Thursday, December 27, 2012

A Jones for MOL #7: Hatches? We Don’t Need No Stinking Hatches!


I addressed the topic of hatches in MOL back in #3 of this series. But there is more to be said on the topic. Plus, I just recently thought of the title for this blog post, which is much better than the title of #3, and could not resist using it. If I compile these individual posts into a longer article or a book chapter, I will use the title from this blog post. For a brief history of the entertaining catch phrase behind this title, see “Stinking Badges” at Wikipedia (ref. 1).

In #6 I explained the evidence for my assertion that the Gemini-B capsule was to be sealed and depressurized for most of the 30-day manned MOL mission. The isolation of the capsule seems straight-forward. One obvious approach would be for the last pilot leaving the Gemini-B to close a hatch at the lab end of the transfer tunnel, permitting the subsequent depressurization of the transfer tunnel and the adjoining open capsule while maintaining the air pressure in the MOL. But as noted in #3, the evidence for a hatch at this location is ambiguous at best. I concluded that there was no independent pressure-seal capability for the transfer tunnel: the tunnel would have been open from the Gemini-B heat shield all the way to the interior of the MOL habitable volume.

This is not unique in spaceflight: the Spacelab modules carried in the payload bay on many Space Shuttle flights were connected by a long transfer tunnel to the Shuttle’s crew compartment, and there was no hatch at the Spacelab module opening (ref. 2). But the crew module of the Shuttle was never intended to separate from the Spacelab while in-flight, while the Gemini-B was required to separate from the MOL for return to Earth. This will be the topic of a future blog post in this series.

If the Gemini-B/MOL tunnel could not be closed off at the lab end, then maybe it would have been possible to seal off just the capsule by reversing the large pressure bulkhead (LPB) hatch, passing it through the hatch opening and then re-installing it from the tunnel side to seal the Gemini-B’s large pressure bulkhead  (see Fig. 1). In this scenario, the separate heat shield (HS) hatch would have stayed stowed in its alcove in the wall of the transfer tunnel. 
Figure 1. McDonnell Douglas illustration of the crew transfer hatches in the Gemini-B: the large pressure bulkhead (LPB) hatch and the heat shield (HS) hatch. Note the common features of both hatches, including the crank handle and the six dog arms. Note also the stowage location for the LPB hatch in the small alcove between the seats.

The following analysis is based on my reading of the collected Gemini-B/MOL documentation, much of it provided by Dr. Dwayne Day from his visit to the U.S. Air Force archives at Maxwell AFB in Alabama in late 1999. The drawings from those archives are detailed and thorough, and bear corporate file numbers but not dates. However, a date may be inferred from the fact that the drawing numbers all contain the abbreviation “MDAC” clearly indicating the McDonnell Douglas Astronautics Corporation. McDonnell Douglas was formed from the April 28, 1967, merger of the McDonnell Aircraft Company, builders of the Mercury and Gemini spacecraft and prime contractor for the Gemini-B, and the Douglas Aircraft Company, prime contractor for the MOL itself (ref. 3). Thus, I have assumed that the drawings represent mature designs for their subject components, since MOL was heading into manufacture by 1967 and was cancelled just over two years later.

There are some problems with the approach of reversing the LPB hatch. First, the hatch was too big to be passed through the hatch opening. The diameter of the hatch was 25.1 in. (63.8 cm.) (per Fig. 1), but the hatch opening diameter was only 24 in. (61.0 cm.) (per Fig. 2). This does not mean that such an approach is impossible under any circumstances: I have watched in amazed disbelief as a large airliner door is swung inward and then snugged up to the door frame from the inside to make an airtight seal, but that apparently requires some rotation of the rectangular doors to pass through the frame and then de-rotate to fit snuggly. No amount of angling or rotation would have allowed the circular hatch to pass through its circular opening that was over an inch smaller.
Figure 2. McDonnell Douglas illustration of the edges and seals of the LPB and HS hatches. (In this illustration, the LPB hatch is labelled as the crew transfer hatch.) I have highlighted the LPB knife edge and the corresponding gasket, neither of which were labelled in the original illustration. I have also indicated the area of interest in the small diagram of the Gemini-B, to the left of the hatch detail.

