Tuesday, August 31, 2010

What Was Not to Be: The Apollo 18, 19 and 20 Landing Sites

From Footprints in the Dust, chapter 11, by Colin Burgess, pg. 336:

Apollo 18’s Lunar module was scheduled to land in Schroter’s Valley, the site of intriguing transient lunar phenomena and possibly even volcanic activity. The two-man landing crew of Apollo 19 would then have explored the collapsed lava tubes of Hyginus Rille. The most hazardous but ultimately benficial mission of all would have been Apollo 20. The Lunar Module was to have been upgraded to allow the two man crew a stay of up to six days on the lunar surface, enabling them to visit and excavate the floor of the immense Copernicus Crater.

How (Not) to Launch the First Satellite

“If Project Orbiter had gone ahead as planned, the United States would have placed a satellite in orbit during the summer of 1956.”

From Rockets, Missiles, and Men in Space by Willy Ley, 1968 edition, pp. 304-323:

In the spring of 1954 the Space-Flight Committee of the American Rocket Society had worked out a satellite proposal which had been submitted through various channels. The time was ripe; the government, in particular the Office of Naval Research, had already realized that scientific instruments were needed in the outer fringes of the earth’s atmosphere. Therefore the response was quick. When Frederick C. Durant, president of the International Astronautical Federation, called the Office of Naval Research to tell them that Wernher von Braun would be in Washington late in June, a meeting date was set for June 25. The meeting place was room 1803, T-3 Building of the Office of Naval Research. Present were, in addition to von Braun, Fred Durant, Dr. Fred Singer, Professor Fred L. Whipple of Harvard, Commander George W. Hoover (U. S. Navy and also A.R.S.), David Young of Aerojet-General and, of course, officers of the Office of Naval Research.

At that time there were quite a number of high-altitude research programs, but no satellite program was on the books—yet. The question was whether a satellite, just large enough to be detected, could really be shot into an orbit, say, 200 miles up, within a reasonable length of time. What was a reasonable length of time? About two or three years. “Sooner than that,” said von Braun, and developed an idea which he had probably worked out in his spare time. (I must mention that from about 1950 on the favorite pastime of everybody interested was combining existing rockets on paper to see whether they might reach satellite-carrier velocities.) If you took a Redstone rocket as the first stage, von Braun explained, and then placed successive clusters of Loki6 rockets on top, you should add up enough velocity to get the top Loki into an orbit around the earth.

This scheme had the major advantage of very nearly predictable cost, since it operated with existing rockets. Every one at the meeting then contributed to the discussion along the lines of his own specialty, and a preliminary decision of the following nature was reached: Dr. Singer’s MOUSE, though no doubt a useful device, had to be considered as the second phase of the program. The first phase should be a much simpler and also a lighter satellite.

A visit of Navy representatives to Redstone Arsenal followed, and after everybody up to and including the Chief of Army Ordnance, General Leslie E. Simon, and the Chief of Naval Research had been notified and consulted, Commander Hoover was assigned as project officer of Project Orbiter.

The Redstone rocket, developed by Wernher von Braun and his team at the Redstone Arsenal near Huntsville, Alabama, was the direct successor of the V-2. It was taller—69 feet 6 inches—but its diameter was only 1 inch more than that of the V-2, probably just a coincidence. The take-off weight was 60,000 pounds, the horizontal range around 200 miles, and the payload capacity several tons. While the Redstone was a powerful rocket, it just barely qualified as a satellite-carrying vehicle. To be certain, every advantage that could be found anywhere had to be utilized, including the fact that the earth rotates on its axis.

Fig. 63. Oberth’s “synergy curve,” showing the most economical take-off maneuver possible.

Oberth had written the theoretical prescription many years earlier and had called it the “synergy curve” (see Fig. 63 and Appendix 3). To take off with the least fuel expenditure, Oberth had reasoned, a rocket should take off horizontally and move eastward. By moving eastward it would utilize the velocity which the launching pad has as part of the turning earth. But taking off horizontally would subject the rocket to too much air resistance. The compromise was to take off vertically for 5 or 6 miles until the densest layers of the atmosphere had been left below. Then the rocket was to be tilted in an easterly direction, leaving the atmosphere more or less horizontally. (Note how this procedure agrees with Newton’s conditions.)

Of course a point at the equator moves faster than a point at latitude 45°. To get all possible assistance from the earth’s rotation, the Redstone rocket should take off from the equator. Since Orbiter was a joint Army-Navy project everything worked out well; the Army would provide the rocket and the Navy would provide a suitably equipped ship to carry the rocket to a point at the equator in the Atlantic for a shipboard launch. One V-2 and one Viking had proved that a shipboard launch, in a reasonably calm sea, could be carried out. If Project Orbiter had gone ahead as planned, the United States would have placed a satellite in orbit during the summer of 1956.

But in the meantime other mills had been grinding elsewhere. The A.R.S. proposal had also gone to the National Science Foundation which was involved in the planning for the International Geophysical Year. Scheduled to start on July 1, 1957, the IGY was to run for 18 months, to the end of 1958. Its purpose, in which all the nations of earth, except a few very small ones, participated, was the exploration of our own planet.7 A satellite shot that would gather information obtainable in no other way could, and even should, be a part of the IGY activities.

Plans for the IGY satellite shot began with an examination of Project Orbiter. The Redstone, even by using all the tricks of the trade, could not put into orbit a satellite that could hold all the instruments that were desirable. A heavier satellite—which naturally implied a more powerful rocket—was wanted. The Department of Defense, then headed by Charles E. Wilson, had rejected many well-considered proposals for a more powerful rocket, always insisting that the proponent show “an immediate military need” for such a rocket. The phrase is meaningless; if there is “an immediate need” it is evidently already too late. But it was a fine phrase for clubbing planners that advanced anything more radical than a change in Defense Department bookkeeping procedures.

The U.S. IGY project, being financed by the National Science Foundation, could neatly bypass Secretary Wilson. And if a new rocket were built, it would not even be necessary to ask the armed services for missiles which were still in short supply.

Incipit tragoedia!—as used to be said in old books, “Here begins the tragedy.” It was a tragedy compounded of fallacious reasoning in some quarters, overconfidence in others, callous indifference in still others, and an over-all lack of understanding.

The IGY satellite project was to be known as Project Vanguard, with the Navy in charge. White House Press Secretary James C. Hagerty made the official announcement on July 29, 1955.

Project Orbiter had been pushed into the background and was later shelved.

Let us begin with a description of the Vanguard rocket. It should be said at the outset that the design of the Vanguard was superior to anything else in existence at the time. The Vanguard would use less fuel per pound of satellite weight than any other rocket, it was to have a superior guidance system, and it would incorporate a number of new concepts. In short, it was to be a new rocket—and for an idea of what a new rocket may do, just reread the Viking section in Chapter 9.

The first decision that had to be made was what instruments should be housed in the satellite, a rather difficult decision that had no precedent to go by, since this was the first artificial satellite to be actually built. The only satellite design available, Dr. Singer’s MOUSE, was much too heavy for the Vanguard. The satellite was to be spherical and at first its diameter was announced as an even yard. Later this was reduced to 24 inches, with a weight of 21.5 pounds. Of course since there were to be several shots, the instrumentation in the different satellites might not always be the same, but there should always be a total payload of 22 pounds (what the extra half pound was for is described shortly). Three stages would be needed to accelerate that payload to satellite velocity.

