Flying Colours: Recapping Research and Development at Mid-Century

Alvin Hansen famously said that one of the reasons that secular stagnation was happening was that technological innovation had stopped.

In 1938. He said that. It seems a little . . . odd. Although no odder than attempting to make the argument that technological change is coming to a halt right now. Don't we live in an era of ever-accelerating technical change, catapulting down the runway?

So that's an experimental gunpowder catapult launch of a Vickers Virginia in 1931.

In case you didn't get a good enough look at it in the video, here's another view of a Vickers Virginia.

High tech, circa 1924! It's got a metal frame, two Napier Lion engines, and can reach 5000ft in only 10 minutes. Technically, the service ceiling is 13,800ft, but it's not like you can expect to be able to bomb German steel plants from such science-fictiony altitudes. I mean, what about low cloud cover? You wouldn't be able to see them! And forget navigating when you can't check your astronomical fixes against ground features and calibrate air to ground speed with drift sights.

Replacing the Vickers Vimy in 1924, the Wikipedia article says that some of the 124 Virginias produced in the 1920s were still in "front line service" in 1938. It appears to  have been predeceased, if anything, by the 97 bomber-freighter-transport variants of Victorias and Valentias. It also says that the Virginia was "somewhat accident-prone, 81 being lost in service." 

You say "accident-prone," I say "widow-maker." Not the plane I'd choose for experiments in catapult-assisted takeoffs. That being said, the job of a "heavy" bomber is to carry more bombs, further, and one of the best ways of doing that is to increase the all-up-weight so that it can carry more fuel and bombs. Catapults are a means to that end. Below the fold, I'm going to talk about the goofy, goofy ways, that people tried to do that in the 1930s so that there could be commercial air service across the Atlantic, and reach for a comparison with the frustratingly long and sorry story of the introduction of colour television. 

It's a comparison that, I think,  has some relevance to the problem of technological innovation and seculary stagnation. Specifically, I am going to compare and contrast a couple of relevant developments in an attempt to show that what's important is not "research and development" taken in isolation, but the broad shoulders of the state, carrying the economy forward, where and when it is willing to be the investor of last resort. Once the research and development problems that could be solved without the state, were solved, all that was left was . . Well. All that's left is everything that still needs to be done in 1939. 

And that's why Alvin Hansen was actually right!

After the jump, some kind of hand gesture in the direction of trying to make the point.

  First, this.

Comparative Research and Development (thousands dollars/ pounds)

Year	NACA	US Navy	US Army	Air Estimates “Research and Development”
1921	200	2,935	5,523	1,835
1922	200	3,705	5,261	1,673
1923	226	1,117	3,2161	1,469
1924	307	1,600	2,888	1,469
1925	470	1,573	2,984	1,469
1926	534	1,550	2,683	1.469
1927	513	1,928	2,208	1,469
1928	550	1,729	2,189	1,469
1929	629	2,000	2,250	1,560
1930	1,508*	2000	2,239	1,576
1931	1,321*	2000	2,265	1,717
1932	1,051*	2,295	2,286	1,501
1933	915	2,210	3,171	1,339
1934	957*	2,189	3,009	1,428
1935	1,245*	1,773	3,487	1,444
1936	1,178	2,498	5.724	1,601
1937	2,545*	2,500	4,506	2,836
1938	1,734	3,500	4,339	3,133
1939	3,869	2,904	3,484	5,024

The source for the American data, as you may have noticed, is the February 1946 number of Aviation. The  British numbers are direct out of the Air Estimates, which include a bottom-line figure for total research and development. Unlike the American numbers, the British one does not include construction. (The asterisked figures for the National Advisory Committee for Aeronautics are for years in which construction is more than a third of the "research budget." Aviation has obligingly broken this out for NACA, but not the Army and Navy.) I would not go very far out on a limb on the strength of these numbers, but they give a sense of scale, and serve the valuable purpose of demonstrating that the British were not neglecting aviation research between the wars. 

One also gets the impression that by 1939 they were just throwing money at anyone who could spell "scientist."