Second, the cross-section of the LPB hatch shows it to have had a type of “knife-edge” seal (ref. 4) which would work only if the hatch were oriented correctly (see Fig. 2). An extended ridge called the knife edge near the rim of the hatch would be compressed a recessed gasket in the rim of the hatch opening to make the seal airtight. The gasket was only on the Gemini-B side of the opening, so if the hatch had been flipped around so that the crank handle was accessible from inside the transfer tunnel, then the knife-edge seal would have been facing away from the gasket.

Third, the crank handle for securing the hatch was only on the Gemini-B side of the hatch (see Fig. 2). This handle was to be turned to drive six dog arms on the hatch into corresponding fittings on the hatch opening rim, mechanically wedging the knife edge of the hatch onto the gasket of the hatch opening, making the seal airtight. Conceivably the hatch could have been left in its usual place on the Gemini-B side of the opening, but the handle moved to the tunnel side, or a second handle could have been inserted on the tunnel side of the hatch. But the available diagrams give no evidence that a handle could be removed and reinserted on the MOL side. In addition, leaving the hatch on the Gemini-B side of the opening meant that it would have been on the vacuum side of the bulkhead when the capsule was depressurized for the 30-day duration of the mission. The hatch was clearly designed to be on the high-pressure side of the bulkhead, using the air pressure of the manned capsule to reinforce the seal against the gasket. Whether it could have provided an adequate seal for 30 days if the air pressure from the MOL was trying to push it into the depressurized Gemini-B capsule is unclear to me, but I suspect not.


Fourth, I just realized that there would have been no need for the permanent structure between the seats to retain the LPB hatch (see Fig. 1), if the hatch were to have been used to seal the tunnel for up to 30 days. A temporary stowage location, perhaps just velcroed to a wall, would have sufficed.

Thus, it appears that the LPB hatch could not have been reversed, installed and sealed from the tunnel side.

Summarizing the key points of several blog posts:
  1. The Gemini-B capsule was to be evacuated, sealed and decompressed for the duration of the MOL missions (discussed in #6 in this series).
  2. There was no hatch or other fixture to seal the transfer tunnel from the Gemini-B to the MOL at the lab end of the tunnel (discussed in #3 in this series).
  3. The LPB hatch of the Gemini-B capsule could not be made to seal nor could it be operated from the MOL side of the hatch (as discussed in this article).
I can only reconcile any two of these three points. First, there would be no need for either a transfer tunnel hatch or a reversible LPB hatch if the Gemini-B were not to be depressurized. (This is the only case for which the title of this blog post is literally correct.) Second, there would be no need for a tunnel hatch if the LPB hatch were operable from the tunnel side of the capsule’s heat shield. Third, there would be no need for a reversible LPB hatch if there were an operable hatch at the MOL end of the tunnel.

But at this juncture, I cannot say how all three of those conditions were to be fulfilled simultaneously on manned MOL missions. It is a mystery, and I hope that someone reading this blog post knows the answer and will share it with me. If that happens, I will pass it on.

Why this was important to the end of the crew-occupied phase of each MOL mission, when the Gemini-B was to separate from the MOL and begin its return to Earth, will be the topic of a future blog post, but first I want to address other biomedical and human factors aspects of the MOL missions.

References
  1. “Stinking Badges,” http://en.wikipedia.org/wiki/Stinking_badges (accessed 17 Dec. 2012).
  2. Spacelab News Reference, NASA Marshall Space Flight Center, document number 14M983 (undated, ca. 1980), Sec. 3.2.2. Tunnel Systems, p. 3-19, http://www.scribd.com/doc/44577396/Spacelab-News-Reference (accessed 29 Aug. 2012).
  3. “McDonnell Douglas,” http://en.wikipedia.org/wiki/McDonnell_Douglas (accessed 27 Dec. 2012). There is surprising ambiguity on the Internet whether the “A” in MDAC” stands for astronautics, aerospace or aircraft. I have opted for astronautics.
  4. For a discussion of seal types in nautical hatches, see Freeman Marine Equipment, http://www.freemanmarine.com/Hatches2400Summary.htm (accessed 6 Aug. 2012), and Standard Equipment Co., http://www.standardequipmentclosures.com/marine-closures/boat-hatches/ (accessed 27 Dec. 2012).

