Our fascinating look at building automation history starts in 270 B.C.

Welcome to the brief building automation history page!

Our emphasis concerning building automation history will be about Direct Digital Controls for HVAC systems, but a quick study on general development history is in order:

The earliest documented example of an automatic control system device happened in the year 270 B.C. In this example, the device was a water clock, shown below.

(I think the Gentleman is slightly out-of-place for the period, but you get the idea!)

More specifically, if we're talking about HVAC related systems, during the period of the 1700's, Rene-Antoine Ferchault de Reaumur (1683-1757, inventor of the Reaumur Scale thermometer) had some ideas for temperature control of incubators.

(Reaumur shown below)

Reaumur's ideas were based on an invention by Cornelius Drebbel (1572-1663). Drebbel's invention used a U-shaped vessel containing mercury, sensing temperature, that operated a lever arm, controlling the draft to a furnace that controlled heat output.

During these early times, there were several things going on at the same time in different areas of automatic control systems development.

(Drebbel shown below)

Automatic control of the steam engine and the various attempts at maintaining accurate control of the steam engine helped to lead the way for improvements on automatic control systems in general (James Clark Maxwell 1831 -- 1879, and his work titled "On Governors" for example, that dealt with linear differential equations for governor speed control). Maxwell also wrote a textbook called "Theory of Heat" in 1871.

Most of the automatic control systems during the 1800's dealt with temperature, pressure, liquid level, and the speed of rotating machinery. So long as the operators had stability, the folks that operated the systems were generally happy.

(Maxwell shown below)

Further improvements began to be made when larger ships, new weapons and propulsion systems began to appear. The systems used pneumatic power to operate position control mechanisms. As we neared the 1900's, electricity also began playing a part in automatic control systems.

One of the first examples of PID-type controls that were developed was by Elmer Sperry. In 1911, this type of system was used for automatic ship steering (Sperry did a lot of work involving gyroscopic compasses). Sperry's device compensated for disturbances in the water as sea conditions changed.

In 1914, one of the Foxboro Instrument Company founders named Edgar Bristol, invented the pneumatic "flapper nozzle amplifier." The action of this flapper nozzle can be likened to a flexure lid on the modern pneumatic controller. Early flapper nozzle systems that attempted to control devices were proportional and because there was no integral or derivative action, made a controlled device act like a binary "on-off" instead of modulating.

Although Elmer Sperry used a type of PID control in 1911, the control law that we commonly associate with the modern PID loop came from Nicholas Minorsky (1885-1970). In 1922, he observed a helmsman controlling a ship and came up with the proportional, integral, and derivative type of control we know of today:

Proportional is the control required to steer the ship based on actual ship direction compared to the desired course setpoint.

Integral is the amount of reset required to correct an amount of error. For example, if the ship is off course by a small amount, and correcting it to the left brings it back on bearing, then turning the wheel all the way to the left is inappropriate. Only a slight adjustment to the left is required.

Derivative is the attempt to see how far a process variable (ship course) has been from the set point in the past, and anticipating where the course correction will need to be in the future.

In the 1920s, several instrument companies began making complete boiler control systems.

During the 1930s, the Foxboro Instrument Company made improvements to the flapper nozzle system that incorporated the integral and derivative. The controllers made during this period were very expensive and also not very well understood in operation.

(An example of a PID Loop shown below)

In 1936, the first professional organization that dealt with automatic control was created and called the Industrial Instruments and Regulators Committee. Standardizing the automatic control design methods and terminology were the primary goals of this early organization.

During the 1940s and 1950s, instrumentation companies continued to improve their products. For example, the Foxboro Instrument Company featured an improved Stabilog controller that was field adjustable, and the Taylor Instrument Company did the same for the Taylor Fulscope. During this time, papers were also published that describe how to adjust the PID values for optimal settings on the controllers.

From the 1930s to the 1950s, some very important work was done by MIT, the IEE, and by several people, both overseas and in the United States regarding electrical and electronic automatic controls. This was a very exciting time period in electricity and electronics, which led to the early systems of modern weapons development, communications, and semi conductor circuitry.

Control systems evolved quickly, such as an example of electric relays, versus transistor controls. Without the fast and accurate control systems that were developed during this period, defensive systems used in the free world would not have been accurate, nor work correctly.

There were many spinoffs to the military technologies that were developed since the 1950s. Small-scale electronics lead to affordable computers, and those computers constantly evolved from the 1950s to the present.

The personal computer that we use today, and the electronics used in today's digital controllers, make it possible to effectively and easily control the environment within a modern building.

1. Stuart Bennett, "A Brief History of Automatic Control," June 1996.

2. Wikipedia, the free online user-generated Encyclopedia.

Share this page:  What's This?