The first stage was based on the design of the Viking rocket and used the same liquid fuels. The main innovation for the first stage, aside from its greater length, was the fact that it was without fins, relying on the gimbaled motor for steering. One other innovation not visible from the outside was the lack of guidance equipment in the first stage—all the guidance equipment for the whole rocket was located in the second stage. The second stage would guide the first stage while the two were still connected, and it would also position the third stage for the final push. The second stage also was a liquid-fuel rocket, but used different fuels. Its oxidizer was nitric acid and the fuel was a substance known to chemists as unsymmetrical dimethyl hydrazine (“Dimazine” became the trade name); these two liquids constituted one of the hypergolic combinations, bursting into flame when touching each other without the need for a special ignition device. The third stage, finally, was to be a solid-fuel rocket.

There was a practical reason for building a three-stage rocket with three different fuel systems. As a rocket using liquid oxygen must be fired soon after the fueling, the fueling procedure might have been quite complicated if all three stages had used liquid oxygen. Since the solid-fuel rocket (third stage) presented no fueling problem and since the fuels of the second stage could stay in the rocket for a considerable length of time without harm, the difficulties, if any, would be confined to the first stage.

Fig. 64. Take-off performance of the Vanguard satellite carrier. Brennschluss of the first stage is at A, that of the second stage at B; the third stage begins firing at C. The dotted line is the trajectory of the second stage after its Brennschluss.

The take-off procedure to be followed was also carefully planned (see Fig. 64). The firing site was to be Cape Canaveral which, being located at latitude 28° 28' N., would contribute 1340 feet per second of rotational velocity. The job assigned to the first stage was vertical lift-off and tilting the whole rocket through an angle of 45 degrees. At the moment of separation the rocket would be 36 miles above sea level; the distance from the firing site, measured horizontally, would be a little less—about 32 miles. After separation the burned-out first stage would continue to climb to about 110 miles and would finally splash into the ocean 230 miles from the firing site.

The second stage was to start burning immediately after separation and to continue the tilt. Simultaneously with the beginning of burning the flight-programming equipment in the second stage would jettison the nose cone that protected the satellite on top of the third stage. At an altitude of 36 miles the satellite would no longer need protection and there was no reason for using up fuel to carry the nose cone any farther. By the time the second stage had used up its fuel supply, the altitude reached would be 140 miles and the direction of motion nearly, but not quite, horizontal. Still connected, the second and the third stage would coast to a peak altitude of between 200 and 300 miles above sea level, the peak altitude being different for different shots. Peak should be reached 10 minutes after lift-off vertically above a point in the ocean 700 miles from the firing site. At the peak, where the direction of motion would be horizontal, the second separation would take place. During the unpowered climb from 140 to 250 miles there would be some loss of velocity which would have to be made up by the third stage. In fact the third stage had to contribute a little more than 2 miles per second. But long before the second stage splashed into the ocean 1400 miles from the firing site, the third stage and the satellite would be in orbit.

Then a final maneuver would take place, involving that extra half pound of the payload, which was a device for separating the satellite from the third stage. It was thought then that if it was connected with the third stage the satellite might not be able to function properly. The separation mechanism was essentially a tensed spring that was “locked” by a small weight. Though the relatively mild accelerations of the two liquid-fuel stages would not dislodge that weight, the high acceleration of the third stage would. Then, after a suitable delay, the expansion of a pressure bellows would release the catch of the spring, and the spring would push the satellite away from the third stage at a rate of only about 3 feet per second. Of course, the third stage, having orbital velocity, would stay in orbit too.

For a while there was a plan for a subsidiary satellite, a plastic balloon covered with aluminum foil which would be carried into orbit folded and inflated after separation. This “subsatellite” was to have the same diameter as the satellite, but while the satellite would weigh 21½ pounds the subsatellite would weigh less than 1 ounce. Both, however, would present the same cross section to any residual air resistance in the orbit. If even a minute amount of resistance were encountered, the subsatellite would be retarded far more because of its lesser mass. The difference in retardation would establish the amount of resistance left. This would have been a very instructive experiment; unfortunately it was never carried out.

This was the design.

Then the muddle started.

Or, more accurately, muddles, for there were several even before the crowning touch of Russian competition materialized.

The first move was a pious announcement to the effect that the American satellite accomplishment would be all the greater because it would be done by means of an “entirely peaceful” rocket that could not be used as a missile. It is said that President Eisenhower was especially pleased with this, but it was as meaningless as the phrase “immediate military need.” “Bon, not only are we going to get an American satellite, but a virtuous American satellite, probably as virtuous as prohibition,” a French magazine jeered.

Seriously, nobody throughout all history had ever cared whether a discovery had been made by civilians or by men in uniform. The fact that the eastern tip of Asia and the western tip of the North American continent are separated by water was discovered in 1728, by a civilian, the Danish explorer Vitus Bering. Later during the same century the Endeavour, commanded by Captain James Cook of the Royal British Navy, made the first circumnavigation of Antarctica, but would anybody have been annoyed if the Bering Sea had been charted by the Endeavour, a military vessel? Or, to cite an actual example: during the last few decades of the nineteenth century many islands and island groups in the Pacific were charted by American destroyers, British cruisers, and German corvettes. Was there ever a sea captain who rejected these charts on the grounds that they had been made with the aid of warships? On the contrary, navigators were happy that the work had been done and willingly paid eighty-five cents, two shillings sixpence, or one mark and forty pfennig for the charts. Or, even closer to home, who had objected to the attempt to explore the upper atmosphere with V-2 rockets?

Of course, Project Vanguard enjoyed tremendous publicity all through the second half of 1955 and through all of 1956. There was so much publicity that the chief of the project, Dr. John P. Hagen, warned publicly that there might be only one success for six attempts. The statement was dismissed by many as “protective official hedging”; actually Dr. Hagen well knew that with so many things which could go wrong the probability of success for a specific shot was small indeed. The final score turned out to be three successes for eleven attempts.

This overwhelming amount of publicity was not caused merely by the fact that this was a new project; there was another reason for it. Both the Army and the Air Force began testing new rockets at Cape Canaveral, and a large crowd of enthusiastic “birdwatchers” with expensive optical equipment began to assemble at a suitable spot along the beach. The White House countered with an angry order that no information at all was to be given to the public, not even the names of new missiles. However, Vanguard had been specifically declared not to be a missile and since it was not a missile it could be talked about. The Navy had stressed the nonmissile nature of Project Vanguard by referring to the Vanguard rocket as a “vehicle,” which resulted in a large volume of mail inquiring who was going to ride in this vehicle, plus a few dozen offers to ride it.

Of course Project Vanguard also was discussed in Russian journals, one of which carried an article by Professor Kyrill Stanyukovitch that contained a significant phrase. After describing Vanguard, Professor Stanyukovitch said that Russian engineers “believe it is possible to build larger satellites than those now being discussed in the Western press.” (Excerpts from this article were printed in Aviation Week, October 31, 1955.)

But during an international conference dealing with the organization of the IGY in Paris in December 1955 the Russian delegates stated that the Soviet Union did not have a satellite project for the IGY.8

Working on Project Vanguard must have been an unbelievably frustrating experience. The demands were so high that the chance of success was marginal indeed. The timetable was too tight. The original appropriation had been too small and additional appropriations were parsimonious and slow in coming. Even the jurisdiction was not as clear as one had a right to expect. And then there was another source of delay about which nobody had the courage to scream aloud in public, because to scream, or even to moan, would have been “un-American.” This was the policy, set by somebody who probably could not be trusted to change a light bulb or to drill a hole through a piece of pinewood, that in “the spirit of democracy” the project must “have the full participation of industry.”