Monday, December 17, 2012

A Jones for MOL #6: Not Quite a Vacuum.


If you have read any of my preceding “A Jones for MOL” blog posts, you might have been wondering where all of this hypertrivial analysis of a defunct spaceflight project was leading. Your patience will now be rewarded, and if your response is, “That was the blockbuster?” I will also tell you why it is so interesting that it motivated these six blog posts and a few more yet to come.

The Revelation

Here it is: during their 30-day mission, while the two MOL pilots were to have been evaluating space reconnaissance techniques as well as being guinea pigs for space biomedical research, their Gemini-B capsule would have been sealed off and depressurized to 0.1 psi (5 mm. Hg, 0.7 kPa, 0.007 atm)—the same as atmospheric pressure at 42 km (26 miles) above Earth’s surface (ref. 1). Humans need supplemental oxygen above about 3 km (10,000 feet) altitude, and space suits above 19 km (12 miles), so the cabin pressure would have been just a smidge better than a vacuum, uninhabitable for the duration of the mission.

I stumbled upon this surprising and unexpected aspect of the MOL’s human factors plan while re-reviewing some declassified MOL documents at NASA’s Technical Reports Server (ntrs.nasa.gov).  As far as I know, this has not been reported anywhere since 1968. I do not remember ever having seen it reported, or even mentioned, anywhere else, certainly not among the rote, repetitive assertions that accompany any new discussion of MOL on the Internet.

Powered Down

It has long been known that the 90 ft3 (2.6 m3) Gemini-B capsule (ref. 2) was to be powered down, that is, many of its electrical systems were to be turned off, during the 30-day operational mission of the MOL reconnaissance station. Such a power-down has been, and continues to be, common practice in space station missions. It was not even unprecedented during the MOL planning period, although the single time it had happened before was unplanned. In August 1965, Gemini 5, the third manned capsule in the series, was testing electricity-producing fuel cells in flight for the first time. These were required for future long-duration missions, including the Apollo moon flights. According to NASA (ref. 3), “About [2 ½ hours after launch], the crew [command pilot Gordon Cooper and co-pilot Charles Conrad] noticed the pressure in the oxygen supply tank of the fuel cell system was dropping. At some point earlier in the flight the oxygen supply heater element had failed, and the pressure dropped from nominal pressure of 850 psia to a low of 65 psia 4 hours and 22 minutes into the flight. This was still above the 22.2 psia minimum but it was decided to […] power the spacecraft down. An analysis was carried out on the ground and a powering up procedure was started on the seventh revolution. Over the rest of the mission the pressure slowly rose in the fuel cells and sufficient power was available at all times.”

After entering the lab module, an early task of the Gemini-B co-pilot would have been to activate the lab’s systems so the command pilot could then shut down most of the Gemini-B’s systems for long-term quiescence. All guidance and navigation and communications systems would be off or on stand-by. Spacecraft power from the laboratory would preserve the capsule’s non-rechargeable batteries for re-entry. During the month-long mission, the Gemini-B would remain quiescent, and the pilots in the lab would monitor 10 parameters of its systems, including the pressure in two gaseous oxygen and two nitrogen tanks, in two fuel and two oxidizer tanks for the maneuvering engines, and the status of two coolant pumps. Meanwhile, 73 Gemini-B parameters would be monitored from Earth via the lab’s communication system whenever its orbit passed over a tracking station (ref. 4).

