When writing the post on the V-2/A4, I fell down the rabbit hole learning how the A4's guidance system worked. I couldn't really work it into that post, as it was already way too long, so I present those details here, both because the details I learned are pretty cool, and also to back up a claim I made: that the guidance system was a hard limit on the Third Reich's missile ambitions. While in theory later missiles could have been developed with the A4's motor, the guidance system would have to be completely redone, an development project that it's difficult to imagine the historical Third Reich ever finishing.
The Obvious, Stated
I know quoting the dictionary is a hoary old cliche, so let me just say my research made me reflect on the phrase "guidance system". Before I started, I assumed a kind of black box that did the guidance thing. But once I got into it, I realized that "system" implies a bunch of mechanisms doing guidance: working, but not necessarily unified. This is the first thing about the A4 guidance: it was a whole series of disparate systems working to bring the A4 to a meaningful definition of guided.
The next word I ended up contemplating was "analog." The A4 development saw one of the early digital computers created, which Peenemunde engineers used to confirm trajectory calculations, but all systems on the A4 would be run by analog electronics. Digital electronics do everything with binary signals, the on/off function acting as a zero and one. Analog
control systems are an entirely different kingdom, the Fungus to more familiar plants, and like fungi, analog covers an array of stuff that can get astonishingly weird.
After that, I found myself going to a high school physics textbook to learn about inertia. Inertia is an object's resistance to
changes in velocity, and is a supremely useful force in the analog era,
as it can be measured quite precisely - usually with a machine called a gyroscope which we will get to.
Still, it takes a slightly different mindset to get into analog, as something trivially simple in digital (like the holding of information) becomes its own custom problem in analog. Because of the bespoke nature of problems in analog, systems become very specific to application. There's one more pre I'd like to amble out: in the last post, I mentioned it took 100 men and about thirty vehicles to set up an A4 for launch. Despite this, engineers prelaunch only had to give two variables to the missile: an azimuth bearing, and a precalculated precise time for engine shutoff.
Back to the dictionary and the physics textbook one more time: 'ballistic missile' as it turns out tells you very well how the missile flies, and the cornerstone of its guidance. In fact, 'ballistics' (the study of projectiles) would be perfectly familiar to artillerymen of Napoleon's time.
Inertia D
Issac Newton discovered that anywhere there was gravity, projectiles would fly in a parabolic curve. A bullet shot from a gun completely level would fly as if it had been shot from its highest point in a parabola. [You can see this demonstrated in a Mythbusters episode, where they demonstrate that a bullet shot from a gun and a bullet dropped from the same height of the gun barrel hit the ground at the same time.]
As Newton was pretty good with math, he soon worked out some simple equations for figuring out the parabola of a given projectile. These were seized upon by gunners and artillerymen the world over, who soon had collected a lot of empirical data on explody projectile flights. So the math about projectile weight, the energy applied to it, and its resulting curve, mathematical or real, was settled into the science of ballistics.
The A4's engineers were going to use ballistics as their guidance system. Instead of a single shove via explosive, the motive energy was being released over time, but the basics and the results were identical: a weight, a quantity of energy, and the resulting parabola. By timing the engine cutoff, the distance traveled could be modified, creating a longer or shorter trajectory as was needed.
Working out HOW to get that timing right was more involved. Initially in flight testing, the shutoff signal was broadcast by a device that used the Doppler shift of the rocket on radar to time the broadcast signal. This first timing system would be used in tests and quite a few operational launches, but when the A4 was being developed, it was always assumed to be at best a stop-gap system, as it would be vulnerable to enemy interference. The permanent solution was the creation of a sensor known as a PIGA [Pendulous Integrating Gyroscope Accelerometer.] This sensor measured the inertia of the A4's acceleration and would be set to engage the engine shutoff once a preset point was reached.
There's one more thing you need to make this system conceptually work. You need to know the precise distance between the launch point and the target, so you know how big a parabola to make. Fortunately, this was once again very simple. The A4 was road-mobile, and used concrete pads for launch. These pads were surveyed when constructed, which made it simple to calculate the distance between the launchpad and the target.
So, you have one-half of an aiming method for your 1940s ballistic missile. Now, you need the other half: the rocket needs to be able to hold an azimuth bearing, IE a direction on the compass. You also need (small note here) to make the rocket able to fly autonomously through its flight. To do these things in the 1930s and '40s, you need a gyroscope.
Gyroscopes are frequently used in Second World War era electro-mechanical magic: they are simple machines that measure angular momentum. What's more, because they spin round and round, gyroscopes can also be used to generate analog control signals.
