| |
Glenn’s Computer Museum |
| Centaur | Home | Old Military | Later Military | Analog Stuff | IBM Stuff | Components | Encryption | Misc | Index |
The museum is incomplete; it was last updated on 2/16/2010 (old military section).
This is a bombing diagram from a great reference book I have: F-4E Inertial Navigation Bombing System: Aircrew Handbook
More interesting are the cases where one or both of the source and target are moving, are at possibly different altitudes, possibly changing direction or altitude, and the calculation must be done in real time. Some obvious examples are dropping a bomb, shooting a gun from an airplane, and shooting a torpedo from a submarine. A closely related application is navigation: calculating the path to get from point A to point B. Another real-time computing application for this period is the air data computer, which calculates the calibrated air speed, mach number, altitude, and climb rate from an airplane's pitot-static system combined with other variables such as reference pressure and temperature.
Starting in the early 1900's, specialized electro-mechanical analog computers were developed to solve these computing problems. Analog computers were used because the primitive digital computers were not fast enough fast enough for real time applications (they were also too med and too flimsy enough for airplane and ship operations). The key computations needed for these applications are addition, multiplication, trigonometry, and numerical integration. These can be easily and quickly performed by mechanical or electrical circuitry used in an analog fashion. For example, the typical mechanical integrator (invented in the mid-1800's) consists of a rotating disc and a small wheel sitting on the disk and rotating with it. The rate of rotation of the wheel is proportional to the product of the disk rotation rate and the radius of the point of contact between the disk and wheel. If the radius y is changed as a function of the wheel rotation, x, then the accumulated rotation of the wheel at a given moment is proportional to the integral of y with respect to x.
The disadvantage of analog computers is, of course, lack of precision. This is not critical where the inputs are imprecise ( human controls, for example) and the outputs control imprecise equipment (hydraulic moving of a gun turret, for example).
I'm particularly interested in these electro-mechanical computing devices since they represent a lost computing technology. While today I work on very fast and complex micro-processors containing hundreds of millions of transistors, the design and manufacturing of these 1940's and 1950's devices seems unfathomable to me. I doubt that anyone could manufacture something like these devices, especially not in high volume.
A great paper on on the early history of mechanical analog computers for military applications is here.
|
One of the most famous devices of WWII is the Norden bombsight, often mentioned along with radar and the atomic bomb as the most umportant technologies in winning the war.
There is lots of good reference material on the web, including this great site
about for bombsights.
We are fortunate to have a complete Norden bombsight (with the optional X-1 reflex sight). A large part of the bombsight consists of two stabilization gyroscopes. The actual analog computing element (called the rate end) is the upper right portion. On the left is a diagram from the Bombardiers Information File (BIF), dated 1944 showing the major components. This very informative document is available from our site here (warning: this is a 50MB file). Below this diagram is is a closer view of the top portion. Using the knobs on the right side that connect into the rate end, the bombardier enters the altitude, airspeed, turn and drift rate of the plane, and the drag factor (“trail”) for the type of bomb to be dropped. Several other alignment controls must be set. The target is sighted through the telescope that looks down through the bombsight. The rate end unit performs continuous calculations (as the sight angle to the target changes or new inputs are made) of when to release the bomb and the course the plane should fly. Near the end of the “bomb run”, the plane’s autopilot is connected to the bombsight and it actually steers the plane. The bomb release can be controlled directly by the computer. Fortunately, we also have a separate (and never used) rate end (the computer component). On the right it is taken apart so you can see the analog computing mechanism. Note the disk and wheel integrator assembly: the wheel is on a splined shaft just below the middle of the left half and the disk is sticking vertically out of the right half. In addition to the bombsight, the BIF, and the extra rate end, we have the maintenance manual that which contains a wealth of detail on how the bombsight works. |
|
|
This World War II bombsight was originally
designed by the British (as the Mark XIV bombsight) and first entered
service in 1942. It was also to Sperry, who made improvements.
