In the early 1930's Ford Instrument Co. introduced the Mark 1 Fire Control Computer. This computer was a purpose-built computer of the electro-mechanical kind, its output being analog in nature.

In my opinion, two important developments made the Mark 1 possible:

One was a device called a Selsyn which was developed about 1925. This was the first really good device that could transmit and receive electrically the mechanical position of anything.

So with this device director information could be transmitted electrically to a remote location, such as a computer. The computer could transmit its output electrically to a gun mount or turret.

The other device that was really important was the mechanical rate integrator. With this device, time and position, for the X, Y and Z axes, could be integrated together.

Integrating time and position together with the other components of the FC problem could predict a future position.

This put Ford Instrument Co. and the USN on the road to producing the finest FC System of the time.

At this time all of the equations and formulae were well known that were necessary to solve the FC problem to the target. This was incorporated into the Mark 1 computer.

Ballistic calculations were available on all sizes of guns in a horizontal plane. I do not know which gun was used for the ballistic calculations for the first Mark 1 computer.

As I understand it, the original Mark 1 computer could solve only the X and Y coordinates of the FC problem.

The Mark 1 has no memory, so it does not remember any past event(s).

The computer is only interested in current events.

Through the use of the mechanical rate integrator, which is the computer's crystal ball, the computer can predict the future. The future position of the coordinates that it is currently measuring.

The ballistic section of the computer contains all of the ballistic information for the weapon that the computer is to control.

All of this information is based on new gun performance and new gun initial muzzle velocity (IV).

This section of the computer has one manual input, Current IV.

Bore information is used to manually compute bore wear, so that you can manually produce a current IV for the gun.

Available weather information, plus current bore wear, produces the current IV for the gun.

The weather information is measured on an hourly basis. From this you calculate and manually input the current IV to the ballistic section of the computer.

The differential is widely used in the Mark 1. The differential is like a car differential except that it is small enough to hold in your hand. The ones in the Mark 1 have a 2:1 gear ratio, which means that the drive shaft input requires 2 turns in order to get the wheel shafts to make one turn.

Just like a car differential, if you hold any one shaft, and turn one of the other shafts, the third shaft turns in response.

If you hold the drive shaft input, and turn one of the wheel shafts, which are opposite each other, the opposite one will turn.

The differential was used in places where there was a need for one shaft to have an input from two places.

For example, the target bearing from the director is fed into one of the shafts of the differential – in this case that would be one of the wheel sides. The opposite shaft might be connected to a manual knob that would be to enter an offset bearing. The drive shaft side, which is going to turn at 1/2 speed of the director input because of the 2:1 gear ratio, then drives some aspect of the computer gear and cam mechanism. If the output of the drive shaft side needs to turn at the same speed as the input shaft, a set of 2:1 gears does this.

The hand knob manual input, for offsetting bearing is held by a friction disk to the computer case, so the knob requires some effort to turn it. This ensures that the input from the director bearing goes out via the drive shaft side of the differential and not back out through the hand knob. There is also a friction clutch in the shaft and gearing between the manual input knob. This slips if the mechanism reaches its maximum travel and encounters a mechanical stop. This is to prevent damage to the internal parts of the computer.

The carriage of the mechanical integrator needs an input from two shafts. One is the time constant that drives the carriage to the center of the disk. The other input is bearing or range that moves the carriage away from center.

The computer is about 3 ft wide, 4 ft tall, and 6 ft long. Looking inside the Mark 1 was like looking into a fine mechanical watch. Everything was very compactly designed and organized. The first look is completely mind bending, making one wonder what is all this stuff and how does it work? How could anyone ever have put together something like this and made it work? Especially in 1930.

The components are so packed inside that you can only stick your finger into the components. There is not enough space for your hand or arm to fit inside. All of the components are either parallel or at right angles to each other.

Each component part is built like the mechanical integrator, all in one piece. The component parts are all reasonably strong and the mounting plate it is built on is usually a sheet of steel 1/2" thick. The component parts are usually mounted using 4 to 6 1/4" socket head cap screws. You do not have to be very careful with these parts, as they do not damage easily.