This is a fine rule for production, but it leads to disaster when experimental work is involved. In practice, this is the way it worked. A dozen or two specimens of a scientific device consisting of a small set of instruments and a number of switches were needed. No single component was completely new; on the other hand, no single component could have been bought in precisely the size, or of precisely the function, that was needed. There is a government-owned research laboratory, well equipped and employing skilled mechanics and competent scientists; its staff could have built the necessary devices in three or four weeks and the Vanguard people would have been absolutely sure of their own project. But no, industry had to participate. Instead of having the job done in three weeks, they had to write forty contracts, in quintuplicate, engage in the correspondence that goes with forty contracts, and then check and test the pieces as they came in.

Small wonder that Project Vanguard moved slowly during 1956; the fact that it moved at all was due to the determination of a handful of men who stubbornly did the job they had said they would do. Around the middle of 1956 rumors of Russian long-range missile firings began to be heard. The existence of Russian missiles was officially denied until a courageous senator—I don’t recall just who it was—stated the fact in public and was probably reprimanded behind the scenes for having violated security.

On September 20, 1956, Redstone Arsenal, which had meanwhile become ABMA (Army Ballistic Missile Agency), achieved a great success. A new rocket, known around ABMA as “Missile 27,” had made a record flight over a distance of 3100 miles (Fig. 65).

It was a Jupiter-C rocket and it came to its name in a roundabout way.

Fig. 65. The Jupiter-C shot of September 20, 1956. The two short lines marked x and y are two radii of the earth, forming a geocentric angle of 60 degrees.

Early during that year ABMA had been charged with the development of a missile that was to have a range of about 1500 miles. Its name was to be Jupiter. Of course new components had to be test-flown and the re-entry nose cone for the Jupiter missile created a special problem. Since ABMA had Redstone rockets available a number of them were assigned for testing Jupiter components. But a very minor amount of cheating was needed to have the testing program move fast. As General John B. Medaris, then commanding officer of ABMA, put it, in his book Countdown for Decision:

I must confess here that the nomenclature of these different missiles was—and is—confusing, but there was a reason for it. In testing components for Jupiter, we were constantly using the Redstone missile as a vehicle. But Jupiter had a much higher priority than Redstone on firing dates at the Cape, and in other ways. We therefore decided to label these Redstones used for testing Jupiter components “Jupiter-A.” They were not Jupiters, of course, but the label identified them properly with the program, and gave them the necessary priority.

Similarly, when we put together a composite missile for testing nose cones, we called it “Jupiter-C.” The first true Jupiter missile was not fired until May 1957.

Actually, for this first long-range shot we were readying two identical missiles, no. 27 and no. 39. We wanted to have a spare in case the first shot failed. These two missiles looked—and in fact were—exactly like the satellite carrier that we were to use over a year later,9 except that for this test shot we were using a dummy fourth stage. If we had put a solid propellant into the fourth stage instead of the inert material we were using to get it to the right weight, we could have fired that particular missile into orbit as a satellite. We didn’t do this for the simple reason that we were forbidden to do so. We had no mission for putting up a satellite—Vanguard had that assignment. And nobody had any intention of giving us that mission. So we put sand aboard the fourth stage instead of powder.


Actually, as many readers may have guessed, the Jupiter-C was the same concept as the original Orbiter, except that the Redstone had been made longer and that the new Recruit rockets were more powerful than the old Lokis.

While Missile 27 was an outstanding success, the spirits of the officers and scientists of ABMA were dampened by a strict order from Washington that this shot must not be mentioned. Fortunately for American morale several magazines (among them Life) did learn about it and published the facts. What they did not know at the moment was that ABMA had pointed out that long-range testing of nose cones might result in an artificial satellite and that a strict order saying “no accidental satellites” had been issued in reply.

Just before the news about the flight of Missile 27 became known, the rumors of Russian preparations became more definite. The magazine Aviation Week (October 29, 1956) published excerpts from an article in the Moscow News by Professor Georgi Pokrowsky, which stated that the Russian satellite would have a diameter of 24 inches (actual: 23.8), that it would weigh over a hundred pounds (actual: 184), and that the perigee of the orbit would be at 185 miles and the apogee at 810 miles (actual: 156 and 560.) Of course anybody could assume that Professor Pokrowsky had just published the results of some private calculations, but if one did not make that assumption the article proved that the Russian design had progressed to the point where figures could be mentioned.

In December 1956 the newspapers proclaimed triumphantly that “Vanguard” had made its first flight on December 8 and that this had been a successful flight. Well, yes and no. It had been the first flight in the Vanguard program and it had been successful, but the rocket had been a Viking XIII, used to test instrumentation. Even if the rocket that took off had been a Vanguard rocket, the program would have been behind schedule.

In the meantime the whole picture had been beautifully confused by an order of the Secretary of Defense stating that the Army was responsible for ballistic missiles up to a range of 200 miles, while longer ranges were the responsibility of the Air Force. No matter what the reason for this order, it should have had the result that the Army’s development of the 1500-mile Jupiter missile would be stopped and that the Air Force’s development of the 1500-mile Thor missile would go ahead. But in reality, the Jupiter program was not stopped, it was just slowed down. Nobody has ever been able to understand the true meaning of that order; as far as subsequent developments went it might just as well not have been issued. All it really did was to lower morale by creating a situation of uncertainty. A committee created to decide between Thor and Jupiter found itself in about the same situation as the German committee that had to decide a dozen years earlier between V-1 and V-2. In fact the situation was strikingly similar: the Air Force missile clearly carried the favor of the policy makers, but the Army missile was further advanced. Since the situation was the same, the outcome was the same: development of both missiles would continue.

With the coming of spring in 1957 the pace of events seemed to quicken. On May 1 the second test shot in the Vanguard program was fired; it was Viking XIV, carrying the third stage of Vanguard for testing. The second stage was not yet ready.

On June 2, 1957, The New York Times reported on an article in Pravda by Professor Alexander Nesmenyanov, which contained the statement: “We have created the rockets and all the instruments and equipment necessary to solve the problem of the artificial earth satellite.” On June 10 the Russians reported to IGY headquarters in Uccle (Belgium) that the satellite project was ready and that the satellite would be fired into an orbit as close to the meridian as feasible. This report was distributed in sixty-four countries and carried by most major newspapers in most of these countries, but after Sputnik-I had gone into orbit many prominent people, including even a few Congressmen, asserted that the Russian satellite shot had “come as a complete surprise.” In June the Air Force distributed a memo (not classified) stating that there was every reason to believe that the Russian satellite shot would be made on the hundredth anniversary of the birth of Konstantin E. Ziolkovsky.

Ziolkovsky’s birth was registered as having taken place on September 5, 1857, as stated on page 93 of this book. But that was an “old style” date (Julian calendar) and the Russians, soon after their revolution, had adopted the Gregorian calendar. Consequently they now refer to earlier events in their history as if the Gregorian calendar had been in use in Russia at that time, with the slightly disconcerting result that an event known as the “October Revolution” is commemorated on November 7. As applied to Ziolkovsky’s birthday the change meant that it fell on September 17, so that September 17, 1957, was clearly the intended firing date. But nothing happened.

On August 3, 1957, Reuter’s reported from Russia that Professor Evgenii Fyedorev had been appointed head of the satellite program. He had announced that the firing would take place in the early morning hours (it did) and that the orbital period of the satellite would be 90 minutes (actual: 93 minutes).

