Today is...
Monday, April 27, 2015
Welcome to, the global online
community of automation professionals.
Featured Video...
Featured Video
EtherCAT with CTC’s master lets your multivendor network play well together...
Our Advertisers
Help keep our servers running...
Patronize our advertisers!
Visit our Post Archive
Digital and Analog I/O - basic info
I need to know the application difference between digital I/O versus analog I/O and advantage and disadvantage in between.
By Daniel Johnson on 8 January, 2001 - 9:43 am

I am not a control engineer. Due to the workload, I need to know the application difference between digital I/O versus analog I/O and advantage and disadvantage in between. Can someone tell me the answer?

By Brian Cervi on 8 January, 2001 - 10:22 am

Very simply... digital signals are either ON (binary 1) or OFF (binary 0). Digital input signals are used to represent items that only have two (2) states, such as... ON or OFF states ALARM or NORMAL conditions Similarily, Digital output signals are used to control items that again only have two states, such as.. START or STOP a device turn an indicator ON or OFF Analog signals are variable, they have multiple states. Analog input signals can represent such items as temperature or level or rate of flow. Analog output signals are also variable and can be used for such things as opening a valve to a desired position. So, use digitals for 2 state devices and analogs for multiple state devices.

If i am building a robot and i need it to stop after my odometer has gone so many feet. would that be an analog input or a regular input?
Please email me at

By Hakan Ozevin on 8 January, 2001 - 12:43 pm

The basic example when I am giving education in control systems is as follows (maybe there are better examples from the list): A digital signal is something like telling if a door is open or not. An analog signal is something like telling how much the door is open (or closed). In the first case you have two alternatives, either closed or not (in binary system we call it either 0 or 1). In the second case you have to give a more accurate number to explain the door's status (say the door is 40% open). The accuracy of the figure you are giving depends on the measurement/observation you make. Measurements from the real world are almost always analog and all analog signals cover digital ones (0%=closed, 100%=open), but digital signals are more suitable for computers (PC/PLC, etc). Therefore analog signals are first transformed to digital signals (or series of digital signals). Signals are made interpretable to the computer by I/O's. An analog signal can only be interpreted by an analog I/O, but a digital signal -theoretically- can both use digital or analog I/O (provided that the I/O accepts that level). In practice, you never connect a digital signal to an analog I/O, because analog I/O's are much more expensive than digital ones. An answer to your advantage/disadvantage of analog and digital signals can be given as: 1. Whenever possible, use a digital I/O to give information about the real world (it is cheap) 2. In the other cases, you have to use analog I/O, provided that you cannot use a series of digital signals (no other choice). I hope this introduction will help. ++++++++++++ Hakan Ozevin

By Daniel Johnson on 8 January, 2001 - 8:11 pm

I realized that digital I/O have different storage space, like 1K, 64Kb, or 128Kb. What's the advantage with this? Do they use different 0 and 1's to describe the system's state instead of using analog signal to do so. Can my example be fit: 0000 stands for 0%, which is pump stopped. 0001 stands for 20% funtion. 0010 stands for 40% function. 0011 stands for 60%, and etc. and finally 1111 stands for 100% motor running. Since this digital I/O only uses 4 bytes, and more bytes of course can express other informations as well. Thanks.

By Hakan Ozevin on 9 January, 2001 - 1:09 pm

I think there is a misunderstanding here. The storage space you mention must be the RAM size or the addressing capacity for the digital I/O's. By the way 128 kB is too much for the addressing capacity, altough there are now some high level PLC's that can reach to this number. This is nothing to do with the accuracy or resolution in interpreting a *digital* signal. The example you give is not correctly figured (see below) and applies to analog I/O's. Analog inputs have analog-to-digital (A/D) converters. In your case it is only 4 bits. In a 4 bit system you can only distinguish a change 1 divided by 2 to the power 4 minus 1, i.e. 1/15. With 4 bits you can only have numbers from 0 to 15. It means that if you have a transmitter that gives 10V at 150 degrees C, 0V at 0 degrees C and connected to an analog input of 4 bits, you cannot distinguish a change in temperature if it is less then 10 C. Commercially available analog I/O's are between 8-18 bits. The higher the A/D convertion bit figure, the resolution is better (and module is more expensive). I corrected your figures as follows: 0000 stands for 0%, which is pump stopped. 0001 stands for 100/15% function. 0010 stands for 200/15% function. 0011 stands for 300/15%, and etc. and finally 1111 stands for 100% motor running. If it is too confusing I recommend you think about these examples about number systems: 345=3*100 + 4*10 + 5*1 (in base 10: the system we are using in daily life) 0011=0*8 + 0*4 + 1*2 + 1*1 (in base 2:binary) Note that the power of the base is increasing in every digit, starting from 0 (x to the power 0 is 1, provided that x is not 0). I hope this will help Hakan Ozevin