The component parts have input and output gearing and shafting which connects them to the other component parts of the computer. Most of the shafting is 1/4" diameter, some is a little larger. The shafting is held in place by ball bearings, which are held in place by various mounting blocks. Miter gears are used to make 90 degree turns of the shafting which allows them to connect to the various component parts.

A tool box is furnished with the computer. This tool box contains the special tools necessary to reach inside the computer. Some of the tools have small lights on the end of a rod about 3 ft long. The light can be swiveled on the end of the rod, so you can put it far enough into the computer to see with. Other tools are 3 feet long with 90 degree bevel gears at the end. This might hold an Allen wrench, so you could turn an Allen screw or socket head cap screw deep in the computer. As you can guess, working on the internals of the computer while the ship was moving was quite difficult.

The miter coupling is used to make adjustments to the input and output shafting of the components.

This coupling consists of two parts, or halves, each of which are mounted on the end of a shaft. The miter coupling connects two shafts together. One of the coupling halves has teeth on its outer diameter, sort of like the teeth on a gear. The other half of the coupling, mounted on the other shaft, has a screw in it that its threads engage the teeth of the gear on the other half of the coupling that is mounted on the other shaft. Turning this screw then moves the coupling halves in rotation, with respect to each other. This, then, is the mechanical adjustment that aligns the shaft input, or output, to and from the different mechanical components. Once the adjustment is made, there is a locking screw in the miter coupling that when tightened does not allow movement of the adjusting screw. This then keeps the adjustment from changing.

The original Selsyn devices were about 8 to 10 inch diameter and were also about that long. Even though they were large, they did not have much power to drive or position anything other than a very light load.

So the Selsyn was sort of re-invented, and made smaller, down to about the size of your fist. These were re-named Synchros, and the term stuck.

The synchro in the Mark 1 positions cams, which close and open electrical contacts, which in turn, start, stop and run servo motors. The servo motors actually provide the power to turn, set or adjust the various elements in the computer, all as accurately as if the servo had actually positioned the element in question.

This gave, or gives, the computer the ability to receive remote inputs of range, bearing, and elevation to the target, from a remote director.

True North is an input from the ships Master Compass. It is also transmitted electrically and received by a synchro, which inputs True North into the computer.

The ships heading or course is another input to the computer. This allows the computer to keep up with the ships relative heading with respect to True North.

On the computer there is an indicator which is an outer ring of a dial. This dial shows True North. Inside this dial there is another dial, which shows the ship's heading. This dial also has an outline of a ship drawn on it, with the bow at zero degrees. When you look at this dial, you do not have to look for the zero to know which way the ship is heading. All you have to do is read the picture of the outline of the ship.

The computer needs to know where True North is with relation to the ship's heading.

This information is necessary because of a naturally occurring thing called Coriolis Force.

Coriolis Force is an apparent force that as a result of the earth's rotation, deflects moving objects, (as projectiles or air currents), to the right in the northern hemisphere and to the left in the southern hemisphere.

The computer does not correct for Coriolis Force, but rather for Coriolis Effect.

Coriolis Effect is the apparent deflection of a moving object that is the result of the Coriolis Force.

The weight of the projectile and its speed must be known to correct for this effect.

The computer must also know the ship's latitude. This is because at different latitudes, the effect from Coriolis Force is different.

The direction the projectile is fired in with relation to True North must also be known. This is because firing at or in different bearing directions with relation to True North the Coriolis Effect is different.

The FC Directors on a Battleship are about 100 feet above the water line. As the ship rolls and pitches, the director is tilted out of the true horizontal plane, of or with the level of the ocean. The director is sort of like mounted on a vertical stick about 100 feet tall, with its pivot point at a point below the water line. So as the ship rolls and pitches, the FC Director swings back and forth at the top of the 100 foot tall stick. Since the director is locked on and tracking a target, this swinging back and forth is being transmitted to the computer as changes in bearing and elevation of the target. This is actually false information, because the target is not making these moves.