There was no news from Project Vanguard but Washington took the official position that the United States was not involved in a satellite race with the Soviet Union. One can only hope that this pronouncement was a considered political lie, for if it was not it was plain stupidity. But at that time Secretary Wilson said in so many words that he did not care if the Russians succeeded with a satellite program ahead of the United States. ABMA, of course, followed the Vanguard program carefully and General Medaris took a preliminary step:

On August 21, therefore, I issued a directive to stop the Jupiter-C test program at once, and to put all Jupiter-C hardware into protected storage so that it would suffer no deterioration. The original program had called for twelve shots. We had used three10 and had nine precious missiles, in various stages of fabrication, to hold for other and more spectacular purposes. After looking the missiles over carefully, and evaluating the state of readiness of each, I advised General [James M.] Gavin, Chief of Research and Development of the Army, that we could hold the two most advanced missiles in such condition that one satellite shot could be attempted on four months’ notice, and a second one a month later.

(Medaris, l. c.)

In September 1957 the Russians broadcast the wavelengths to be used by their satellite to the countries in the Far East and on September 30 the Soviet Embassy distributed the same information to the American press. (But some people, later, still called the satellite shot a surprise.)

The first Russian satellite went into orbit when it was evening in the Western Hemisphere on October 4, 1957. Washington had a deservedly rude awakening. It still tried to stick to the slogan, “we are not in a satellite race,” but public opinion shouted down its own government. American scientists were fairly unanimous in denouncing a policy that had caused such a setback for their country. General Toftoy, who happened to be in Madrid at the moment, simply said what he thought; General Medaris decided that his ABMA would fare best if he kept quiet. He had his special reason: it had been announced that Secretary Wilson was resigning and that Neil McElroy would be his successor. And by a weird coincidence McElroy had been at ABMA when Sputnik-I began orbiting the earth.

Before dinner Wemher von Braun spoke to McElroy, undiplomatically but truthfully, all the pent-up resentments tumbling out: “We knew the Russians were going to do it! Vanguard will never make it. We have the hardware on the shelf. For God’s sake turn us loose and let us do something. We can put up a satellite in sixty days, Mr. McElroy! Just give us a green light and sixty days.”

Medaris interposed: “Ninety days.”

It turned out to be eighty.

In addition to creating general turmoil and unhappiness in the West, the Soviet satellite added a new word to the English language: sputnik. Actually the Russian term was sputniki zemlí, pronounced SPOOT-nik zem-LEE; the Rusian plural is sputniki. Sputnik is a standard Russian noun, composed of s (meaning “with”), put (pronounced poot, meaning “road”), and the suffix nik, which can refer to either an object or a person. The meaning, then, is “travel companion.” Russian astronomers have been using sputnik since the turn of the century as a term for natural satellites, especially if the satellite is small in size. As for the second word zemlí, it is derived from the Russian word for “earth” (zemlyá, as in the name of the island Novaya Zemlyá, “new land”) so that the whole means “companion of earth”; the use of sputnik to mean an artificial satellite originated with Ziolkovsky.

Some time after the launching, the astronomer Fred L. Whipple suggested an international system for designating artificial satellites, modeled after the astronomical system of designating comets. Whipple’s system made the first sputnik 1957 alpha; the next satellite would be 1957 beta if fired in 1957. Since 1957 alpha consisted of several bodies, a finer distinction became necessary. Naturally the orbiting rocket casing, being much larger than the satellite, had a greater visual brightness. The orbiting bodies were labeled according to their brightness as alpha-1 (the rocket), alpha-2 (the satellite) and alpha-3 (the nose cone, which was also in orbit). Orbital decay proceeded as expected, but a little bit faster, indicating that the residual air resistance at altitudes above 100 miles was greater than had been expected from calculations. Since the rocket body was larger than the satellite, it had a more pronounced orbital decay and assumed shorter orbits sooner. When alpha-2 (alias Sputnik-I) made its 345th revolution, alpha-1 was on its 346th. Alpha-3 (the nose cone) seems to have re-entered and burned up unnoticed; alpha-1 experienced burn-up during the first week of December 1957, over the northern Pacific.

Alpha-2 transmitted on two wavelengths—15 meters and 7.5 meters, corresponding respectively to 20.005 and 40.002 megacycles; the choice of wavelengths was probably determined by the availability of receiving equipment on the ground. After having completed 326 revolutions (22 days after launch) the batteries ran down and Sputnik-I fell silent. It re-entered and burned up during the first week of January 1958; tracking indicated that it broke into several pieces during its last few revolutions.

The latter part of 1957 was devoid of good news for Americans. One of the very few American space experiments made miscarried during September and October. This was Project Farside, which apparently had started as a moon-shot project but had been cut back to a shot to 4000 miles, or one earth radius. The Farside rockets, of which six were built, were four-stage solid-fuel rockets, launched from a Skyhook balloon, or more precisely, through a Skyhook balloon. The first stage of the assembly was a cluster of four Recruit rockets, the second stage one such rocket, the third a cluster of four Asp motors, and the fourth stage a single Asp with instrumentation. The rockets were fired by radio command when the balloon had reached the necessary altitude. Because considerable drift could be expected, the firings took place over the Pacific Ocean in the vicinity of Eniwetok atoll.

Of the six shots four were complete failures; all the stages seem to have ignited more or less simultaneously. One seems to have worked but could not be tracked because its radio transmitter failed. Only one was a limited success; at an altitude of 2700 miles, while the rocket was still climbing, the transmitter ceased to work. The rocket probably reached a peak altitude of 3000 miles.

Meanwhile, what had happened to Project Vanguard, the official American satellite project, while all this was going on? The first shot of the actual Vanguard rocket, first stage only with dummy second and third stages, took place on October 23, 1957. The official designation was TV-2 (for test vehicle), and during the preliminary tests it had given trouble at every stage of the procedure—so much so that the engineers working on it called it “a lump of garbage” and worse names. But it did make a beautiful flight, proving that the design was workable and really needed nothing but time, the one commodity that was not available.

The successful shot was a very mixed blessing for the Vanguard engineers. Public reaction was: “See, it works—even if a little late. Now for the real satellite shot.” Official reaction, though phrased differently, was basically the same. In the meantime somebody had had a new thought. The whole Vanguard program consisted of “test vehicles,” or TVs, and Satellite Launch Vehicles, or SLVs. The two Viking rockets had been labeled TV-0 and TV-1, and after TV-2 there were to be at least three more TVs. Of course the actual launching of the American IGY satellite had to wait for the SLVs, but the TVs could launch an “experimental satellite” weighing 3¼ pounds—the one that Khrushchev later called “the grapefruit”—that emitted only an electronic whistle. Since this satellite would not be the American IGY satellite, the fact that it would carry no instrumentation was not important.

The chief of Project Vanguard, Dr. John P. Hagen, was summoned to the White House for a conference with President Elsenhower to explain the Vanguard situation. Hagen emphasized that TV-3 was strictly experimental and that the orbiting of the test satellite should not be announced before-hand. Then he left for the West Coast to confer on the second stage—and the White House handed out a press release that TV-3 would launch the American satellite! The country was now clearly nervous from the top on down.

Before TV-3 was even ready for a full-scale test, the Russians launched Sputnik-II, which received the astronomical designation 1957-beta. It was much larger than Sputnik-I and the satellite was not separated from the top stage of the launch vehicle. The whole orbited as one body, with a weight of about 7000 pounds. The satellite section consisted of two parts, one a duplicate of Sputnik-I, and the second an animal capsule holding an 11-pound dog named Laika (“barker”). The perigee of 1957-beta was 145 miles, the apogee 1056 miles, the orbital period 103.7 minutes. Among other things the heartbeat of Laika was telemetered; it had gone up considerably during the ascent but settled back to normal within a number of minutes after the satellite had attained orbit. As had been hoped by some, and doubted by others, the weightlessness of an orbiting body did not interfere with the bodily functions of the animal.