It was used in the B-25 and deHavilland Mosquito bombers.
Here is the front and the back with the cover (we have it) removed.
Note that there are a couple of components missing: the large hole and a component
is also missing in the upper left of the reverse side.
There is a separate sighting head (with attached gyro), which I do not have, that communicated to this computing element. The attachment between the two devices was mechanical (flexible cables). The computation was performed mechanically (with much done pneumatically). A set of replaceable cams were provided which adjusted the unit for each type of plane. Here is an illustration of the complete system including the sighting head. |
|
|
It turns out that the famous Norden bombsight was
not only U.S. bombsight of World War II.
The first bombsight accepted by the Government was the Sperry S-1 that was developed in
the 1930s.
It was designated as "standard" equipment in March 1941 and was
used in some U.S. Army Air Forces bombers early in World War II;
however, all contracts for production of the Sperry sight were ordered
canceled in late 1943.
Many claim that the Sperry S-1 was a superior
design to the Norden and clearly politics were involved in the
procurement decisions.
The September 1999 issue of IEEE Spectrum has a very good article on this topic, appropriately called The Bombsight war: Norden vs. Sperry. As is obvious, our S-1 arrived in pieces and I have gradually been reassembling it (as is also obvious, progress of this project is slow). Here is a picture of a completed nit at the Air Force museum |
|
|
The British-designed RAF Mark IX bomb sight was first introduced in 1939.
It was used on Canadian and Great Britain planes in World War II:
in particular Lancaster, Wellington and Sterling bombers, and Mosquitos, Beauforts and Beaufighters fighters.
This was an early preset vector bombsight that involved squinting through wires that had to be manually set based on aircraft speed, altitude and bombload.
The sight lacked tactical flexibility as it had to be manually adjusted if any of the parameters changed.
Our piece includes its original container (including several accesories within the containers drawers), and clearly was clearly in Canadian in 1942. Our particular version, the Mark IXa, was intended for use in land bombers. However, it has the "moving target" attachment, which was usually being reserved for those planes doing maritime duty, where the ship they were aiming at would be moving. |
|
|
The famous B-29 bomber was a critical factor in ending World War II.
It incorporated many technical innovations including a Central Station Fire Control System (CSFC) for its five gun turrets.
Prior to the B-29, gun turrets in bomber such as the B-17 and B-24 contained a gunner responsible
for aiming and firing the guns in the particular turret (more on these in a subsequent section on gunsights).
The B-29, however, had four unmanned turrets (the tail gun was still manned)
that could be (as shown in the left diagram) controlled from several remote locations: a central sighting station near the tail,
a nose gunner statiom and the tail gunner.
The advantages of the CSFC were several: smaller gun turrets (less drag), improved sighting accuracy, easier maintenance, redundancy if a sighting station is incapacitated,
improved comfort for the gunners, and so forth.
The CSFC comprised many components (e.g., servo amplifiers, motor generators, sighting heads, the gun turrets themselves), but the critical component was the computer itself, which we have. There was one computer for each of the five sighting stations. Based on information from the sighting station along with the plane's altitude, air-speed, and temperature, the computer calculated the proper aiming of the turret considering ballistic factors (wind and gravity), parallax (offset between gun and sighting head) and the appropriate lead to the target based on range and relative velocity of the target. Shown here are
|
|
|
Here is the nose gunsight used in the giant Convair B-36 bomber.
It is heavy (about 150 pounds); on the left is the sight in my office.
On the right is a picture of a sight mounted in the nose of a B-36.