The phrase “powering down a spacecraft” brings to mind the images of the inert command module in Apollo 13 (both the movie and the actual spaceflight). The cabin would not be continually heated by its electronics, and would get cool. Then it would get damp, too, as the pilots’ warm, moist breath diffused throughout the open tunnel from the laboratory module into the capsule. This had been a problem on Gordon Cooper’s 34-hour Mercury flight in 1963: the short-circuiting of critical re-entry control systems led to improved packaging for electronics on the Gemini spacecraft, as well as relocating much of those systems outside the crew cabin, in the vacuum of space. Ironically, Cooper faced the challenge of a power-down on his second spaceflight on Gemini 5.

Closed and Depressurized

After its power-down, the Gemini-B was to be sealed off for its long dormancy and then depressurized. The MOL pilots would have lived in the 400 ft3 (11.3 m3) pressurized laboratory module (ref. 5) for all but the first and last few hours of the flight. This plan for the isolation and prolonged decompression of a portion of a manned spacecraft during flight was and is unprecedented in the annals of human spaceflight, but reflects the knowledge, experience and concerns of the manned spaceflight community of the day, 1964.

Of course, the Gemini capsule was designed to be depressurized because space walking, or extravehicular activity, was a goal of the Gemini program. The capsule needed to be opened to the vacuum of space so an astronaut could exit it and perform useful tasks outside. But these were episodic, rare and brief events, not continuous for weeks at a time.

In other spaceflight experience, during Skylab missions, the unpowered Apollo command module was accessible continuously and was used for privacy by the astronauts because temporary ducts circulated fresh, dry air through the capsule. The Russian Soyuz was also accessible during Salyut and Mir missions—and remains so during International Space Station (ISS) missions. Ditto for the Chinese Shenzhou-9 while docked with Tiangong-1. Of course, the Space Shuttle was open to the Mir and ISS on all of its docking missions, and was fully powered the whole time, even drawing current from the ISS’s solar panels to extend its own orbital lifetime.

However, the depressurization is mentioned specifically in only one document (ref. 4), a report produced by Bellcom, Inc., a Washington-based contractor which analyzed Apollo program technical decisions for NASA Headquarters. Early in 1968, Bellcom’s R.K. McFarland met with representatives of McDonnell–Douglas to learn how the Gemini-B capsule was being modified for long-duration quiescence, in a mode comparable to that planned for the Apollo command-service modules during upcoming NASA space station missions. Among other things, McFarland was told that the Gemini-B would be depressurized to 0.1 psia, with the crew not re-entering the capsule until the end of the 30-day mission.

Independent, albeit indirect, evidence for the long-term isolation of the Gemini-B is found in a fire safety report (ref. 6) by The Aerospace Corporation, the systems engineering contractor for MOL, in mid 1967, before McFarland’s visit. This report documented the responses of the MOL Program to the recommendations of the review boards for the Apollo 1 fire and the Brooks AFB chamber fire (ref. 7), both of which occurred within a period of four days at the end of January 1967.

The report noted that, “[u]nder the original baseline, the Gemini B was [to be] repressurized with 100 percent oxygen prior to crew transfer from the Laboratory. In case of a fire emergency in the Laboratory requiring the crew to abort to the Gemini, this pure oxygen atmosphere would have made a hazardous situation worse. A capability is being added to permit repressurization of the Gemini B with a two-gas atmosphere from the laboratory atmosphere supply source to minimize this hazard. In addition Gemini B emergency repressurization time is being sharply reduced.”

This is a telling comment: a repressurization of the Gemini-B capsule would have been necessary only if it had been depressurized earlier in the mission.

Repressurization

As originally designed, the Gemini-B’s life support system would have required remote control—either from Earth or from the MOL—to repressurize its cabin by releasing pure oxygen from on-board storage tanks. The added capability mentioned in the Aerospace report might have been as simple as a valve in the hatch between the MOL and the Gemini-B which would allow MOL atmosphere to flow into the Gemini-B.