After about 100 hours the batteries in Sputnik-II ran down and the dog was painlessly put to death. Because of its great mass the orbital decay of 1957-beta was slow, even though its perigee was lower than that of 1957-alpha. It stayed in orbit for 2366 revolutions and was seen by many millions of people. It finally re-entered over the Caribbean Sea late in the evening of Sunday, April 13, 1958, which happened to be Russian Easter, old style.

During the first days of December 1957, the time for the launch of “America’s answer to the sputniks,” the little test satellite in the nose of TV-3, drew near. Two American launch crews were busy; the Huntsville crew readied the fourth Jupiter-C for firing, while the Vanguard crew worked on TV-3. The Army had not set a date yet for its Jupiter-C shot, but it was clear that Vanguard would be first. Press, radio, TV, and plain spectators assembled around the Cape en masse.

It so happened that I had a lecture engagement in Baton Rouge, Louisiana, for the evening of December 4. Press and radio reported during that day that the countdown was proceeding—it had been started during the very early morning hours. The chairman of the lecture committee stationed his teen-age son backstage with a radio set and told him to interrupt me as soon as Vanguard had reached orbit. I was not interrupted; the launch attempt had been “scrubbed” (called off) after a countdown that lasted so long that some of the launch crew members simply fell asleep from utter exhaustion.

The next morning it was learned that the launch had been set for December 6. I waited around in Baton Rouge during the 5th, spending part of the day finding out how to get from Baton Rouge to the Cape without hiring an air taxi—the weather was such that I had no confidence in single-engined airplanes, though it was good enough for normal operations of large airliners. From Baton Rouge I could get a plane to New Orleans and another from there to Miami, but I would risk a so-called “lay-over” of more than 4 hours in Miami. It was a better bet to fly via New Orleans to Atlanta and south from there. I landed in Atlanta at about 12:30 P.M. on the 6th. While looking for the information counter I ran into the local manager of Eastern Air Lines, whom I happened to know. “Where are you bound?” he greeted me. “For Cape Canaveral if I can get a connection.” “Why? Haven’t you heard? Vanguard blew up an hour ago.” “In that case . . . please confirm my return flight to New York.”

Again the countdown had started during the very early morning hours. The earlier long countdown had been plagued by many small leaks in the system, but this one seemed reasonably troublefree. The firing button was pushed at 11:44:35 A.M. on December 6, 1957. The rocket engine of the main stage of TV-3 caught fire, burned a little half-heartedly for about half a second, and then steadied, and the rocket began to lift. After about 2 seconds the launch crew saw a stiletto-like flame issuing from the side of the rocket, in the area of the upper part of the rocket engine. Simultaneously the thrust dropped below the weight of the rocket, which settled down, then toppled over. The fuel tanks split open, and the launch pad became a mass of flames. After it was all over the launch crew found to their amazement that the solid-fuel third stage, though surrounded by flames, had not ignited and that the little test satellite was whistling away as if it were in orbit.

“America’s answer to the sputniks” had turned into the most publicized failure in history. The country was shocked; there were the customary cries for an investigation (as if such an event could come to pass without being investigated); and the foreign press was scornful, even the British press. If it had been the first attempt at a satellite launching anywhere, there might have been some sympathy, but two Russian satellites were orbiting overhead. A West German magazine compared the triumphant advance announcement to the “discovery” that Easter Island had disappeared beneath the waves.11

The actual reason for the failure has never been fully established. Of course the cause had been a fuel leak somewhere between fuel tank and rocket engine and the most probable cause of a fuel leak is a misalignment. The engineers from General Electric (which had built the engine) blamed the Martin Company (which had built the rocket); the Martin engineers took the opposite view. While the argument was never resolved it is worth noting that, if the rocket had worked properly, it would have been the very first flight of the second stage, which had never been test-flown.

A few days after my return to New York I faced the television cameras of the ABC network and its commentator Quincy Howe. To what did I ascribe the failure of Vanguard? To the fact that it is a new and untried rocket. Did I think the Jupiter-C rocket that had been announced as being readied for a satellite shot would make it? I said I did not doubt that it would—having in mind that a Jupiter-C rocket had already almost made it. When, in my opinion, would the Jupiter-C make it? This was a difficult question and I thought fast. Christmas was only a few weeks away. I doubted that the rocket could be ready that soon, and for some reason I felt sure that nobody would shoot between Christmas and New Year’s. I said: “In January, Mr. Howe, in January.”

There was one interesting factor of which I was not aware at that time. If “Missile 29” had been permitted to go into action in 1956, it would have produced the first artificial satellite but this would have had nothing to do with the IGY satellite which was to be launched as part of the Vanguard program. Now, when nobody believed any longer that Vanguard would ever make it, the Jupiter-C shot became the IGY shot and the satellite had to carry instruments. The Vanguard satellite, spherical in shape, was not too heavy for the Jupiter-C, but it was too large. The instruments had to be rearranged to fit into a cylindrical tube.

Everything was ready late in January 1958; except for one factor the shot could have been made on January 28. That factor was the “jet stream,” a very fast constant current of cold clear air that moves at high altitudes above the United States. It travels always from west to east, but over different latitudes, sometimes farther north than at other times. Late in January 1958 it happened to be exceptionally far south. If the rocket had to go through the jet stream the shot might be ruined. On the 30th it was decided to make the rocket ready to the point where the liquid oxygen had to be put into the tank, and then, on the basis of the jet-stream weather forecast, decide whether to go ahead. Balloons were flown into the jet stream and the computer at ABMA ran an analysis.12 At 9:20 P.M. the message came: Highly marginal—we do not recommend that you try it. The firing was canceled. The next morning the jet stream started shifting northward; in the evening its velocity over the Cape was expected to be down to about 110 miles per hour. The Jupiter-C, it was known, could cope with a 110-mile-per-hour jet stream, but the night before it had been about twice that.

Jupiter-C no. 4 lifted off at 10:48:16 P.M. on January 31, 1958. Two minutes and 36 seconds later the first stage had Brennschluss, 60 miles up. The upper stages coasted to the peak of the trajectory to be fired by radio command. Two minutes after this command the top stage was in orbit. The men at the Cape as well as those in Washington could know at the moment only that the top stage had the necessary velocity. The instrumentation that could be carried did not reveal whether it was also pointed in the right direction. The ground tracking network had not yet been developed to the point where one can tell within 10 minutes that orbit has been attained; a wait of about 1½ hours was required until one could be certain. But orbit had been attained. The word came: “Goldstone has the bird,” meaning that the big Goldstone radar in California had picked up the satellite as it crossed the West Coast, having nearly completed its first orbit.

At 1 A.M. on February 1, America was jubilant. The Huntsville radio urged all citizens to come to the town square for a celebration. They came, with signs reading “Our Missiles Never Miss”; they also hanged ex-Secretary Wilson in effigy, apparently not wishing to have their feelings misinterpreted, and changed a sign at the city limits from The Missile Center of the USA to Space Capital of the Universe.