That's a mouthful, so let me explain. The heart of a gyroscope is a rotor that spins at a set speed; like a top, it resists movement at a 90 degree axis to its spin as it possesses angular momentum. This resistance which means in a gyroscope precise measurement of that resistance is measured, as force in the measured axis either speeds up the revolution of the rotor, or slows it down; usually measured by the rotor making an A/C electrical wave as it spins. The A4 used two or three gyros to measure acceleration in the three axis of movement, and thus could correct itself in flight. The control gyroscopes of the A4 spun at 20,000 RPM, generating a control frequency of 333 Hz. Shifts in flight in the direction of spin would speed up the revolutions of the gyro, and thus boost the control frequency to for example 336 Hz; while against spin would slow down the control frequency, say 330 Hz. These analog signals are then used by the control surfaces to correct against the detected movement, which exerts the opposite force, thus bringing the signal back to 333 Hz.
Just to make things a little less clear for later researchers, the A4 had two different gyroscope control sets. One used two gyroscopes, with one measuring yawl and roll, and the other measuring pitch and tilt. The other used three gyroscopes; a gyro for each axis of movement. (In aviation these are pitch, roll, and yawl, but you could think of the axis as X, Y, and Z if that works better for you.)
To get back to where we started, an additional gyroscope was used to set the azimuth bearing. This used polar coordinates, which you may know as degrees of a compass, with North as zero or three hundred and sixty degrees, East as ninety degrees, etc. Once set, like the control gyros, the navigation gyro would keep the rocket on a given bearing, correcting via inertial shifts.
And we're still not done! All the control inputs passed through a device which the German called a mischgerät, a mixer. The mixer took signal inputs and if necessary modified them for additional variables.These variables included the shift in the rocket's center of gravity as it consumed fuel, shifts in air density, and the rotation of the rocket as it flew: the rocket for stability spun six degrees per second, for much the same reason a bullet, shell, or arrow is set to spinning. The mixer also presumably handled the shift between the two different sets of control surfaces, initially graphite vanes, then the control fins. Then the corrected control signals we sent along to the actual control mechanisms, which took these signals and translated them into actual motion.
So, what can we learn from all this?
First, understanding this stuff does a lot to explaining why A4 improvement was not really in the cards in the Third Reich Planning horizon, and why all Aggregat models post A4 were half-baked at best. The most difficult job of the A4 program was arguably the creation of this analog control system and then testing it till it worked. Essentially a clockwork computer, it was not built with easy modification in mind.
Still, the Germans did try. Since internal mods to the A4 were out, they attempted to use external systems to improve guidance, and settled on creating radio guidance beams for the rocket's ascent phase so the speed and engine cutoff variables could be refined.
The Third Reich contracted
electronics giant Lorentz to work on this, a move that made sense. In the early
1930s, Lorentz did something fairly amazing:
it developed a blind landing system. Since radar was not a thing, how
it worked was two guidance beams would be broadcast on either side of
the runway. The landing aircraft has a radio set receiving both signals.
When out of alignment it'd produce noise, but when in the approach
path, the two signals would create a continuous tone.
So a system very similar to this was built to improve V-2 guidance. Two
signals broadcast into the sky from trucks 15 or so kilometers
behind the launch site. A radio receiver on the V-2 to steer the
missile into the continuous signal. This Doppler radar control signal from tests also seems to
have adapted for this new method. The Germans calculated that in order to achieve an accuracy of 250m circular error probable at a distance of 250 km, "the speed at burn stop had to be 0.5%
exact." One thing I couldn't find: if this system managed to achieve this metric.
So this accuracy assist system appears to be completely done, ready for
production in the end of 1943. But fortunately, it took an entire year
for the equipment to be manufactured and issued to missile units. As it
happens, it was only the SS's missile unit, SS Abteilung 500 who was
issued the equipment. This unit would attempt the most credible tactical
attack with A4s during the war, when in early March 1945 it was ordered
to destroy Ludendorff Bridge, a bridge across the Rhine accidentally left
intact that advancing Allied Forces seized upon. The resulting bombardment only saw one missile land within a
mile of the bridge, but the US Army post-war reckoned it was an
impressive performance in what was after all a contrabassoon solo.
Despite the fact I'm kicking dirt on the notion the Nazis could have improved guidance, don't mistake that as a problem of analog systems generally. Inertial guidance systems would see extensive research and development post Second World War. This interview with a former Indian MiG-25 pilot has the interesting detail that the MiG-25 Foxbat had an inertial autopilot that could be programmed to fly an entire reconnaissance mission, minus takeoff and landing. Another fun example is Advanced Inertial Reference Sphere. The last purely inertial ICBM guidance system built in America, it gave an accuracy of tens of meters over thousands of kilometers, and looks like a scifi prop to boot.
There are now also solid-state electronic inertial tools like this ring laser gyroscope. Instead of a rotor, it uses light to measure the shifts of momentum along a given axis. I'd tell you more, but the last time I read it my brain exploded and I'm still finding little bits of it in odd corners when I sweep up.
