Wednesday, July 8, 2020

A Technological Appendix to Postblogging Technology, March 1950: We Need To Talk About Martin

Martin B-26 Marauder: The "Baltimore Widomaker"

On 29 August 1948, Northwest Airlines Flight 421, a Martin 2-0-2 crashed near Winona, Minnesota, killing all 37 aboard. The plane went down on the slopes of a steep gully (I didn't know they had those in Minnesota!) and body recovery was extremely traumatic to would-be rescuers. 

When an observant captain asked that another Northwest Airline 2-0-2 with similar flying hours be inspected for handling issues that same day, attention was drawn to a stress failure at a wing skin-spar joint, and the cause of loss in the Winona accident was determined to be a wing falling off due to stress failure. 

The appellate judge does a good job of summarising the technical details in the 1955 judgement that let Martin off the hook for  negligence. (Northwest oversaw production in detail, so no one principal was more negligent than the other.) Martin adopted an high-strength aluminum-magnesium alloys that was relatively new to the aviation industry due to its high notch sensitivity. It had been used with great success in the Mitsubishi A6M Zero, and was a definite late-war fad, although so were the stress-corrosion related groundings of the late Forties, and considering the way the Zero was used in service it's not hard to believe that stress corrosion problems were overlooked. 

Call me lazy or call me a victim of industry capture of aviation journalism, but I was surprised to read, last month about the 7 March 1950 loss of Northwest Orient Airlines Flight 307, another Martin 2-0-2, while trying to land at Minneapolis. Although the crash was determined to be the result of the pilot losing ground reference while landing in a snowstorm, the actual cause of the crash was a seventy-foot flagpole, causing a section of the left wing to detach. Obviously no-one is asking that an aircraft wing survive a collision like this, but the 2-0-2s had been grounded, at great expense to the airline while the problem was being fixed, and the suspicion has to arise that the fix had been less than completely successful. 

We can also, of course, reasonably ask why there was only a pilot and co-pilot in the cockpit at the time. CAA regulations did allow for this, and it was a pretty straightforward implication of the idea of a "DC-3 replacement" that the Martin 2-0-2 be operable by two crew. Tens of thousands of DC-3s had landed safely in similar conditions across Canada and the United States, but it has to be said that airlines picked approach routes that didn't have 70ft flagpoles on them. The Minneapolis approach route didn't have one, either, but the plane was off course, if only by a little more than a mile; which would seem to be at least as important a factor in the crash as the fact that it was too low. 

So much for one accident involving a half-full plane. It  turns out that a Northwest Orient training flight crashed near Almelund, Minnesota, on 13 October 1950, with six casualties; that Northwest Orient Flight 115 crashed near Butte, Montana, on 7 November 1950, with 21 fatalities; that another Flight 115 2-0-2 crashed near Reardon, Washington, on 16 January 1951 after unexplained loss of control in flight. This last had only ten people on board, which suggests that the news was getting out to the travelling public, although that didn't stop a strange Japan Airlines/Northwest Flight 301 crash in Japan with 37 fatalities. Rearden was the last straw for Northwest, which sold its remaining 2-0-2 fleet to Allegheny Airlines. Since Allegheny (and Transworld) operated the 2-0-2 happily for another ten years, either the problems really were ironed out, or it was all just a terrible series of coincidences. 

The problem is that in 1950, Allegheny's fine record was in the future, and the Glenn L. Martin Aviation Company looks really, really bad. While it's not the only company saddled with a dangerous airliner in this era, but the stunning series of 2-0-2 accidents is hardly going to help a company that operated, after all, in what was practically a suburb of Washington, DC. People are going to talk!

It's also not the the end of the story from March of 1950. We've still got an (undersourced) internal, blistering report on the Matador/Mace program, already prefigured in Aviation Week's recent lament that no-one could possibly expect industry prime contractors to master all the science involved in a guided missile. And although that report is still several years away, it turns out that the guided missile that's setting records of reliability and usefulness down at White Sands according to various anonymous sources is the Martin Matador.
The Matador belongs to the generation of missiles that the American aviation industry is just beginning to realise are going a bit wrong due to lack of proper managerial control. In the light of these rumblings, which will eventually lead to  a major reform of missile management,
it's probably not a bad idea for the company to leak encouraging reports about the weapon system's promising early flights, because they are promising. 

The Matador was the first "operational surface-to-surface cruise missile designed and built by the United States." As a young adult of the 1980s (remember when video stores with adult sections was a serious matter for left activism?), I'm just a little surprised to learn that there was an earlier generation of cruise missiles, decades before Litton's Tomahawk. However, the Matador, and its Mace counterpart, was just a first generation successor to the V-1, and even used some technologies originally developed for post-WWI "aerial torpedoes." It was a bit improved over the buzz bomb. It used a version of the Whittle/GE/Allison centrifugal turbojet rather than the V-1's pulse jet, RATO bottles rather than the V-1's goofy liquid-explosive boost propellant, had explosive cord laid in the wing seam to make sure that it entered its "terminal drive" with the right (straight down) trajectory, and had some modest control arrangements running off a Sperry autopilot.