At 1:30 A.M. a long-distance call from Washington woke me up; the office of the Associated Press told me that the satellite was in orbit, adding: “We thought you would like to know.” They had also ruined my sleep for the next four or five hours. I sent a wire to Wernher von Braun, saying: “Congratulations! And thank you for having kept me an honest man by the margin of one hour and twelve minutes. I had told everyone that Jupiter would jump into space in January.”

The first American satellite received the name Explorer-I and the astronomical designation 1958-alpha. As with Sputnik-II, there was no separation of the instrument package from the top stage of the carrier rocket; they orbited as a unit. The first apogee was at 1573 miles, the perigee at 224 miles, and the orbital period 114.8 minutes. The lifetime in orbit, predicted from these figures, was 3 years, but Explorer-I had a number of surprises in store. One was that it would still be in orbit in 1967, though by that time the orbit had decayed to some extent. The figures for Explorer-I for January 1967 read: apogee down from 1573 miles to 890 miles, perigee down from 224 miles to 210 miles, orbital period 102.6 minutes.

The Vanguard rocket labeled “TV-3 back-up” was on its pad when Jupiter-C jumped into space. It had been scheduled for February 3, because the IGY committee had informed the National Science Foundation that they could handle the tracking of satellites only if they were fired at least three days apart. But valve trouble forced a postponement to February 5. For the second stage of Vanguard this was still the first flight, and again the second stage had no chance to prove itself. “TV-3 back-up” lifted off on February 5 and looked beautiful, but 57 seconds after lift-off it broke apart in midair and exploded.

The next Jupiter-C shot, carrying Explorer-II, was scheduled for March 5. There was an unexpected “hold” for 10 minutes in the countdown, because the first Explorer was passing overhead and existing instrumentation would have had difficulty in disentangling the signals from two satellites relatively close together in space. But Explorer-II did not reach space, in spite of a perfect take-off. After the necessary wait it became clear that there was no orbiting satellite. For unknown reasons the top stage had not ignited and the shot had turned into an unwanted repetition of the shot of “Missile 27.”

On March 17 Vanguard TV-4 was ready. This time the first stage behaved, the second stage made a flawless maiden flight, the third stage performed as expected, and the little test satellite became Vanguard-I and 1958-beta. It was put into such a high orbit (perigee 409 miles, apogee 2453 miles) that it would never re-enter on its own accord. After Explorer-I finally decays, Vanguard-I will be the oldest artificial satellite in orbit and will remain the oldest until it is removed by a manned spacecraft to be put on exhibit in the National Air and Space Museum in Washington.

On March 26 another Jupiter-C put Explorer-III into orbit as 1958-gamma.

Meanwhile, where were the Russians?

The Russians were getting ready to show the world, and especially the United States, what they could do. The orbiting test satellite of the Vanguard program weighed 3¼ pounds; the Explorers weighed about 31 pounds. Sputnik-III, fired on May 15, 1958, weighed 2925 pounds. The total weight in orbit was well over 8000 pounds.

The American public was flabbergasted once more. Everybody actively engaged in the space program received a large volume of mail which could not even be dismissed as “crank mail” because in most cases the letter writers were honest—and worried—citizens who sincerely tried to help. Of course they talked nonsense in 90 per cent of the cases, but this was due merely to lack of scientific or engineering knowledge. I was not a member of any government organization or even of a semiofficial committee—in fact I had no connection with the space shots that were going on or being planned except that most of the people who were shooting Jupiters and Vanguards were good personal friends—but I got my share of such mail too.

Of course I don’t remember all the letters, but those I do remember had one thing in common: their authors had sat back and “thought.” They had “reasoned out” what the Russians were using and what we, therefore, should also use. A fairly common conclusion was that the Russians must be using an enormous catapult to throw their rockets into the air. My reply to this kind of letter was almost standardized: I pointed out that a catapult capable of imparting a useful velocity to a rocket probably weighing 100,000 pounds would have to be very large and enormously expensive, provided that it could be built at all. Another common conclusion was that the Russians were using atomic energy, and why did our atomic energy “boys” sit back, resting on their laurels? If it was not atomic energy, then it was a new fuel. One letter writer who “knew” that the Russians were using a new fuel also knew that our own chemists were deliberately ignoring the problem, being clandestine members of the Communist Party. He had reasons for saying so; he had written to the research department of a big chemical company and the company had answered that there was no reason to believe that the Russians were using an unknown fuel. One man wrote that all the people who were guessing at new fuels or catapults were barking up several wrong trees. The secret of the Russian successes was that they had discovered a metal that did not weigh anything!

Not many of the letter writers were women, but the few who were introduced a number of fascinating thoughts. One wrote to say that her nephew, who was in college at the time, had heard the true story from a fellow student who was a German refugee. The Germans, late in 1944, had developed a fuel that was five times as powerful as any fuel we used because it was a solid, not a liquid. And the man who had this fuel had been captured by the Russians and collaborated with them because he was a Communist. My suggestion that this German refugee had probably amused himself by inventing a tall tale was brushed aside with the hint that I was probably a Communist too.

I thought, then, that this letter represented a kind of apogee, but some years later I had to change my mind. Another woman wrote, enclosing a reprint of an article by the Russian physicist Pyotr Leonidovitch Kapitsa. Professor Kapitsa’s article dealt with ball lightning, a rare electrical phenomenon. He tried to evolve a theory of ball lightning, to find what conditions have to prevail during the appearance of a lightning ball. Kapitsa did not succeed and said so in the introductory paragraph of his article. But never mind what he said; according to my correspondent, the secret of the Russian space shots was that they don’t use ordinary fuels at all. They were propelling their rockets by ball lightning!

I don’t recall how often I wrote (or said after lectures during the question and answer period) that the “secret” was not a new metal or a new fuel, and certainly not a catapult or a vertical tunnel in a mountain. The secret, I kept repeating, was that the Russians had bigger rockets.

I hope that at least a few people believed me.

6. The Loki was a development of the unfinished Peenemünde Taifun, but for solid fuels. It has since been dropped from the list of potential or operational weapons and trans- formed into the HASP (High Altitude Sounding Projectile).

7. The date of the IGY was chosen because, 25 years earlier, there had been a South Polar Year, also with international cooperation. And the South Polar Year had been preceded, 50 years before it, by a Polar Year in which those countries near the Arctic Ocean cooperated in the exploration of the Arctic regions.

8. This probably was a truthful statement; the Russians’ intercontinental missile program and their satellite program were two sides of the same coin. By December 1955 they might not have known how far their missile program would progress during 1956 so that they could not tell whether a satellite program was feasible.

9. The Jupiter-C missile had an elongated Redstone as its first stage; the second stage consisted of 5 solid-fuel Recruit rockets arranged in a circle around a large center hole, the third stage was a cluster of 3 Recruit rockets fitting into that hole, and the fourth stage was a single such rocket on top of the third stage. The whole was enclosed in what was called the “launch tub,” or “the basket,” a cylindrical metal container on top of the Redstone. Prior to take-off the “basket” was rotated rapidly by an electric motor.

10. President Elsenhower, in a TV appearance, referred to one of these shots as “an artificial comet” but never explained what he had meant by that term.

11. In 1922 or 1923 an American destroyer made port with the announcement that it had failed to find Easter Island and that the island therefore was assumed to have “sunk.” The fact was that, in very bad weather, the destroyer had simply failed to find it.

12. General Medaris, the launch chief Dr. Kurt Debus, and the launch crew were at the Cape, while Wernher von Braun and Dr. Pickering, who was responsible for the instrumentation in the satellite, were in Washington. In view of the political importance of the shot this was necessary, but it resulted in an incredible number of long-distance telephone calls.

