volts right here. That was a great asset for portable instruments like the PDR-27. And also the process of sealing the mica windows on the ends of tubes but using powdered glass seals was also part of that development. So it turned out, I believe, that some of the halogen fillings had been discovered independently in Germany, but it was certainly an invention of Friedman, as far as his discovering it in this country. So then, really, the ultraviolet program came out of studying the response of these counters to ultraviolet radiation, and it was found that they had the unusual property that they were dead to near ultraviolet, but were sensitive, somewhat sensitive, to far ultraviolet, and so they could provide a unique type of sensor if we looked at far ultraviolet radiation, radiation below 2000 Angstroms. One of them was flown in V-2 49, along with a beryllium X-ray tube, and these were the first to show both the solar X-rays and the for-certain evidence of far ultraviolet radiation. The rocket went up and tumbled around, and every time the detector looked at the sun, you saw a signal. When you didn't look at the sun, you didn't see a signal. So that was the first for-sure detection of both X-rays and ultraviolet. So that preceded me by about a year. (Laughs)

DeVorkin:

How do you test for plateau characteristics on this setup?

Chubb:

Well, you don't use this setup, really, in a quantitative test for plateau, except you determine when you first get the operating — you don't measure the slope of the plateau on this instrument, but what you do, you can measure the length of the plateau. You can first determine when you get these uniform pulses or pulses of a given small amplitude, and then just turn up the voltage until you see evidences of double pulsing or triple pulsing. You don't want to go into the discharge region, because that generally ruins the tube.

DeVorkin:

Now, we're looking at Geiger counters here, essentially, straight counters. How would the test procedure be different for proportional counters?

Chubb:

Okay. In the early days, the proportional counters were used mainly with mica windows for detection of alpha particles, for example, in the laboratory. And they really weren't used in the X-ray astronomy program, because we did not have the sensitive electronics to take advantage of the energy discrimination of the proportional counters. In fact, they were just not practical to use in those days on rockets. So they're operated with preamplifiers rather than just a simple lead to a scope. You put a preamplifier right on the tube, where you're not loading your pulses capacitively until you get them amplified, and then you typically look at the energy. You use typically a different source for activating a proportional counter if you're studying it. Generally speaking, you would use an iron 55 source, which puts in a monoenergetic X-ray, a 5.9 kilovolt X-ray into the tube. Then you measure what the distribution of amplitudes of the pulses that are produced by that 5.9 kilovolt X-ray, and from the spread in that voltage, that distribution of pulse amplitudes from that spread, you can determine the energy resolution of the proportional counter.

DeVorkin:

This is tape number three with Dr. Talbot Chubb. It's July 8, 1987, in Dr. Carruthers' lab. Dr. Chubb, now that we have seen the testing process for sensitivity and threshold and the basic characteristics of the tube, I want to ask you to describe why you need different quenching gases for the tube and how you go about designing the specific mixtures for each tube.

Chubb:

Okay. Well, let me just very briefly talk about X-ray tubes first. [In] X-ray tubes the sensitivity characteristics, are almost entirely determined by the window and, in part, by the absorbing gas and the thickness of the absorbing gas. But you can calculate the sensitivity of an X-ray tube if you know the composition of the window and the thickness or mass density of the window, and the composition and mass thickness of the absorbing gas. But when you come to using these tubes as we also used them, for studying ultraviolet radiation, then the sensitivity characteristics are determined by other properties of the gas. And in particular, they're determined by the properties of the quench agent, which can be photo-ionized by light. Actually, it's determined by the quench agent and the crystal window, which is used to prevent [permit] the ultraviolet light to get into the gas. For example, you use a crystal window of calcium fluoride, then only radiation longer than 1230 Angstroms passes through that window into the tube and, therefore, can be counted. But if you use the same gas with a lithium fluoride window, then light longward of 1050 Angstroms can get into that tube. So it's a choice — the photo-ionization threshold and the crystal window, which are the main things to determine the characteristics of an ultraviolet photon counter. Now, we used these photon counters in the ultraviolet to look at narrow bands of radiation in the sky. You can build a photon counter using these gas/crystal techniques that will be sensitive for maybe only a range of 80 Angstroms at a pre-selected wave length. The other thing in a photon counter, for ultraviolet light purposes, detection purposes, you're concerned with the sensitivity of the detector. Now, whereas these initial halogen-filled tubes were blind to near ultraviolet light and only sensitive to far ultraviolet light, they were very insensitive as a whole. Their quantum efficiency is the order 10-4, 10-5, whereas some of the detectors that we developed later in the program had sensitivities of the order of 50 percent or more. In other words, two photons would come in and produce a single count on the average. So it was really to develop these special band-sensitive detectors that was the purpose of a considerable amount of our work. In the X-ray area, we were concerned with the plateau characteristics, and these were, to some extent, affected by the choice of quench agent. For example, ethyl formate was a good quench agent for general purpose use, and we used it with an inert gas like helium or neon for many years in the X-ray area of work. There are other gases, like butydyene, that would give you a very long plateau initially, but if they ever sparked or went into discharge for a moment, they would never be good again, because they left on the walls of the tube a polymer film, which became a source of electrons that kept the tube in self-excitation.