These were intended to solve the problems the V-1 and especially the V-2 had experienced with airframe disintegration at supersonic speeds, and led to major revisions in the Matador airframe. I suppose that's a good thing when compared with V-1's "stop and tip over when the autopilot says you're there." The problem is that at this early stage of development, there wasn't really a weapon for the Martinet. The programme was associated with the Army and Tactical Air Command, and the Air Force brass were shitting bricks over the very suggestion that they might get a share of SAC's atom bombs. (And given that we are talking about atom bombs, this isn't necessarily unreasonable.) The upshot is that flight tests were going ahead on the basis that the Matador would be used to dispense chemical and biological weapons; the case can be made that loads of nerve gas anthrax spores shouldn't just be disintegating in mid-air, even if dispersal at supersonic speeds at 6000ft isn't necessarily conducive to spore survival and molecular structure. 

Before pursuing the technological history aspects of this month's news that I want to bring out, I'm going to stop and root around in the "they were crazy back then" files and pull this out.
While a picture from Eisenhower's second Inauguration parade, the picture above is scraped from a website selling a self-published book about the Mace and Martinet from which I have already quoted far too much in this post. Here's the link, if you're at all interested. you'll have to follow the link to get to the authors' writeup of the Teracruzer, which was quite something in its own right. 

So, to make a long story shorter, for me, because I'm not inclined to research this aspect of Eisenhower's defence policy today, Tactical Air Command won its fight, and was awarded atomic bombs all of its own, in large part because the alternative would have been giving the Army some. (Which eventually happened anyway, of course, but you can only fight the battles of the day.) Familiar as I am with the Navy's frenetic efforts to get hold of some of them new-fangled A-bombs to stay relevant, of which a bit more below, it had honestly never even occurred to me that tactical operations got "tactical" atomic bombs before tactical atomic bombs actually existed. For while we've all heard the tales of the  Davy Crockett atomic recoilless rifle and atomic demolition munitions and such, those were the creations of a slightly later generation of mad scientists at Los Alamos and (especially) Livermore. The Matador/Mace generation carried an early dial-a-yield W38 that could not be reduced in yield below 70kt, although on the other hand they could get up to3 megaton bombs, which is definitely getting to the point where you're not going to be missing your "tactical" target, or any cities that might happen to be near it. 

To be fair, the Matador was reserved to the lower figure, as it  only carried the Mark 5, a pure fission weapon at the low end of the spectrum. Still, at almost six times the size of the Hiroshima bomb, 70kt is a big boom. It will therefore be some consolation that it had permissive command guidance. Which is to say that it was controlled from the ground by an early tactical radio control system involving a radar and an analogue computer that could keep the controlled aircraft on course. Unfortunately, hand-off to another control station was never reliable, and the Matador was limited in its lifetime to radius of four hundred miles from a mobile control cab; and, practically, that cab was limited by its control radius to the hardstand needed to calibrate the Matador's gyroscope before launching. Because presumably this cutting edge Atomic Age weapon had to be physically swung by its tail around a compass rose so that the gyros would know where they were. And maybe even where they were  going!

This was all not terribly satisfactory, and someone (Tactical Air Command) had a solution, which was its own proprietary version of DECCA/LORAN, Shanicle. Wikipedia is extremely terse on the Air Force "Short Range Navigation Vehicle. Apparently it worked flawlessly on the Matador, and then was removed within two years, never to be heard from again. This might be because a "radar ground map matching" navigational aid was on the way. This would allow the Matador to "hug the ground" and avoid the kind of interception that was pretty much a foregone conclusion with a missile doodling along at 6000ft on a straight course to whatever tactical target might be in its range. This ATRANs system is what differentiates the Matador from the Mace, and in the Air Force's immortal phrasing, made the Mace a "fire and forget" weapon. 

Because I don't think anyone wants to "forget" where a 70kt bomb might be. 
(Because the music of your twenties is the best music.)

Meanwhile some more, the inglorious failure of one flying boat strike aircraft after another from at least the Coronado through the Tradewind had failed to persuade the Navy that the flying boat had no share of the future of the atomic striking force. The official specification for a Mach 0.9 low altitude flying boat penetration bomber would not be  issued for another 13 months as of March of 1950, but the "Advanced Striking Force" was already percolating through the Navy Department.

There's a lot to be said about the Martin P6M Seamaster, and if you have the time, this fourteen minute video will bring you as far up to date with it as I am with one minor and not-at-all-tangential exception, the Goodyear ASQ-29 Automatic Navigation and Minelaying System, which was essentially computerised dead-reckoning. (But it sounds a lot more futuristic if it is an analogue computer integrating the inputs from gyros and accelerometers than if it is a navigator working on a chart board.) 

Like Shanicle, ASQ-29 seems to have disappeared pretty quickly once the Navy decided not to go ahead with the P6M, in favour of Polaris, as well it might. At this point I think we join the received "inventing accuracy" story already in progress. 

With all due respect to Dr. Mackenzie's classic work, it seems to me that starting with a "loss of visual reference accident" that killed two preteens makes the pursuit of accurate aerial navigation seem a bit less like a complicated sociological interplay of institutional interests and invented concerns; and more like something that we really should be doing if we want to fly commercially while still not killing preteens. But what do I know? Considering the spirit of black humour that seems to have informed the whole Matador/Mace programme, my cynicism may yet be restored. 


No comments:

Post a Comment