Where Can I learn to write programs from flow charts ?

Hi,

I have most of the basics down for the instructions of PLCS but I need to learn how to write programs from flow charts using the transition and function logic. I found a book by Hugh Jack and I would like to find more resources. Does anybody recommend and good reading that will help me develop programs?

Thank You

Hello Austin;
What you describe resembles IEC 61131-3 language SFC (Sequential File Chart), which uses the transition and step (function) logic.
Have a look at a presentation of the language here:

http://www.61131.com/sfc.htm

An implementation from Siemens is called S7-Graph (its name indicated it refers to the original development of sequential controls for PLCs, named Grafcet).
Here is a link to the manual, so you can have an idea of the programming:
http://support.automation.siemens.com/WW/view/en/1137630

Hope this helps,
Daniel Chartier

Try searching for "State" and "Transition". There is some good info out there.

Nick

This may be of some help:

http://www.youtube.com/watch?v=rCa_U6YATTA

It uses LogixPro simulator to go through exactly what you describe.

That is a great video on youtube. This is exactly the kind of programming I am trying to learn. Why did he not have separate state and transition logic? Where can I learn more on programming methods? Is this how all of you tackle programming?

Thanks again

Archie

Is the flow chart diagram in that video link made with visio or something else?

I noticed with this type of programming two output states are active for one scan. Is this ever a problem and how do you get around it

Is the flow chart diagram in that video link made with visio or something else?

Click to expand...

It seemed to me a bit like OpenOffice.Org Draw.

Just wanted to thank everybody again it makes sense now! I did the logixpro silo simulation following the you tube video that Archie posted. I noticed that the product would still build up on the floor. After putting the simulation in single scan mode I noticed that two outputs were on at one time for a single scan. After looking at some of the comments on Youtube it appeared others noticed the same problem. I wrote another ladder logic with separate transition and state logic's. Now only one output is activated at a time and the pink goo doesn't build up on the floor in the simulation

http://www.youtube.com/watch?v=FZpE0dCKu9g&feature=channel

If anybody gets a chance to watch this video on youtube regarding ladder logic, is there an error at 23:50 minutes into the video? It seems like the two transitions D and C should be in series

Or in Archie's states another could have been added - 'waiting for flow to stop' - before restarting the belt. One of the tricks of writing for real world devices is that turn-on and turn-off take actual time and aren't instantaneous.

bernie_carlton said:

Or in Archie's states another could have been added - 'waiting for flow to stop' - before restarting the belt. One of the tricks of writing for real world devices is that turn-on and turn-off take actual time and aren't instantaneous.

Click to expand...

Yup. I write state logic (using ladder) almost exclusively. I always think of things in terms of "logical state," which applies after one scan, and ""physical state" which is going to take a while to happen and catch up with the logic.

Often it doesn't matter. On systems that have motion - like a cell I have that uses an X-Z gantry to move product around the cell - it's often critical. That is, the logical state has just set all the flags and the output to move a load, and that's what has to be used to IK other actions that can't occur until the load moves from A to B. If I waited for the motion controller or drive or whatnot to change it's state, too, well there could be a lot of collisions.

I guess the moral is that you can get yourself in trouble no matter what programming system you use!