There is another indication that the Gemini-B would have been left unattended, if not actually sealed and depressurized, for the 30-day MOL mission. In a memo reporting the results of a visit by MOL engineers to the GE space simulation facility in Valley Forge, Pa., astronaut Scott Carpenter is quoted as saying he wouldn’t leave the Gemini unattended for 30 days without several checks on its condition (ref. 8). The fact that he was saying anything of the sort is evidence that someone, either GE or the Air Force, must have asked his opinion about leaving the Gemini unattended for 30 days. In a vehicle with only 400 cubic feet of volume, if the secluded the Gemini cabin plus the attraction of its two windows were available, it would undoubtedly have prompted frequent visits from off-duty MOL pilots, who could then almost casually have checked on its systems—unless they were physically blocked from such visits.

In the absence of any explicit contemporary reports to the contrary, the two contractor reports from Bellcom and The Aerospace Corporation establish that the Gemini-B was indeed to be depressurized while in quiescent in-flight storage. All of the other primary and secondary sources are silent on the pressurized state of the Gemini-B during the MOL mission. Judging from discussions on some authoritative listservs, this plan to isolate and decompress the Gemini-B capsule during the MOL missions is literally unknown by space history buffs today.

Why the Depressurization?

The available documents don’t mention the reason or reasons for the prolonged depressurization, but I can suggest a couple. First, as previously mentioned, the pilots would have been the source of substantial humidity. The unpowered Gemini-B capsule would have been a natural cold-trap for that humidity, and would probably have become quite damp quite quickly. This subsequently happened on Apollo 13, whose astronauts actually experience a bit of a rain shower as their damp command module entered the Earth’s atmosphere and the condensation dripped onto them from the control panel and other overhead structures.

Sealing off the Gemini-B would also have reduced the air leakage that was an inevitable feature of any manned spacecraft, especially one with two large crewmember hatches.

As I said, I have not seen any documentation of the reasons for the decompression, but these may have been contributing factors.

How the Depressurization?

But how would the Gemini-B capsule have been sealed off? There were hatches between the Gemini-B cabin and the MOL’s pressurized laboratory—surely they were used for that purpose. In my next post I will describe how it might have been done, or not.

References
  1. Calculated using the equations at http://www.ehow.com/how_7916900_convert-pressure-altitude.html (accessed 16 Dec. 2012).
  2. Cohen, M., “Testing the Celentano Curve: An Empirical Survey of Predictions for Human Spacecraft Pressurized Volume,” 38th International Conference on Environmental Systems, San Francisco, California, June 29-July 2, 2008, SAE TECHNICAL PAPER SERIES 2008-01-2027, http://www.astrotecture.com/Human_System_Integration_files/SAE-2008-01-2027%20(2).pdf (accessed 28 Aug. 2012).
  3.  __, “Gemini 5,” National Space Science Data Center, http://nssdc.gsfc.nasa.gov/nmc/spacecraftDisplay.do?id=1965-068A (accessed 16 Dec. 2012).
  4. McFarland, R.K., , Bellcom, Inc. “Subsystem Modification to Develop Quiescent Operation for Gemini B, Case 620”, February 28, 1968, http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19790073017_1979073017.pdf (accessed 6 Aug. 2012). McFarland briefed C.W. Matthews, Director, Apollo Applications Program, on Feb. 15.
  5. Wade, M., “MOL,” Encyclopedia Astronautica, http://www.astronautix.com/craft/mol.htm (accessed 28 Aug. 2012).
  6. __, “MOL safety evaluation based on Apollo 204 Review Board findings and recommendations and Brooks Air Force Base Accident Investigation Board Conclusions,” Aerospace Corporation, AD-856687L, 58p., September 1967.
  7. __, “Apollo-1 (204),” http://history.nasa.gov/Apollo204/ (accessed 17 Dec. 2012).
  8. Anderson, J.J., Col., AFRMO, “Memorandum for the Record, Trip Report to Valley Forge, GE, Simulation Tests,” 20 Aug. 1964, courtesy of Dr. Dwayne Day.

Sunday, December 9, 2012

A Jones for MOL #5: A MOL, not a chipmunk.