Wednesday, August 25, 2010

Man in Space? “Utter Bilge!”

From Rockets, Missiles, and Men in Space by Willy Ley, 1968 edition, pg. 361:

In 1959 the newly appointed Astronomer Royal of England, Australian-born Richard van de Riet Wooley, told the British press that space travel was “utter bilge.” Ever since, the British Interplanetary Society has had a fine time giving him reports such as “An American named Carpenter has penetrated utter bilge for the fourth time.”

Tuesday, August 24, 2010

Knocking the Atlas Missile – With Hammers

The Atlas missile was a highly capable launch vehicle due to its very low structural mass. The place where the greatest mass saving was realized was in the fuel tanks which were, in effect, giant, load-bearing, stainless steel balloons whose strength came not from their dime-thin walls, but from their internal pressure. Without that pressure, the Atlas would have collapsed under its own weight, but, kept properly pressurized, it could throw a circa 1954 hydrogen bomb 5,500 nautical miles, or, more significantly, launch Mercury capsules, and countless other non-lethal payloads into Earth orbit, or beyond (for instance, they launched the first probes to Venus and Mars). But, while it was being developed, the “balloon tank” made a lot of people nervous.

From Rockets, Missiles, and Men in Space by Willy Ley, 1968 edition, pg. 328:

The tank of the Atlas consisted of stainless steel which was no thicker than a dime at any point. Generals—and later congressmen—worrying about what would happen if somebody dropped a wrench on such a tank were conducted to a test version of the “stainless steel balloon” that was stiffened by pressure and offered a choice of assorted mallets to see whether they could dent it. In each case the general or congressman grew tired before he had even succeeded in producing a mark that could be seen.

As I understand it, to this day, if you see an Atlas missile on display, hidden in the background somewhere will be a compressor working (or ready to switch on automatically) 24/7 to maintain the necessary level of pressure in its tanks.

Monday, August 23, 2010

Astronaut Benefits

From Walter Cunningham’s forward to In the Shadow of the Moon, page xii:

[....] When I went to work as an astronaut, in 1963, I earned a little over $13,000 a year. I once calculated that, during my Apollo 7 mission, I had earned the great sum of $660. But we weren’t doing it for the money—nobody does a job like that for the money. Any one of us would have paid NASA to have the job!


Something else we did not have and were unable to obtain in those days was life insurance. The second year I was there, we thought we might get coverage when NASA went out for quotes to renew the agency’s health and accident policy. They actually solicited two quotes. Insurers had to submit one bid with the thirty astronauts included for death benefits and another bid if we were not covered. The difference was, apparently, quite significant because NASA, bless its sweet little heart, accepted the bid with astronauts not included!

“Astronaut” as a pre-existing condition?

Saturday, August 21, 2010

Video: Bats’ Morning Return to the Bamberger Chiroptorium

I had the opportunity last week to watch the bats return to the Bamberger Ranch Preserve’s chiroptorium. The morning return isn’t as dramatic as the evening emergence, but it’s still something to see. Realizing that, for the first time, I had a device that could capture high definition (720p) video, I propped up my iPhone 4 on the fence at the mouth of the chiroptorium and let it record for about 16 minutes. I’ve edited that recording down to two minutes in the following video. It is best seen at 720p - at lower resolution, the distant bats in the descending stream aren’t visible.

As the bats swooped down and maneuvered aggressively into the cave mouth, they produced a sound that I can best describe as a rapid vibration. My guess is that the trailing edges of their wing membranes vibrate violently during those maneuvers. (Much like the end of a piece of fabric held out of the window of a fast-moving car.) Unfortunately, that sound seems to represent some sort of pathological case for the iPhone 4’s audio capture system, and, as you’ll hear in the movie, it is distorted into something that sounds very much like static. That’s disappointing, but slightly better than no sound at all, I think. So, apologies about the static. It’s not an accurate reproduction of the sound produced by the bats, but its presence and intensity does correlate with the number of bats passing the camera at any given moment, so it has some value.

It is, of course, better to experience this first-hand, but one needs special permission to do so. Being fortunate enough to have that permission, it seems the least I can do to try and share some of that unusual experience. Enjoy.

Thursday, August 19, 2010

Java Text – A Closer Look at the Ugly

Because the failings of the native Java text renderer have had such an adverse impact on my recent work, I think additional examination of the issue is merited. So, to further illustrate the difference between text rendered by Java 6 on Mac OS X and Linux, I’ve taken two paragraphs of text and written a program that fits them on-the-fly into an image of a specified size. Text fitting is vital to the operation of the application from which yesterday’s examples were taken, so it seemed appropriate to create new examples based on text fitting as well.

[By the way, this is in no sense a dig at Linux. I believe the text rendering seen on Linux is representative of Java’s built-in text rendering capabilities, whereas I believe the text rendering seen on Mac OS X is representative of OS X’s native text renderer, which I suspect Apple engineers substituted for Java’s built-in text renderer. Running the code on Linux just happens to let the ugly truth shine through. As with yesterday’s examples, the version of Java used on the Linux box was “Java(TM) SE Runtime Environment (build 1.6.0_12-b04)” and the version on Mac OS X was “Java(TM) SE Runtime Environment (build 1.6.0_20-b02-279-10M3065)”.]

The font used in all cases is Georgia, and the font binary is identical on both installations. The quotation used is from Aldo Leopold’s book A Sand County Almanac, and Sketches Here and There, which I highly recommend. (More of the quote can be found on my page of random Leopold quotes.)

The fitting errors cited below are the difference between the height, in pixels, of the area into which the text was fit, and the actual height of the text. The closer the error is to zero, the better. Considering that the errors are spread over 14 lines of text in the small font example, and 30 lines in the large font example, even the big errors aren’t significant. Nonetheless, the errors consistently show a better fit is achieved when the code is run on Mac OS X, which suggests, at a minimum, the availability of more precise font metrics, and possibly the ability of the font renderer to actually produce (subtly) different glyphs for fonts whose sizes vary by only small fractions of a point.

Small Text Rendering

This is where the difference between Java’s text rendering on Mac OS X and Linux is most obvious. Figure 1 shows a 4X magnification of approximately 12 point text when the Java test program is run on Linux. Figure 2 shows the same magnification of the text produced by the test program when run on Mac OS X. (Figure 3 shows the full renderings at normal size, in an alternating animation.)

The difference in the antialiasing algorithms is striking. In figure 2, Mac OS X shows that antialiasing is applied to every pixel of every glyph, and that the sub-pixel glyph placement this permits allows for more consistent spacing between glyphs (although one can readily find places where either renderer could do a better job of kerning). It also provides for a more consistent “color” (uniform darkness or lightness) of text. Figure 1, by comparison, shows a minimal use of antialiasing, primitive glyph shapes, some highly irregular glyph spacing, and poor color.

An examination with Adobe Photoshop shows only 127 shades of gray are used in the text from which figure 1 was taken. In contrast (no pun intended), the same examination of the text from which figure 2 was taken shows 249 shades of gray are used. So, in this case the default Java font renderer is using half of the available shades of gray to form its glyphs, while the Mac OS X renderer is using almost all of the available shades. That alone gives the Mac OS X renderer almost twice the ability to represent sub-pixel glyph features, an ability of which it seems to take full advantage. Every glyph in these two examples demonstrates this, but, if you want a specific example, take the word “July” on the sixth line. Observe that in figure 1 the bottom of the J’s curve is flat against the baseline (so much for the curve) and very nearly black, while in figure 2 its curve is carefully nuanced right down to two faint gray pixels dropping below the baseline (one is relatively obvious; the other you will probably need further magnification, or a pixel sampler, to see). The descender on the letter “y” is also worth a look. As is plainly visible in figures 4 and 5, that descender is meant to be terminated with a feature similar to a dot. However, in figure 1, there’s no trace of that dot, though it’s clearly present in figure 2.