The nose sighting station is a horizontally-mounted, double-prism periscopic sight that gave the gunner a complete hemisphere of vision when sighting through the eyepiece. The sight has at its forward end a spherical glass dome head which projected through the nose of the B-36. Rotation of the gunner's hand grips positioned scanning prisms located in the head. This sight was manufactured by Farrand Corporation; General Electric was the integrator of the B-36 gunnery system. Apparently, the ability to view an entire hemisphere without moving the observer's head was an non-trivial invention and Farrand patented it (under the strange name of "Scanning Telescope Having Asigmatized Pupil", U.S. Patent 2719457). Here's a quote from this patent: "The system illustrated is useful as a gun sight for the control of remotely operated guns which customarily use the polar coordinate system. This is particularly true of aircraft installations." (Shortly after this patent issued, Farrand sued the government for "unauthorized use" [D.C.S.D.N.Y., 175 F.Supp. 230; D.C., 197 F.Supp. 756]. The result was appealed to the U.S. Circuit Court of Appeals and is an interesting case relative to the patent concept of "reduction to practice". But I digress...). Even though it seems obivious that this is a B-36 nose sighting station, there are some mysteries. Instead of looking directly ahead, the optical path is slanted as if the observer looks at a 60 degree angle to the nose. The pictures seem to imply a more straightforward alignment. Also, what about the other arm of the Y-shaped nose? I believe it is a port for a camera attachment, but I don't really know. |
|
|
Here is a Y-4 optical bombsight used in the Boeing B-47 bomber (specifically the B-47B and B-47E models).
It is huge; on the left is the bombsight being wheeeled into my office.
It's about 6 feet long and weighs about 200 lbs (the sight is actually upside down in this picture, on the right is a picture of it right side up).
Next down on the left is a drawing from the Air Force Museum site showing the location of the Y-4 in the bombardier-navigator compartment in the nose of the B-47. The Y-4 bombsight provided an optical view of bombing targets through the glass bubble which produded from the nose of the aircraft. The geared lens mount inside the bubble (a closeup is shown on the right) moved as the plane moved to keep the line-of-sight on the objective. Unlike the World War II bombers of the previous decade, this allowed the pilot to take evasive action as he approached the target without the operator losing his aiming point. The left picture on bottom row shows the bombardier's viewing port. The left eyepiece show the optical view from the external sight. The right eyepiece show a copy of the azimuth view of the bombing radar. |
|
|
This is part of a complex of devices that integrated navigation, radar, and bombing functions for the
giant 10-engine B-36
bomber.
We have two components: the APA-59 navigation computer shown here and the A-1A bombing computer (shown as the next item).
On the right is a picture of the navigator/bombardiers station in the B-36 showing a very recognizable APA-59 (outlined in red) just beyond the bombsight periscope. |
|
|
xxxx xxx x x x x xxxxxx x xxxxxxxxxxxxx xxx xxxxxx xxxxxxxxx xxxx xxxx xxxx |
|
|
This is a navigation computer used in the RB-57,
the C-135,
the F-100,
and the RF-101.
(Here
is a site summarizing the U.S. military equipment
designations, such as AN/ANS-7).
The ANS-7 first flew in the late1950's. It was made by the Ford Instrument Co. (not affliated with Ford motor companies). At the time Ford was one of the premier manufactirers of electro-mechanical analog ocmputers, including the most impressive monster: the Mark I, which weighed over 3,000 lbs. (More on the Mark I below). To the right is a closeup of two of the three disk integrators (as described in the intro to this page) contained within the ANS-7. |
|
|
|
|
|
|
|
|
The K-3 is a computing gunsight used in used in Sperry upper gun turrets used on the B-17 bomber.
The K-4 is the same device with a different mount used in the Sperry lower gun turrets on the B-17.
The gunner inputs range information by estimating the size of the plane and adjusting its image in the attached optical sight so that the image fits withing reticles. The gunner then tracks the target with the optical sight by moving the K-3 (mounted on a movable head) keeping the plane image centered in the reticles. The sight movements cause the computing unit, which, based on the range information and built-in ballistics data, to calculates the deflection, or lead, for aiming the guns and moves the turret accordingly. On the right is a picture for the Gunner's Information File illustrating the lower turret with the K-4 sight right in the middle. |
|
|
The Sperry K-9 is a computing gunsight used in Marin upper gun turrets on the B-26 and B-24 bombers. It operates similarly to the Sperry K-3 except that the computer is fixed and gets information from a separate movable sighting head. |
|
|
This is a navigation related computer and display apparently made for the Royal Australian Navy (see the label) by Huyck Sysyems (label on the side, not shown).