This is the only image I could find of something close to a
chipmunk in space,even though it would probably have
been breathing pure oxygen in the A7L NASA-style spacesuit.
Credit: Avanti Press, 2011, avantipress.com.

The presence of helium in the gas mixture of the Manned Orbiting Laboratory’s internal atmosphere has enabled four decades of amused speculation that the two MOL pilots would have sounded like Alvin, Simon and Theodore when they spoke to each other or with their ground controllers.

Substituting helium for nitrogen in breathing gas mixtures for deep-sea diving (to help prevent oxygen toxicity and nitrogen narcosis) is well-known to cause speech distortion, especially under high atmospheric pressures (ref. 1), so it has been assumed the same would be true onboard MOL. The famous example of astronaut-turned-aquanaut Scott Carpenter’s telephone conversation with President Lyndon Johnson, in which the president betrayed no hint of any struggle to understand the explorer’s squeaky voice, was the result of Carpenter’s presence in a decompression chamber at 89 psi of heliox at 20% oxygen and 80% helium, after his 30-day sojourn at a depth of 200 feet (61 m) in Sealab II (refs. 2,3) (I was surprised to find out that the high pressure alone would not have increased the frequency; ref. 4).

The U.S. Air Force investigated voice quality and other aspects of candidate MOL atmospheres before settling on the final gas mixture of 70% oxygen and 30% helium at 5 pounds per square inch (psi). A complex of connected altitude chambers at Brooks AFB in San Antonio, Texas, for prolonged habitation was laid out like the MOL with an adjacent area to accommodate “ground control” (ref. 5). A smaller “space cabin” at Wright-Patterson AFB in Dayton, Ohio, was used for brief studies (ref. 6). Subjects for these studies, at least at Brooks, were “informed volunteers” drawn from the available population of young, fit basic trainees at nearby Lackland AFB (ref. 7).

My last blog described the Air Force’s process of selecting helium as the diluent for MOL’s atmosphere. Helium’s relative inertness, unusual heat coefficient and reduced solubility in body water and fat all recommended it as a substitute for atmospheric nitrogen even at reduced cabin pressures (ref. 1). Voice quality characteristics of interest included distortion, attenuation and error rate between airmen within the cabin and by external monitors using headsets (ref. 1). An Air Force study (ref. 1) of a 50% oxygen, 50% helium mix at 7 psi (slightly less than ½ atm) for up to 12 hours found no greater incidence of communication problems than in pure oxygen at 3.5 psi or 5 psi (as would be used in US EVA suits and US spacecraft, resp., through the end of Apollo). Fatigue of the test subjects, who were awakened early and spent a long day in the chamber, was judged to cause the same degree of problems. Another study (ref. 8) evaluated almost the same atmosphere, 44% oxygen and 56% helium at 7.8 psi (½ atm), in 4 subjects but for 14 days this time. On average, the fundamental and second formant frequencies of the voice, which would largely determine voice pitch, were higher in helium than in air, but the fundamental decreased toward its pre-exposure value over the two-week confinement. In fact, the fundamental frequency in one subject dropped and remained at or below baseline for most of the study. A third study (ref. 6) of bioacoustics during 5-8 hours at 30%, 50% and 70% helium atmospheres at 258, 360 and 760 mm. Hg (0.34, 0.5 and 1 atm) found over 50% increase in voice formant frequency at both 30% and 50% helium, and 90% increase at 70% helium, consistent with other studies.

Those studies dealt mostly with higher helium fractions and shorter exposures than a MOL mission would entail. A more realistic Air Force study (ref. 9) confined 4 subjects for 56 days at the selected MOL atmosphere. Only 1 subject showed a marked rise in the fundamental frequency of his voice over the 2-month confinement, although all 4 exhibited increases in their second formant frequencies.

Soviet investigators had also investigated an unspecified helium atmosphere, even before the Air Force tests, in which the voice pitch of the test subjects was reported to be increased “seven tenths of an octave, but the distortion was tolerable” (ref. 10). I spent a little time Googling articles about voice quality to understand what “seven-tenths of an octave” corresponded to in terms of “cycles per second” but just came away confused.