Which probably explains this. By Source, Fair use, https://en.wikipedia.org/w/index.php?curid=20854211

And this. Seriously. Way to bhusband the taxpayers' dollar, guys. Fair use, https://en.wikipedia.org/w/index.php?curid=37099839

Just to put this in perspective, in 1929, when the Air Ministry spent 1,576,000 pounds on research and development, the lowest monthly GDP figure estimated by this nice-'n-mathy paper, is 380,420,000 pounds. It's not that research and development were unimportant to the Air Ministry. On the contrary, it's very proud of what it is doing at the RAE and supporting establishments. That's why it's waving these numbers at us. The other two services do not break the numbers out at all --so much for getting a sense of what the Navy's Engineeering Branch was doing-- and I'm not entirely sure where to go in the Estimates to find the numbers for the NPL

It's just. . . Well, okay. I started out with the Vickers Vimy because it is not only the RAF's main immediately-post-WWI bomber, but also the first plane to fly across the Atlantic. In June of 1919, RAF ace airmen John Alcock and Arthur Brown got into a "specially modified" (that is, it  had ferry tanks in the saddle between the two cockpits) Vickers Vimy, and, after just barely getting off the ground in St. John's, landed in an Irish bog in Galway, sixteen hours later. 

This was, of course, in no way a precursor to commercial Atlantic flying. But consider: The Vimy was not only a very-hastily designed aircraft. Its engine was also a ridiculously under-engineered machine. Work began on the Rolls-Royce Eagle less than three years before this flight! Its successor, the Kestrel, by contrast, was gestating for virtually the whole of the 1920s, finally entering service in 1927. The Virginia used the Napier Lion rather than the Kestrel, which was not much employed as a bomber power plant, but the same point may be made here. The Lion was a considerable advance on the Eagle, and used in a considerably better plane. (I won't repeat the figures of merit here. This is going to be a link fest, anyway.)

The story of commercial Atlantic flying starts with survey flights by Short and Sikorsky flying boats in 1937. For various reasons, there was a lull in 1938, and commercial service finally began in the summer of 1939. Practically intended as a summer-only-service, it was scheduled as three flights a week, either way, by very large flying boats labouring to carry 16 passengers that might well be in the air for as many as 22 hours. 

If this sounds a little, uhm, boutiquey, consider the alternatives that were under development in 1937:

--A diesel-powered German flying boat that was going to be sea-launched by catapult ships. (The Admiralty had a brief dizzy spell over the idea of German commerce raiders operating flying-boat bombers this way.

Seems legit.

--A "composite" aircraft, in which the actual trans-Atlantic stage would be lifted into the air by a giant flying boat.

By Source, Fair use, https://en.wikipedia.org/w/index.php?curid=17126434

--A refuelled-in-mid-air service, in which the actual trans-Atlantic stage would take off, circle the airport while being fuelled up in mid-air, and then proceed. 

There were also a bevy of gigantic flying boats, but that's more or less a chronic disease of mid-century aviation, and I would be astonished if the Latecoere and Blohm-und-Voss six-engines were seriously motivated by the idea that they would be useful in trans-Atlantic operations. 

What all of these have in common, of course, is that the designers are clearly facing what seems to them to be an insuperable power deficit. Otherwise, they wouldn't be doing all of these goofy things to get into the air. They just can't seem to muster up the surplus of engine power to aircraft weight to get across the Atlantic, economically. 

Then, of course, World War II happens, and it suddenly becomes easy. What the Hell?

It's worth going back to the engine story, although things are a tiny bit complicated by the fact that most of the innovators of the interwar period were aging men by the time World War II came around, and, to perhaps summarise too breezily, they all died in the 1950s, when it was impossible to break censorship and give a clear account of who was responsible for what. 

At least, that'll do as an excuse for the "story" of the Kestrel, which, if you've followed the links, is apparently that Arthur Rowledge designed the Napier Lion for Napier, got tired of taking flack from Napier management, and took his giant, design-y brain off to Rolls-Royce to dash off the Kestrel for them. Then he died in 1955, leaving no memoirs, and any papers that might shed some light on the story locked up in the PRO, and if, reading this, you're getting the sense that it's a journalist's account, it is. Thank Heavens for Bill Gunston, but let's not mistake him for a real historian.

Arthur Rowledge. By Source, Fair use, https://en.wikipedia.org/w/index.php?curid=24775447

Supposedly, all innovation stopped at Napier the moment Rowledge walked out the door. It's not that the Lion was an absolutely dominant engine, and that anything else the company built would just cannibalise its sales, it's that Napier just wasn't into that whole "research and development" thing. Until, of course, it was. Leaving the Sabre aside, here are links to the Rapier, Dagger, and Culverin pages.  All very interesting, although of dubious relevance. Well, the Culverin is relevant, in that it was a chosen Atlantic engine --but that's because it was just a British licensed version of the uhm, innovative German Jumo 205 two-stroke, double-piston, double crankshaft engine that powered the Ha/BV138. (Then, because that wasn't goofy enough, went on to be the active component of the Nomad and the Deltic. And, for good measure, served as a truck engine.)

