Bye-Bye, Windows? So Long, Ladder Programming?

Shift in Programming Approaches Push Automation Suppliers and Integrators Away From Hardware to Software

By Hank Hogan

A shift in the approach to programming could be pushed along by a technology change. Increasingly, the automation world will not be Windows-based, but instead will run on Linux, predicts Wade Jensen, CEO of automation integrator Wadsen, Temecula, Calif. He argues that ladder logic is on its way out. As for what will replace it, he notes that Linux is everywhere, most notably as the basis for Google's Android and as the operating system for Raspberry Pi. A $25 single-board computer, a Raspberry Pi can be the hardware behind a very inexpensive controller.

"Right now I'm writing code in Android, and I'm building a little frequency generator out of an embedded controller," Jensen says, as an indication of where things are headed.

SEE ALSO: Programming With Old and New 

In addition to knowing higher-level computer languages, he adds that other skills that
will prove beneficial are proficiency in web design and HTML. Knowledge of how to form-ulate simple queries in SQL and its variants is also likely to become essential, Jensen says.

The arrival of Raspberry Pi is just another example of the forces pushing automation suppliers and integrators away from hardware to software, says Barrett Davis, partner at system integrator AutoMate of St. Louis. In the past, hardware was king because dedicated controllers costing thousands of dollars were needed. Now, there are reliable and inexpensive PCs, and even some systems that cost a few tens of dollars like the Raspberry, that can do the job. "A CPU is a CPU, so it really comes down to software," Davis says. "Now, that's only to a point. Some people think you can do everything with software.

Well, that's not really true. You have to have hardware that interfaces to instrumentation, and is capable of getting you the correct information."

This sidebar is part of the October 2013 cover story, "Old Skills, New Skills."

Show Comments
Hide Comments

Join the discussion

We welcome your thoughtful comments.
All comments will display your user name.

Want to participate in the discussion?

Register for free

Log in for complete access.


  • <p>The Raspberry Pi is a good example of a very inexpensive piece of control hardware. But I think scaling may turn the economics of this small hardware on its head. For a production line with 1000 (or 10,000) I/O points, the engineering costs involved in networking all those imbedded controllers becomes significant.</p> <p>As for Windows and ladder logic, I say good riddance. :-)</p>


  • <p>I would concur that Linux may make some inroads in the control world against Windows and, as M Robbins said, good riddance. BUT the problem that stands in the way of extensive selection of Linux or any Linux based hardware is one of control and version standardization. Today one can buy a Windows version or I suppose an iOS version and be very confident that the behavior of that OS is (reasonably) well documented and very predictible down to the level of fine details. If you are on the scale of a GM or Unilever you might even be able to get some support out of them for the OS. But Linux on the other hand, even a specific Linux version, is not really one OS it is a hundred OS versions that have tinkered and tweaked by numerous people and organizations with little or any documentation. Fine for the small operator who is building one off products or systems, but anathema for any large system or one that has to be able to be redeployed again and again over several years. </p> <p>And as far as ladder logic is concerned, my feeling is that it will be with us for a long time to come. There is simply no other method of depiicting logic that is so intuitively easy to understand and so little subject to interpretation about the fine details of the function of each element. Yes, sometimes a Statement Language or Function Block Diagram format program might be a bit easier and more compact to code, but there are a lot more maintenance people who can follow a ladder diagram than there ar etose who can understand SL or FBD. in addition it is much easier to follow the signal flow when online monitoring a program in Ladder than in SL or FBD. The other programming languages have gained popularity as more and more number crunching applications are being ported to PLCs and PACs and as the programmers who write programming software discover that it is easier to put a complex analog function into FBD or SL format than it is to turn that function into a ladder element. Bottom line: ladder can do discrete logic better than any other method of visualization AND can support complex function blocks, while the other languages have problems with visualization. They may handle analog better than ladder (arguable), but they are much more troublesome for discrete logic. </p>


  • <p>Some were predicting that ladder logic was dying 15 years ago. It will never die. As Kim pointed out, there is nothing else as easy to understand for logical operations. I can teach a maintenance person to proficiently read ladder logic in less than a day. Not so for textual languages. FBD works best for those in process control that are used to loop diagrams. Sure, controllers will support the other languages. I do not advocate using ladder for complex calculations. But, that does not mean it will disappear.</p> <p>Also, in an industrial environment, the controllers are expected to run continuously for decades. I would not trust the Raspberry Pi to be capable of running continuously for even one year.</p>


  • People will use every form of programming available to write industrial control software. VB was going to be dead 10 years ago and people still write industrial HMI software using VB. Ladder has been predicted to die since the day it came out. Java is on the plant floor. The only dead language seems to be Pascal, the favorite language of Computer Scientist in the 1970's and 80's. To make a project work I'll use anything and everything that's available, this man's experience is limited to the industries he's done projects for. Raspberry Pi will be everywhere but it will no more take the control market than Java did, it is for all intensive purposes another dead end on the software evolution tree. A living branch, probably, but a revolution, hardly. I remember when Bernoulli drives were going to replace all other forms of storage, heck I even remember when bubble memory was going to take over. I'll be bold and predict that if we don't have an apocalyptic event that robs us of all technological advances for a thousand years that there will be brand new installations of controls that still use ladder logic then. Nothing competes with the speed of boolean logic, and unlike most other technologies the bootup time is in milliseconds rather than seconds or even minutes, so don't buy this BS this easily and beware all such shade tree programmers that would make such rash statements. While they may have excellent academics remember that to be a industrial control expert all you have to do is say you're an industrial control expert, there's no actual certification. Sad isn't it.


RSS feed for comments on this page | RSS feed for all comments