It is filled with a variety of electro-mechanical components (gears, servos, electrical clutches, potentiomemeters, etc.)
as well as some electronics (high-gain and low-gain amplifiers, quadrature rejectors, and a lot of resistors and capacitors).
Of particular interest is a 19 position mechanical analog memory. The "memory set" knob rotates a carrier with two cams for each of the 19 positions. The selected cams each move an arm whcih moves an electrical resolver to provide the electrical value for the position. The back of the carrier and its cams and the sensing arm are shown in bottom left picture (there is a cam and arm on both sides if the carrier). What is not shown is a mechanism at the front of the carrier where two servos engage the two screws that adjust the length of the selected cam. This is how the memory values are set. Exactly what this device does is not clear. The NS and EW knobs on the right move the crosshairs on the display and set values via potentiometers. When not being set, the crosshairs are driven by the computing mechanism. Also, the display is presented in degrees (not NS-EW) and is labeled "wind". Based on my piloting experience, this seems too complicated (oe weird) just to be indicating wind direction. A center knob sets a variance, but variance of what to what? Again, when not being set, the variance display is being driven. Below this knob is a switch (not easily seen) labeled "wind" with positions "man" (surely manual) and "dop" (Doppler?). |
|
|
This is the computing element from an unknown gunsight. It is particularly interesting since it makes extensive use of cams, which are an analog version of a ROM. In particular, its has two two-dimensional cams, one of which drives the axial position of the other. In the top right picture one of two cam followers stick up which drive "logic" above this assembly. |
|
|
|
|
|
|
The Mark I was first introduced around 1930 with variations up
until the 1950‘s (our piece is a Mod 7).
Used on almost all US Navy
ships in WWII.
Acomplete Mark I is monstrous—it weighs about 3,150 lbs.! Here it is in action aboard USS Hornet… http://dcoward.best.vwh.net/analog/ford.htm http://www.eugeneleeslover.com/USNAVY/CHAPTER-25-C.html |
|
|
|
|
This is a "torpedo director" used on the WWII Douglas A-20 bomber.
According to the Douglas A-20 Havoc Pilot's Operating Instructions (which we have), a torpedo director "is an instrument operated by the pilot
to help him approach a moving target in such a manner that the torpedo will intercept the target on its course."
The computation mechanism of the B-2 is simple, but it does compute in a way. On the right is the first page from the 1945 patent for the B-2, which describes the operation of the B-2. |
|
|
|
|
Before GPS, a primary navigationa method for ships, aircraft, spacecraft and missles was celestial tracking, also known as astronavigation.
The basic approach is to sight on the sun or a particular star and calculate the position of the aircraft, missle or ship from the sighting angles to the star.
The calculation is based upon highly accurate data about the positioning of the sun and certain reference stars at various time and locations.
For modern applications, the star tracking location is used to calibrate and update an inertial navigation system.
Early Air Force aircraft such as early B-52 bombers relied on a human using a sextant through a bubble in the top of the plane. Some time in the late '60s, manual sighting was replaced on the E and F models by this automatic astro tracker mechanism and its associated analog computer. At some point, the astro navigator was replaced by GPS. Our device has a maintenance tag dated 1983, so the system was used at least until then.
Similar "star trackers" were and are used on many other planes such as the SR-71 and RC-135 reconnaissance aircraft, the B-58 bomber, and the P-3 Orion patrol plane. Of special interest is the fact that the most modern USAF plane, the B-2 bomber, still has a automatic star tracker navigation system. Many missiles, including some currently operational ones, also use automated start tracker navigation: the Polaris, Poseidon, Trident, MX, SM-62 Snark and the AGM-28 Hound Dog, and others.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
