Glory Days Military

Analog Fire Control

When I was in the Navy, the ship that I was on had a 48 inch diameter carbon arc searchlight on it. The searchlight worked by taking what was essentially two welding rods, pressing them together, and maintaining an electrical arc in order to create a searchlight beam that was bright enough to be seen for miles. In fact, by shining that light at a cloud, it was possible to send Morse code signals to other ships over the horizon.

When my ship was built in the mid 70’s, these lights weren’t made any longer, and the one that was on my ship had been salvaged from a WW2 era destroyer that had been decommissioned. Built in the days before electronics, the system that ran this searchlight was incredibly complicated. It was an analog power supply that ran on a system of motors and gears, with lenses focusing beams of light on various parts of the system that turned motors on and off, pushing the rods closer together, or pulling them apart, as needed to maintain the light beam. A technical manual for a 24 inch example can be found here.

By the time I reported aboard the ship, the light no longer worked and no one knew how to fix it. At one point as a young E-4, I took an interest in this searchlight and decided to get it working. I made a project out of it. I found a manual in the ship’s tech library, brought the control unit down, and spent several weeks rebuilding it. When we finally got the thing lit, it was amazingly bright. The light hadn’t worked in years, and I didn’t get so much as an “attaboy” for getting it working. Nowadays, it seems like you would get a Navy Achievement Medal for fixing that thing.

I tell you this as a setup and explanation of where I got this interest in how early electrical engineers solved problems that seem easy today using electronics. The focus today is on the Ford Mark I fire control system.

The Navy needed a way to calculate the elevation and deflection of Naval guns so as to put shells on target. This was no trivial exercise in math. Both the target and the gun platform were likely moving, the target might even be airborne, the platform might be rocking in heavy seas. Different shells were of different weights and ballistic coefficients. Or you might want to put a starburst shell 50 feet over the target for illumination. Ranges were sometimes 30 or more miles away. All of these factors required math in three axes in order to be overcome: direction, distance, and elevation. Enter the Ford fire control computer.

A frigate might have one. Destroyers had two, allowing multiple batteries to engage different targets. An Iowa class battleship had four of them. They were accurate enough that this computer was still in use until the battleships were retired in the mid 90s. 50 years old is not bad for an analog computer living in the age of transistors.

Check out this video on how the system worked to direct the secondary batteries on the 5 inch guns of the battleship New Jersey.

What can be done today with a laptop computer took an entire room of switches and a 3,000 pound box filled with motors, switches, relays, and gears. It was bulky, heavy, and more complicated than a box full of Swiss watches, but it worked. It worked quite well, in fact.

I consider myself lucky to have worked on that searchlight. It was one of the most interesting projects that I have ever taken on.

12 replies on “Analog Fire Control”

Geniuses, oh yes!
You look at some of the things that have been done in the past using mechanical computers, and think about implementing the same thing using a $1 microcontroller and some analog I/O…
… And realize that, while the MCU has plenty of crunchpower, the math involved is positively mind-bending, so trying to express it first in equations and then in code, making allowances for the finite precision of numerical representations, becomes a daunting task indeed.
(I got a serious headache a few years back trying to cope with the output of a 3-axis rate gyro. It looks dead easy, until you try to do it. The inputs, and the state variables, interact HOW?????)

Old movie projectors worked the same way. I got to use one when I worked in a drive-in theater. The movement of the rods was automatic, but you had to watch them and adjust as necessary.

Electronic analog computers, using operational amplifiers, have supplanted the older mechanical systems, but I disagree that a laptop would do a good job across the board replacing analog systems. The truth is that digital computers are easier to program, not that they are superior for these real-time high speed computing tasks.

My father has a piece of one of the analog for control computer for a 5 in gun.

It takes as input number of rounds fired. Barrel wear.

It has a 6 inch long cam that rotates and one of the inputs moves a switch along the cam.

Amazing piece of kit.

It is truly amazing the ingenuity of men back in the day. The mechanical distributor with points and condenser and automatic ignition advance was genius. The mechanical computer you described is beyond words. I work in the chemical industry and got to see the older units that ran on pneumatic controllers that have all been replaced by electronics (that still mimic the 3-15 psi signal of the original pneumatics).

Good note about the laptop. If you have serious need for control, you need a dedicated system that responds in milliseconds. Laptops can’t give the same level of response or reliability.

My point about the laptop was to illustrate that what once required the resources of a nation state and weighed more than a ton, can now be done by a device that weighs less than 4 pounds and can be had by an average person for the cost of a week’s salary.

Totally understand. My phone has more power than anything shown on Star Trek. Your point is valid.

The reason that the Iowa’s still used analog computers in the 90s was that they couldn’t do the job on a digit computer then. I’m not sure they could now – as mentioned above, the complexity and speed of calculations is still a huge task.
Don’t forget that ALL digital systems have a built in delay; analog systems are real time unless a delay is specifically added.
In this case, while the processing power is significant, the real issue is all the inputs and outputs and the throughput to connect the two.

Last year,, went inside the USS Alabama at Mobile Alabama.
I was amazed at all of the equipment in the fire control rooms for the guns.
That, and how it all was working together with the ammo feed systems.
Busy and complex below deck.

All that bulk has some side effects. The USN made the Cleveland class cruiser with 4 main battery mounts instead of the 5 on a Brooklyn class to make room below decks for a new AA fire control system.

Comments are closed.