DeVorkin:

Now that we understand why there are different types of mixtures, the way I understand it, you know already what type of spectral characteristics you want in a particular tube, and you have determined empirically or theoretically — that's what I want to ask you now — the specific mixture that will give the spectral response you want. Is it an empirical process?

Chubb:

Okay. Actually, what we did in the early days, we had a subsidiary program where we measured the ionization yields of gases, and also the absorption spectrum of gases in an ultraviolet spectrograph like this one.

DeVorkin:

So that's what we should turn to next.

Chubb:

It wasn't just done with the tubes. We studied the properties of these gases, and then that studying of properties was done in even greater detail by the [United States] Air Force geophysics laboratory, [Kenichi] Watanabe and [Yoshio] Tanaka, people of that sort, those scientists. And once we had this additional material, we used their results as guides to the development of further types of photon counters.

DeVorkin:

So in your operational program, once you filled a tube with a certain mixture of gases, you would use this vacuum spectrograph to test that it does have the spectral characteristics that you want.

Chubb:

That is correct.

DeVorkin:

Okay.

Chubb:

We actually used a different unit, understand. We had a large Baird monochrometer bigger than this one set up in a separate laboratory, but still, the principle was the same. This is the unit that George Carruthers uses to test and measure his photometers, just as the one that was used to calibrate the intensity of the Lyman alpha glow on Halley's comet.

DeVorkin:

Could you identify the major components of this vacuum spectrograph?

Chubb:

Okay. Well, [in] a vacuum spectrograph, I should say, the most important thing is vacuum, because the radiation that we're interested in does not travel through air, and some of the bands that we're interested in, even less than a millimeter [of pressure] would completely snuff out the radiation. It's amazing how opaque air is at certain wave lengths of the ultraviolet. So we have, to start with, a vacuum pump, which removes the gas from the arms and body of the monochrometer.

DeVorkin:

And is that what we're hearing now?

Chubb:

That's what we're hearing now. Right.

DeVorkin:

Okay.

Chubb:

Then we have a lamp, a discharge lamp Dr. Carruthers used as an RF power supply and an RF lamp. The light from this lamp travels and hits a grating. This is an old beat-up grating of the type that's used in this instrument. This is called a Seyea-Namioka spectrometer design.

DeVorkin:

And this is a half-meter grating?

Chubb:

I would guess so.

DeVorkin:

Something of that order. Okay.

Chubb:

Then the light diffracts off the grating down to another arm, through a slit, and there we have the detector, and we measure the response of the detector as we scan through the spectrum. Now, we scan through the spectrum by just changing the angle of the grating, and you can see this crank turning around. The spectrometer we had in the other building was also mechanized, and it would just rotate the grating. That changes the color of light hitting the detector tube, and we measure its response.

DeVorkin:

So the grating is at the — right in the middle.

Chubb:

Right in the middle where these two arms meet.

DeVorkin:

And that arm would simply rotate the grating back and forth.

Chubb:

Yes. You can see this arm actually moves back and forth, and that actually rotates the grating very slowly back and forth, and the numbers can be read off here. The way we did it before, the instrument went to a recorder, and we read off the wave length scales on the recorder, at the same time as we were measuring the response of the tube.

DeVorkin:

Where does the tube sit again?

Chubb:

The tube sits on this arm here. Actually, this system of Dr. Carruthers is designed for differential pumping, so that there's a slit, and he pumps on the detector side of the slit. In the old days, we really didn't have this differential pumping, so that system wasn't really available, and we didn't use it.

DeVorkin:

Does that have something to do with the strange geometry of those tubes that flatten out?

Chubb:

Yes. This is just to give you a narrow area here and a round area here, with more or less the same cross-sectional area. Yes. In fact, that looks like this has got two stages of differential pumping, or could be run with two stages of differential pumping. That's important when you're using windowless tubes, but our detectors had windows, so we really didn't have that problem. It wasn't a problem with that.

DeVorkin:

Do you know if this particular spectrograph was modified in the NRL shop significantly?

Chubb:

I would say not, really. It's been mounted on angle-arm system, which is definitely NRL, so [it's] possible that — the valving and pumping system looks like an NRL specialty, rather than maybe one that came with the original instrument. But this framework which holds the tank, that's certainly part of the original system, and this focusing arm adjustment is part of the original spectrometer. I believe that this differential pumping unit is made by a different manufacturer than the spectrograph.

DeVorkin:

So this is what he put together out of several standard sources.

Chubb:

Oh, yes, I would say so, and some shop work. I mean, here's clearly a flange that has been — and a housing for a parallel plate. This looks like a parallel-plate ionization chamber housing, where you take the light from the chamber and run it between two parallel plates until you totally absorb the light, and if you use methyl iodide in this parallel-plate ionization chamber, you get almost 100 percent efficiency, so you can measure the number of electrons collected, and you know the number of photons coming out of the spectrometer. That's true, say, at Lyman alpha radiation — 1216 Angstroms.