To correct this false information, a true horizontal plane is generated by a device called a stable vertical on Battleships and Cruisers. On Destroyers, a stable element provides this information. There is actually very little difference in the two devices.

The true horizontal plane generated is transmitted mechanically to the computer. The stable vertical or stable element sits about 18" away from the computer, and mechanical shafting transmits the true horizontal plane directly into the computer.

Inside the computer, this input is applied to the director inputs for bearing and elevation by the mechanical differentials. They add or subtract as necessary from the FC Director inputs, to convince the computer that the FC Director is not moving and is staying in a vertical and horizontal position with relation to the ship and the target.

The Mark 1 uses true north to keep up with the ships heading and to keep up with the target in relation to the ship and true north.

It is necessary to keep up with true north, the ships heading, and target bearing, in order for the computer to make the correct adjustment to correct for Coriolis effect.

The movement of the ship in course and speed are transmitted into the gun, so when the gun is fired, this movement is imparted to the projectile, and becomes a part of it's flight pattern or trajectory.

The computer then corrects the position of the gun, to remove this imparted motion in the projectile, so that the projectile will land on target.

If the target is moving, for the computer to correctly calculate the true course and speed of the target, true north must be known.

As the FC Director continually inputs bearing and range information into the Mark 1 computer, the computer is continuously producing an output to position the guns to a position that will hit the target.

So if the target is stationary, accurate gun fire can be done, or commence, at any time because the computer does not have to predict a future position of the target. The FC solution is instantaneous, with a stationary target.

If the target is moving, it will take the computer a maximum of 30 seconds, to predict the course and speed of the target, from the time, or moment of target acquisition.

If the target changes course or speed, it will take the computer only a few seconds to come to a perfect solution for the new course and speed, but not an entire 30 seconds, as it needs for a new target solution.

Range and bearing to the target can be put into the computer manually, without using the FC director.

The computer has mechanical rate integrators that can correct the manual input of range and bearing for the course and speed of the ship. The ship can then make any change in course and speed but the range and bearing manually set into the computer are automatically reset by the mechanical rate integrators. With these inputs, the computer can keep track of the target's position relative to the firing ship at all times.

The manual inputs of range and bearing are normally used in shore bombardment, and in particular when there is no line of sight between the FC director and the target. The actual range and bearing figures used usually come from CIC (Combat Information Center).

They can, however, be transmitted from a lookout or any other place directly to the FC officer in the FC plot room with the computer. In which case the information might just be their best guess. However, after seeing where the rounds impact in relation to the target, offset corrections can be easily made and the second salvo will be much closer to the target..

The starshell section of the computer allows you to take one gun mount or turret and reposition it to fire starshells for target illumination for the other weapons.

This section of the computer takes the generated or computed firing solution and allows the operator the control necessary to offset the gun from the target position.

Offsets can be set into the starshell section to correct for range, bearing, elevation and the prevailing wind so as to correctly position the exploding starshell.

The A and B tests checked the computer for accuracy.

The A tests checked the input information against the computed output information. These tests determined if the computer could correctly solve a particular FC problem.

There was a table of about 50 FC problems with the input figures for each problem. The table also contained the output figures for each problem and the allowable tolerances for error of each problem. The operator entered the input figures into the computer and checked that the output figures agreed with the table values within a small tolerance envelope.

The tests checked the computer for many bearing and range inputs, so as to make sure that the computer could solve correctly a firing solution to any target.

These test were run, let's say, on Monday, Wednesday, and Friday. Usually about 6 tests on each day. So about every 30 days, the complete set of A tests were completed.

The B tests checked the mechanical rate integrators, to make sure that the computer could apply time to the computer inputs and correctly predict the present course and speed of the target and produce a gun output position for the future position of the target.

The B tests were run on the days between the A tests. Like the A tests, a table gave the inputs to be used and what the output solution should be, and gave the allowable tolerances for error.

