• Welcome to the new Internet Infidels Discussion Board, formerly Talk Freethought.

Control Systems

steve_bank

Diabetic retinopathy and poor eyesight. Typos ...
Joined
Nov 9, 2017
Messages
13,817
Location
seattle
Basic Beliefs
secular-skeptic
I put this in math forum because it is heavily mathematical. It is one of the most math rich areas in engineering. I dwelled here a lot.

Basics. In the first block diagram on the right imagine a car speed control system. The block diagram form is the common way to depict a system.

https://en.wikipedia.org/wiki/Control_theory

The PID controle4 in in industry is as common as rain. The user programs in constants for the three terms. The circle with a summation signal is a basic symbol. The output of the symbol is the algebraic summation of the signed inputs.

https://en.wikipedia.org/wiki/PID_controller
http://ctms.engin.umich.edu/CTMS/index.php?example=Introduction&section=ControlPID
http://www.cds.caltech.edu/~murray/books/AM08/pdf/am06-pid_16Sep06.pdf

'bang bang' control is common. When a romm is cold the thermostat turns on the heat full. Say the set point is 70. When the temperature hits 70 the heat is turned off by the temperature keeps rising a bit until the room begins to cool down. When the temperature hits say 65 the heat turns back on. The 5 degree difference is called hysteresis, It keeps t\the system from constantly cycling the heater.

https://en.wikipedia.org/wiki/Bang–bang_control

The top level most general model is sate variables. In a complex digital control system it invokes solving systems of differential-integral equations and matrix operations on the fly. S represents complex variables, poles and zeros.

A complex system can have multiple loops and summation points multi[le inputs, and multiple outputs..

https://en.wikipedia.org/wiki/State-space_representation
https://en.wikipedia.org/wiki/State_variable

Some bed time reading for you.
 
Cool topic!

I'm old enough to have built and used analog PID control systems with moving parts....synchros and servos and the electronics was discrete components around very old thermionic valves.
 
Personally, I am a big fan of Watt's centrifugal governor. It was ingenious, simple, and damned effective. Although no math is needed (unless someone wants a mathematical understanding), only an understanding of physical principles.

 
Agreed - Watt was incredible. And, of course, a Scot! One of his other major contributions was the separate steam condenser which radically improved the efficiency of steam engines at the time, and essentially making mobile engines feasible....thus ushering in the railway age.
 
I was thinking of developing some kind of general analog fractal compilation device.

One of the ideas was real time inputs that calculated some values and outputted them as a signal to be used to draw on a screen (probably analog to digital input).

I eventually realized the millions of calculations required for one image would require >= a billion analog devices to generate the image (assuming a minimum 1000x1000x1000 voxel resolution). Digital it still is. Although, I still think some cool effects could be generated cool and fast (thinking old psychedelic shows probably used analog display devices).
 
Modern control systems came out of WWII in weapons controls mostly out of Bell Labs. Bode pronounced bow-d, a mathematician produced the first comprehensive book with theoretical foundations. The Bode Plot is a primary tool.

https://en.wikipedia.org/wiki/Bode_plot

https://en.wikipedia.org/wiki/Hendrik_Wade_Bode

Another stability analysis technique.

https://en.wikipedia.org/wiki/Root_locus

Analog computers calculated torpedo firing solutions and compensated for pitch and roll for surface ship guns.

A good basic book. I used to have a copy, if anyone is interested. There should be used copies.

https://www.amazon.com/Art-Control-Engineering-Ken-Dutton/dp/0201175452
 
I was thinking of developing some kind of general analog fractal compilation device.

One of the ideas was real time inputs that calculated some values and outputted them as a signal to be used to draw on a screen (probably analog to digital input).

I eventually realized the millions of calculations required for one image would require >= a billion analog devices to generate the image (assuming a minimum 1000x1000x1000 voxel resolution). Digital it still is. Although, I still think some cool effects could be generated cool and fast (thinking old psychedelic shows probably used analog display devices).

Yor could have a box with potentiometer connected via USB to a PC. The PC app takes pot voltages and converts to fractal constants. I charge by the hour :D
 
I was thinking of developing some kind of general analog fractal compilation device.

One of the ideas was real time inputs that calculated some values and outputted them as a signal to be used to draw on a screen (probably analog to digital input).

I eventually realized the millions of calculations required for one image would require >= a billion analog devices to generate the image (assuming a minimum 1000x1000x1000 voxel resolution). Digital it still is. Although, I still think some cool effects could be generated cool and fast (thinking old psychedelic shows probably used analog display devices).

Yor could have a box with potentiometer connected via USB to a PC. The PC app takes pot voltages and converts to fractal constants. I charge by the hour :D

When I was in a garage band in junior high I made a light display by connecting guitar and vocal outputs to the deflection coils of an old color tv.

Anyone know of a windows program for creating fractal displays? I had one many years ago that allowed assigning colors or shades of a color to various levels that could be made to cycle so as to simulate movement. You could pan and zoom into the mandelbrot set or any other and explore infinite and beautiful landscapes.
 
Modern control systems came out of WWII in weapons controls mostly out of Bell Labs. Bode pronounced bow-d, a mathematician produced the first comprehensive book with theoretical foundations. The Bode Plot is a primary tool.

Yep...that's where I learned...I worked for the UK Navy at the time. Most of them were for Radar and Sonar transducer control. Bode plots were valuable tools.
 
I was a gun fire control technician in the early sixties. Didn't care much for any of it then except for the Mark 56's capability to keep the director level fore and aft so I wasn't as sea sick as I'd be when cleaning johns.

Later, much later I designed, built, and used a three dimensional stepper motor hydraulic cylinder system driven by computer sound system to find minimum audible angles in an anechoic environment. I must have learned something from my Navy experience. There I had to compute startup to more or less linearly overcome travel rates to minimize speaker wobble from rest and moving inertias to stop and to drive at rates resulting in particular transits for sound to be moving 2, 4, 6, and 8 degrees per second in an arc about an observer's head holding a constant distance from her near ear.

Interfaces, energy conversions, appreciation, all in a single experimentals paradigm.

Let me tell your doppler effect is a lot of fun to study as a means for converting energy to information in a nervous system.
 
Modern control systems came out of WWII in weapons controls mostly out of Bell Labs. Bode pronounced bow-d, a mathematician produced the first comprehensive book with theoretical foundations. The Bode Plot is a primary tool.

Yep...that's where I learned...I worked for the UK Navy at the time. Most of them were for Radar and Sonar transducer control. Bode plots were valuable tools.

If you include Smith Charts then we are really takin' .
 
Modern control systems came out of WWII in weapons controls mostly out of Bell Labs. Bode pronounced bow-d, a mathematician produced the first comprehensive book with theoretical foundations. The Bode Plot is a primary tool.

Yep...that's where I learned...I worked for the UK Navy at the time. Most of them were for Radar and Sonar transducer control. Bode plots were valuable tools.

If you include Smith Charts then we are really takin' .


Yep - did my first Smith Chart at age 17. on RADAR waveguides, Magnetrons and Klystrons.
 
Back
Top Bottom