Another difference in this rendering of small text is most obvious in the figure 3 animation: While the two renderers agree about overall line heights, they disagree on the height of the glyphs by a full pixel. Given that the lines are about 14 pixels high (leading included), that’s a 7% error, though the lack of subtlety in the rendering produced on the Linux machine results in an even greater apparent difference, at least to my eyes.

Next, there’s the matter of large text rendering.

Figure 1. Text rendered by Java 6 on Linux at 11.968737 points
(fitting error: -0.107483), magnified four times.

Figure 2. Text rendered by Java 6 on Mac OS X at 11.944245 points
(fitting error: -0.000177), magnified four times.

Small text rendering examples.

Figure 3. An animated comparison of small text rendering by Java 6 on Mac OS X and Linux. The magnified text seen in figures 1 and 2 was taken from the two renderings from which this animation was produced, seen here at their original size.

Large Text Rendering

The differences between what I’m assuming to be the native Mac OS X text renderer as used by Java 6 on Mac OS X and Java 6’s built-in text renderer, as used on Linux, are less significant at the relatively large font sizes (almost 26.5 points) used in the following figures, but they’re real enough. The figure 6 animation clearly shows many differences in letter spacing, and, while neither renderer is producing ideal results, Mac OS X looks to my eyes to be doing better overall.

Also, the magnifications shown in figures 4 and 5 suggest that, once again, the Java renderer is using fewer shades of gray when it antialiases text than the Mac OS X renderer. And that is, in fact, the case. An examination with Photoshop shows only 123 shades are used in the Java-on-Linux rendering, while the Java-on-Mac-OS-X rendering uses all 256 shades (both numbers include black and white).

Once again, that—and perhaps other failings of the renderer—leads to a lack of subtle detail in the text. Observe that in figure 5 almost every curve or vertical serif that touches the baseline actually extends around half a pixel below it. That doesn’t happen in figure 4. Yet, extending a curve slightly below the baseline is a basic trick of typography – as a curve reaches its bottom it becomes less and less visually significant. If it bottoms-out precisely on the baseline, then it touches the baseline, in principle, at an infinitely small point, and—though an infinitely small point can’t be rendered—the curve nonetheless has reduced contact with the baseline relative to surrounding features that have large areas in contact with the baseline (horizontal serifs, for instance). And that difference in the degree of baseline contact tends to make the glyph with a curved bottom appear to float slightly above the baseline. So, typographers design their fonts to correct for this perceptual error, by dropping the bottom of curves slightly below the baseline. Typographers have a lot of little tricks like that – more, I’m sure, than I’m aware of. And that’s why antialiasing is so important to accurately preserving the look of text on computer displays, at least until our displays exceed something on the order of 300 dots-per-inch of resolution.

The folks behind the text renderer used by Java on Mac OS X clearly demonstrate an understanding of these issues and their importance. There are, nonetheless, strange imperfections in their results, but those seem to stand-out because so much else is done well. The folks responsible for the text renderer used by Java on Linux, which I assume to be Java’s native text renderer, don’t seem to care about these issues. Or, if this article is correct (and I have no reason to doubt it), the Java text engineers get it, but their management doesn’t. (Or it hasn’t – we’ll see if the new Oracle management is any better than the old Sun management, though I see no reason for optimism.) Thus, not only the subtle aspects of the art of typography—those that bend type around the particulars of human perception—but the grossest qualities of glyph formation and spacing are … well, I struggle for the right word. “Trashed,” isn’t quite it, but it’s in the correct neighborhood. And that neighborhood should also include either a pervasive failure of perception, or a cretinous disregard for the quality of what one perceives. Maybe both.

Figure 4. Text rendered by Java 6 on Linux at 26.406231 points
(fitting error: -0.957642), magnified four times.

Figure 5. Text rendered by Java 6 on Mac OS X at 26.403095 points
(fitting error: 0.000000), magnified four times.

Large text rendering examples.

Figure 6. An animated comparison of large text rendering by Java 6 on Mac OS X and Linux. The magnified text seen in figures 4 and 5 was taken from the two renderings from which this animation was produced, seen here at their original size.

The Original Renderings

If you’d like to see the original renderings from which the magnifications and animations were generated, here they are.

Wednesday, August 18, 2010

Java Text - When Platform Independence Fails

I‘ve recently been working on a Java web application that does server-side rendering of complex graphics that include text. Having sweated all the details to maximize the quality of the application’s output; having finally made the application sufficiently stable and feature-complete to enter beta testing; and having jumped through some unrelated hoops to make the production host usable, I finally found myself at the point where I could deploy the application to its production server.

I thought I was on the brink of success at long last, but when I tried the web application on its production Linux VM, the results were terrible. The text was rendered very differently between Java 1.6 on Mac OS X, which is my development environment, and Java 1.6 on Linux, which is the deployment environment. Specifically, the resolution-independent, anti-aliased, fractional-sized text was rendered beautifully by Java on Mac OS X, and horribly by Java on Linux. You can see for yourself in the examples below. (Due to confidentiality rules, I can’t show the entire rendering, which is unfortunate in this case, because the differences are even more striking when seen en masse.)

Text rendered by Java on Mac OS X, using
“Java(TM) SE Runtime Environment (build 1.6.0_20-b02-279-10M3065)”.

Text rendered by Java on Linux, using
“Java(TM) SE Runtime Environment (build 1.6.0_12-b04)”.

Above, the two samples converted to an animated GIF and perpetually alternating.

Note that in both cases, exactly the same typefaces are being used: Arial (bold and plain) in the yellow area, and Georgia (plain) for the names above. Also in both cases, the rendering was performed in a Graphics2D environment with all of its hints set to maximum quality, and the backing BufferedImage was of the same type (TYPE_INT_RGB). Both systems are even using exactly the same TrueType fonts; I copied them from my Mac to the JRE on the Linux box. Nonetheless, Java’s font rendering under Linux looks like some early, unfinished attempt at supporting TrueType, anti-aliasing and fractional font sizes/metrics. Java on Mac OS X, on the other hand, handles all of the text beautifully.

My best guess is that the results on Linux are representative of the true capabilities of Java's built-in text renderer, while the very different results on Mac OS X are probably due to Apple’s Java engineers substituting OS X’s native font renderer for Java’s. Others seem to have reached similar conclusions in the past.

So, because of these platform dependent text rendering behaviors, I’ve gone from believing that I was ready to deploy an application whose quality was going to reflect well on myself and my organization, while delighting its users, to being so ashamed of the graphics quality that I don’t want to proceed with the deployment; I don’t want to be associated with such poor quality, and I don’t want it being seen as representative of the care that my organization puts into its work.

Arguably, the platform independence was undone by Apple (if the scenario above is correct), but Java’s native text rendering is so bad, I can’t blame them. In fact, I’m glad there’s at least one platform that can run Java and render text well. The real fault seems to be with Sun’s (now Oracle’s) management for not giving Java’s text engine the attention it deserves.

Of course, if anyone knows that I’m wrong about those conclusions, and can show me how to get the same quality of text rendering out of Java on Linux as I get from Java on Mac OS X, I’d be very grateful. And so, I think, would others.

Help if you can, please. Beware of this issue, if you can’t.