These B tests, like the A tests, were to check the computers ability to solve any FC problem at any range and bearing.

Once the Mark 1 computer came into being, the use of ladders for ranging went out the window. As long as there was a line of sight to the target or if the correct grid coordinates of a shore bombardment target existed, the Mark 1 eliminated any need for “ranging ladder salvos."

In test firings, or firing practice, ladders were still used to teach how to use ladders to obtain range.

Firing practice trained the crews of the entire ship in their respective jobs.

Firing practice checked out all of the equipment, to make sure that all was in working order.

Firing practice checked battery alignment and the pattern of the shells in the target or impact area.

If the pattern size, range, or bearing was wrong, at the first opportunity, battery alignment was done to correct the errors.

Offsets, set into the computer, could correct for bearing and range misalignment, and the computer would keep these corrections.

The computer could not correct for pattern errors. If one gun was off it had to be manually corrected in battery alignment.

A bridge is a target 10 miles inland and out of the line of sight of the firing ship. Through the use of spotters, range and bearing, usually derived from a grid map, are transmitted to the firing ship.

CIC provides Fire Control with the range and bearing to the target. The spotter gives corrected information to the firing ship, if range and bearing corrections are necessary, to adjust the fire to hit the target.

CIC records ship position and range and bearing to the target as each salvo is fired. Once correct range and bearing to the target are obtained, which is evident by your shells hitting the target. CIC records the correct position of the target on the grid map.

This ship, or any other ship, can come back a year later, and with the previous CIC records, can open fire and hit the bridge, with no spotting necessary.

Using the Mark 1 computer to amass fire.

Accurate fire on a single target, from several ships became possible, with the Mark 1 computer.

With one ship firing on a target, that ship transmits relative range and bearing from it to the target.

CIC of each ship, then uses relative range and bearing of the firing ship, to provide their Fire Control plot room, with their own range and bearing to the target.

The ships do not have to be in a battle line, they can be in any position, although steaming in a battle line, is what was generally done.

About 1935 Ford Instrument Co. added the ability to the Mark 1 computer to compute rate changes in elevation or the z coordinate.

The Mark 1 could compute in elevation, it just could not compute rate changes in terms of elevation. It knew if the target was on top of a mountain, but that was about all it could do in elevation calculations. It could not tell if the target was moving in elevation. The ballistics cut for the Mark 1 were only for surface-to-surface firing and there were no provisions for the surface-to-air calculations needed for the anti-aircraft role.

The Mark 1A overcame this little problem.

It also changed the maximum allowed target speed of about 200 knots up to about 450 knots.

The new computer design became the Navy's first Dual-Purpose computer, as it could handle aircraft targets as well as surface targets.

When coupled to the 5"/38 gun, the Navy now had the finest anti-aircraft fire control system in the world. No other Navy could match this FC system through out World-War II. The British purchased this system for use on their ships, but only a single anti-aircraft cruiser actually saw service with it during the war.

The Mark 1A dual purpose computer together with the 5"/38 in single and twin mounts were deployed on almost every USN ship from destroyers through battleships and carriers by the end of the war.

The Mark 1A is what made the 5"/38 the most successful dual-purpose gun of the war as it gave those guns true anti-aircraft capability. No other nation came close to developing the computer-gun combination so well.

As the Jet-age came about in the late 40's, the computer was modified to allow maximum target speed about 650 knots. Our jets were approaching the breakneck speed of 550 knots. There were many who thought that the sound barrier would not be broken.

As far as I know the only ballistics that were ever cut and installed in the Mark 1A were for the 5"/38 gun until just about wars end. At that time, the 5"/54 was being introduced on the USS Midway class carriers. This weapon had different ballistic characteristics and needed different computations than did the 5"/38.

There was no scheduled maintenance, recommended or performed on these computers. They were quite reliable and lasted for years without breakdowns.

• "Ford Instrument Company Inc." at Doug Coward's Analog Computer Museum and History Center

29 May 2000

Updated.