Only it wasn't really irrelevant. Napier returned to engine design in the mid-30s because it came to a brain-sharing arrangement with fellow London-areas aeroengine manufacturer, de Havilland. Both firms employed the independent consulting engineer, Frank Halford to do the detailed work of design, and like his colleague and rival, Harry Ricardo, Halford was part of a tightly-linked British engine design community which could, very precisely, be led in the direction that the RAE wanted to go, engine-wise. The Royal Aeronautical Establishment was very interested in building an engine which could carry a fighter up to interception heights in very short times. In the mid-30s, this seemed as though it would require a small-bore, multi-cylinder, high-speed engine with sleeve valves to suppress the "hot spot" a conventional poppet valve would produce. This is how the Dagger and Rapier came to be. 

Or, it could mean a two-stroke diesel. I'm not sure why: it strikes me that the main advantages of the diesel are thrown away when the engine has to accelerate rapidly. But Harry Ricardo spent his entire career working on an automoible diesel engine, and it took actual combat experience to disabuse the Luftwaffe of the idea that a diesel was a viable bomber engine.  As the recent Volkswagen fiasco shows, there are ideas, including the "high performance automotive diesel," which are like zombies. They just won't stay dead.

Why all these misteps? The answer, as I've said before around here, is that ultimately, gasoline piston engines could potentially achieve performances unimaginable to all ther cheaters and fakers of 1938 --provided that they had access to high-enough octane fuel. In fact, higher octane fuels (and here were are talking about progress from 50 octane to 65) stands behind the Alcock-and-Brown achievement. Without rehearsing the numbers, it should be clear that flying with 100/115 octane is a huge improvement on 87. But to get there, the aviation industry had to piggyback on a new kind of refinery, and dedicated storage, so that lower octane gasolines did not get mixed in, and the TEL did not come out of solution. 

With 87 octane, Atlantic flying is basically impossible. Invest in a new infrastructure, and it becomes easy. In retrospect, what looks like technological progress, the bizarre explosion of innovation of the Atlantic contenders of 1937, are attempts to work around the fact that the money for this new infrastructure isn't coming.

I know that this is what I'm always criticising Flight's military correspondent for --writing about something that is actually happening in 1946 entirely by reference to something that happened eight years before, but I am going to stand on the defence that, in this case, "research and development" is a fig leaf for a failure of state-led spending on a scale sufficient to actually carry commercial trans-Atlantic flying forward on the broad back of the armed forces. 

Anyway, colour television. Colour television is, if you are an old person like me, something that you are vaguely aware came in during your childhood. As the Wikipedia says, the first all-electronic television broadcasting equipment came in during 1953, the first broadcast was on 1 January 1954, and you can still find black-and-white shows on cable TV (well worth the $170/month I'm paying!) today. This might seem like the story of progress if it weren't also the story of scandal. Imminent colour television was promised in this month's Postblog. That is, in February of 1946. This is Peter Goldmark, hawking a version of the "field-sequential" mechanical technology through the Columbia Broacasting System. As usual in an American publication, he is treated as the "inventor" of the technology, which, we are told, is imminent.  

As it happens, of course, it was not. It might have had its day, were it not for the fact that sets had to have additional equipment to pick up colour broadcasts. All-electronic colour tvs, in contrast, could pick up black-and-white easily enough, so having a colour set didn't get in the way of watching the Honeymooners, and served perfectly well as a "positional good." That is, you could impress the neighbours with the fact that you could afford a television that was three times as expensive as theirs, even if the only thing you could pick up was a parade in Pasadena. 

The weird thing about this is that at the same time that field-sequential television was being shot down over the problem of additional equipment, that weird, pointless UHF dial was being added to American televisions as standard equipment.  Why? Because of a regulatory mandate by the FCC, which stepped in to make sure that every American was equipped to watch UHF, just in case the laws of physics changed and made it practical. (Okay, I overstate things a bit; but there's a reason for this.)

What's the difference between early colour television contraptions and the UHF dial? By putting a UHF dial on American televisions, the FCC ensured that there could be UHF television stations. AKA "licenses to print money." Hey, did you know that Herbert Hoover's son's political career was scuttled by a scandal over the allocation of radio broadcast licenses? Reference to it is exceedingly oblique in the Wikipedia article, but I'm going to let it serve, because it's late. 

UHF was never really a thing. Well, maybe it was a huge thing in New York City, but for most of us, it was never a thing, but we all paid for it, anyway, because people who mattered wanted broadcasting licenses. Colour television, on the other hand, came in slowly and in its own good time --a good thing I don't have to recap it, because it hasn't happened yet! However, an acceptable colour television technology died on the vine in 1946 for want of a willingness to spend a fairly small additional amount nof money.