The Soviet study aside, at least the Air Force studies were consistent in their conclusions that, even if the voice frequency were increased, intelligibility was more influenced by background noise, and crewmembers learned to modify their speaking styles to minimize the effects of the helium, the noise and the lower cabin pressure.

What’s more, the frequencies in the MOL pilots’ voices would be increased only 9% compared to air, much less than that observed in the divers (see table below) (ref. 11). The MOL pilots would have been breathing almost exactly the opposite ratio of the gases used by deep sea divers whose voice quality derives from the higher proportion of helium with its low gas density and high velocity of sound.

So, no chipmunks or Donald Duck on MOL.

The pitch of the MOL pilots' voices would have been determined by the
relative frequency of sound in the cabin's 70% oxygen, 30% helium atmosphere,
about 9% higher than in normal air. This is less of an increase than
under cabin atmospheres that were considered, and much less than
deep-sea divers breathing heliox.

References
  1. Cooke, J.P., and S.E. Beard, “Verbal Communication Intelligibility in Oxygen-Helium, and Other Breathing Gas Mixtures, at Low Atmospheric Pressures,” Aerospace Medicine, 36(12): 1167-1172, Dec. 1965.
  2. “LBJ & the Helium Filled Astronaut,” Produced by Larry Massett, http://www.npr.org/programs/lnfsound/stories/991015.stories.html (accessed 22 Mar 2006). The NPR website incorrectly listed the date as 1964, but it was September 27, 1965. See also Carpenter, S., and K. Stoever, For Spacious Skies (Orlando: Harcourt, Inc., 2002), p. 321.
  3. Speed of sound formulas, http://www.engineeringtoolbox.com/speed-sound-d_82.html (accessed 14 Aug. 2012).
  4. “Speed of sound - temperature matters, not air pressure”, http://www.sengpielaudio.com/SpeedOfSoundPressure.pdf (accessed 21 Aug. 2012)
  5. Nunneley, S.A., and J.T. Webb, “Aerospace Medicine at Brooks AFB, TX: Hail and Farewell,” Aviation, Space and Environmental Medicine 82(5, sec. 1): 567-570, May 2011.
  6. Nixon, C.W., and H.C. Sommer, “Subjective Analysis of Speech in Helium Environments,” Aerospace Medicine 39(2):  139-144, Feb. 1968.
  7. Barry, S.J., and J.E. Endicott, “Comparison of Speech Materials Recorded in Room Air at Ground Level and in a Helium-Oxygen Mixture at a Simulated Altitude of 18,000 Feet,” Aerospace Medicine 40(4): 368-371, Apr. 1969.
  8. Nixon, C.W., W.E. Mabson, F. Trimboli, J.E. Endicott and B.E. Welch, “Observations on Man in an Oxygen-helium Environment at 380 mm. Hg Total Pressure: IV. Communications,” Aerospace Medicine, 39(1): 1-9, Jan. 1968.  
  9. Nixon, C.W., W.E. Mabson, F. Trimboli and B.E. Welch, “Study of Man During a 56-Day Exposure to an Oxygen-helium Atmosphere at 258 mm. Hg Total Pressure: XIV: Communications, Aerospace Medicine 40(2): 113-123, Feb. 1969
  10.  __, “Helium Test,” Aviation Week & Space Technology, Nov. 23, 1964, p. 25.
  11. This conclusion is drawn from calculations based on information in two sources: Hess, D.R., J.B. Fink, S.T. Venkataraman, I.K. Kim, T.R. Myers and B.D. Tano, “The History and Physics of Heliox”, Respiratory Care, 51(6): 608-612, 2006, http://www.rcjournal.com/contents/06.06/06.06.0608.pdf (accessed 7 Aug. 2012); and Speed of Sound in a Gas http://www.electronicsteacher.com/succeed-in-physical-science/sound/speed-of-sound-in-a-gas.php (accessed 7 Aug. 2012).