DeVorkin:

Was this the situation also for the Baird monochrometer that you used? I know Baird is a well-known standard laboratory device. Did you have to modify it significantly in any way?

Chubb:

We just put flanges and special outside components on it. The light source, for example, didn't come with the monochrometer. We used the same kind of parallel plate ionization chamber for absolute photometry. We had photo multipliers with sodium silicilate coatings as a means of going from one wave length to another at more or less constant quantum efficiency, which was used to sort of bootstrap us from an absolute photometric measurement at Lyman alpha radiation to knowing the number of photons, say, coming out of the instrument at 1500 Angstroms. Sodium silicilate has a constant quantum efficiency over a wide range of wave lengths in the ultraviolet.

DeVorkin:

How much time would a specific test take, assuming this was the Baird monochrometer, and you had your tube all set up? Did it have an automatic scan rate as well?

Chubb:

Yes, it had an automatic scan rate. I believe we actually, in the first days, read a dial and wrote the wave length on the individual piece, but I'm not quite certain about that. It was a constant scan rate so you could interpolate between those numbers. The spectrum that we use is not a continuous light source; we used a hydrogen light source, and it has bands of molecular hydrogen, so you get these peaks of emission, multiple peaks of emission. And, of course, after you run for a while, you learn which peaks are at what wave length, and you pretty well learn your way around the spectrum with a hydrogen light source. (Laughs)

DeVorkin:

When you started using this system or the Baird system — I assume that was 1949-1950 — as you were developing the program with Dr. Friedman, did you find that you had to learn new techniques, say, with operating a filling station or making one?

Chubb:

Well, of course, it was all new to me, because I was right out of graduate school. I had never even run recorders before, you know, let alone vacuum spectrometers. So it was all new to me, and I think it was maybe a couple of years before I made very effective contributions to the group. You have to learn. Definitely a learning curve in any kind of this work.

DeVorkin:

And what were the basic techniques, as you recall them, that you had to learn? What kind of familiarity did you have to get from what we've seen here?

Chubb:

Well, firstly, you just had to learn how to operate the lamps, run the vacuum system, run the recorders. Those old recorders had servo mechanisms that made the thing track the voltage signal. They were basically Brown recorders, very good recorders. We had to learn what photo-ionization was. (Laughs) And [we] learned what quantum yield was. We had to learn how to hook up the counters, first to amplifiers, but, rather quickly, to rate meters, and operate the rate meters. I don't know. It's just bits and pieces of laboratory technology that you pick up pretty quickly, actually. I mean, it wasn't long before we were really looking at the gases and the band structure of ammonia, for example, as it absorbed some of the radiation between 1800 and 2000 Angstroms. We had to learn how to measure the efficiency of a counter. For example, you can measure — you can expose a counter to an X-ray counter or — well, you had to learn how to measure the charge per pulse of the counter. If you take a counter and excite it with a radioactive source, you can measure the number of counts per second, and you can also measure the currents per second if it's a very low current. So you have to learn how to run electrometer amplifiers, and if you know the charge per pulse, you're also at the same time measuring the gain of the counter, because the counter starts off with maybe — if it's a photon counter in the UV, it starts off with a single electron, and you may end up with 10,000 electrons if it's a gas-gain counter, or a proportional counter, in other words, a type of a proportional counter, or if it's a Geiger counter, you may end up with 107 coulomb charges, electron charges, in that single pulse. So I mean, there are lots of little problems like that that you encounter, and you have to learn. I mean, we had to learn to calibrate these detectors for near ultraviolet radiation. There we used mercury lamps that were, in turn, calibrated against thermocouples, because we had a lot more light in the mercury lamp region, a lot more energy to work with than you could work with heat detectors — thermocouples, basically. We had to learn techniques of how to attenuate a signal, because the light level from the mercury lamp might be orders of magnitude too large. Well, with [a] mercury lamp, which is light which goes through air, there's no problem with putting things [at] a large distance and using the inverse square law to reduce the intensity. But if you're working further in the ultraviolet, where air absorption is a problem, then you need to use grids or some opaque material of a finite transmission. You have to learn to measure the density of the grid. How much does it transmit? Is it a 10 percent grid, a 1 percent grid, or a 3 percent grid? I don't know. There are all kinds of things that go into any kind of laboratory work, and you learn those that you need to do your job. (Laughs)

DeVorkin:

Well, thank you very much. This has been very marvelous, getting to know these pieces of equipment.

Chubb:

I enjoyed doing it. These little ultraviolet detectors, I might say, were among the most sensitive energy detectors that have ever been developed. The little ultraviolet photon counters could detect about 2 x 10-19 watts per square centimeter, that order of magnitude. I don't think there are any detectors really available today that are any better than those were.

DeVorkin:

Those are the ones that started it all.

Chubb:

Thank you.

DeVorkin:

Thank you.