On a Clear Day You Can See Peenemünde
There's one more thing I'd like to add. In the process of researching this stuff, I started reading Rockets and People, a four volume series of memoirs by Soviet rocket engineer Boris Chertok. Available for free via NASA, the books tell the story of the Soviet Rocket program. There are so many good stories in it that more stuff related to it will likely show up on this blog, but the story from the Soviets raiding the defeated Germany for rocket technology to the launching of Sputnik via the R-7 should be read by anyone who wants to know how much work is needed to build a functional space rocket/ICBM. The Soviet experience is especially relevant, as their missile program stuck with the Nazi propellants of ethanol/water and LOX through to their first ICBM, which didn't last long as an ICBM, but would go on to be the most successful space launch rocket in history.
That said, postwar Soviet research revealed some interesting things about the A4; namely how hard the design had been locked down. The A4 was from the middle of 1943 one of the Third Reich's top projects, along with jet turbines. The A4 was deployed in the fall of 1944, far too late to have a serious effect. And the A4 only made it to deployment that quickly with some serious flaws unaddressed.
The first of these the Soviet rocketeers discovered for themselves while still in Germany. Like the Allies, the Soviets had a program for collecting Nazi war technology, especially anything (or anyone) to do with the A4. Messing about with A4 engines on captured test stands, they discovered that the A4 design had considerable thrust reserves untapped. Once optimized, the A4's engine could make an extra 10 tons of thrust, an increase of 40%! This is especially notable as it would have doubled the A4's maximum range to 600 km. Actually making a missile that could use this extra potential was more complected, and would have meant extensive redesign of the A4, so much so that a new design would have been more practical.
There testing moved in 1947 to the Soviet's first test range at Kapustin Yar, about 800 km east of Moscow on the Russian Steppe. The Soviets had laid down a methodical testing program, both to test that they actually understood the technology and to test the first products of their infant rocket industry. The program started with firing actual A4s, captured and reassembled in the east, then moving on to Soviet copies of A4s - the R-1 - as a test of the Soviet design and building capacities, before moving on to the first indigenous design, the R-2, as a test of fixing issues identified in A4 and R-1 testing.
Two further flaws in the A4 were identified. The first was its longstanding 'random explosion' problem; the second was that the guidance system sometimes failed badly. The random explosion problem was that approximately 10% of A4s launched would explode prematurely on their downward trajectory. Soviet and German scientists soon identified the cause: Peenemunde engineers had underestimated the heat buildup on the front of the rocket. These 'exploding in a bad way' A4s were exposed to enough heat that their warheads were evaporating into gas, would quickly make a rupture somewhere, which would cause catastrophic failure. Adding more thermal isolation solved the problem.
The other problem was guidance-based. Test A4s would sometimes hit their X coordinate fine, but miss their Y coordinate to an almost ludicrous degree: sometimes 100 km off in a 300 km flight. This problem was handed to the German engineers on site. (The Soviets in their race for captured Third Reich technology had captured an almost complete missile train, that is, a rail mobile A4 launcher. In an extremely canny move, they completed the train and then ordered a second copy constructed. These two trains would be the field offices of the missile researchers in the 1940s.) The scientists, Dr. Kurt Magnus and Dr. Hans Hoch quickly found that on a test rig they could create noise on the control signal line with certain frequencies of vibration. Magnus and Hoch then whipped up a line filter between the gyro and the mixer from spare parts on hand which completely solved the problem. This greatly impressed the Soviets, who gave the two engineers a cash bonus and an entire jerry can of A4 rocket fuel, IE 75% ethanol. That's a lot even for all the Germans at the site, so it was shared with the Soviets that night in a particularly memorable party.
I bring all this up to underline how much had been ignored to bring the A4 to production, and even as it was it was a sort of freestanding miracle that it was deployed at all. If you follow the A4's development, even in its conception the A4 avoided ideas that Herman Oberath's book predicted, such as the rocket engine being mounted so it could vector its thrust, eliminating the need for early flight control vanes, detachable warheads for greater accuracy, or the use of the pressurized tanks of propellant as a structural component. All these ideas were put aside, rather reasonably, as technology that could be developed later.
The Soviet copy of the A4 would be called the R-1, and would enter Red Army service - entirely as an exercise in training Red Army units and as a manufacturing trial. The Soviets considered the R-1 useless as a weapon.
Part of the America Bombers Series
Part 1: Black Gay Hitler
Part 2: Vague Plans and Flying Boats
Part 3: Walking on Sunshine
Part 4: Stuffing arrogant mouths
Part 5: Eris is Goddess
Part 6: Ragnarocky Road
Part 7: Look Busy and Hope Americans Capture You
No comments:
Post a Comment