Ladder programs for PLCs

PLCs only make sense when you have sequential events that are dependent on previous events, like motion fault detection. From what I have gleaned from previous postings is that your equipment is mechanically timed. In my humble opinion I would concentrate on using electronic trigger technology, solid state relays and motion validation logic (feedback). The value of doing this is first reliability and secondly immediate system fault identification. You should be able to use a common family of components for all your solutions. Steve

machine is using 6 to 8, 3 or 4 pole DT relays, 6 to 8

valves. Should be ideal for PLCs. I need to learn

I've seen or tried is lacking in translating the

stuff is old and doesn't have schematics yet alone a

Thanks! I have different kinds of equipment for different families of products. Most of my latest posts have been for mechanical equipment. The other equipment I'm addressing is all hydraulic and pneumatic actuated. The relay logic controls the valves.

man I hear ya. I used to do A LOT of this stuff. The real key is learning exactly how your machine is working. It can be written up in ladder logic form. I doubt you'll find a tutorial on this. It takes that right combination of person that knows machines and has a lot of PLC programming already under his belt.

I was never a ladder logic writer type. But, i could read and follow it if the programmer walked me through it. So, my approach was to really watch the machine and understand it. Write it all down in english. Then double check the ladder BEFORE taking the machine apart and installing the PLC.

I'm sure you understand your machines. I'd suggest you hire a fella to interview you and write the ladder logic of what you're got. keying it into the PLC is trivial by comparison.

karl

"Buerste" fired this volley in news:hv4d8c$i3d$ snipped-for-privacy@speranza.aioe.org:

IF you decide to use a PLC to do this, you can ease your task by choosing one of the types of controller that incorporates a crippled BASIC or BASIC-like programming language that can run behind the ladder.

Ladder logic is pretty foreign stuff to someone not schooled in thinking in those terms. It becomes a lot easier for some, if the only things the ladder code has to do is detect conditions, and the actual control is done with a more-or-less easy to understand algorithmic language.

I do quite complex hydraulic/pneumatic systems around PLCs; The typical machine has as many as thirty or forty limit and cycle sensing conditions, and fifteen or more directional or proportional control valves. In that sort of environment having a programming language available is a boon. (even if I did have to write my own floating-point library to support the app in an integer-only operating system).

LLoyd

Please remember that if you use solid state relays to drive inductive loads, always use R/C filters across AC loads and back EMF diodes across DC driven loads. Steve

previous events, like motion fault detection. From

mechanically timed. In my humble opinion I would

motion validation logic (feedback). The value of

identification. You should be able to use a common family

products. Most of my latest posts have been for

pneumatic actuated. The relay logic controls the

I learned by designing controls with actual relays before PLCs came out. We didn't document other than the schematic, designers and repairmen were expected to stare at it until it all made sense. Once I understood 3-wire control (latching relays) the rest wasn't too bad.

This is another approach to understanding event-driven process flow:

jsw

Jim Wilkins fired this volley in news: snipped-for-privacy@a42g2000vbl.googlegroups.com:

Another way one can learn is by purchasing a low-cost PLC and using the associated tools.

One family of boards that comes with a compiler and hundreds of programming examples is the TriLogic M-series.

Buy an 8x8 (8 in, 8 out) or a 16x16, and you'll have all the power you'd likely need for such a project, along with all the tools and examples to use the device. This series (as do others like it in other manufacturers' catalogs) has PWM and Stepper control routines either built-in or as free libraries. You don't even need a stepper controller, just drivers. The board can handle it all in real-time while running your application.

Look at the M-series for high-power capabilities; all the way down to the Nano and E-series for smaller aps.

I have a number of these in hot, humid environments. The oldest is over

6 years running now, with never a down moment due to hardware breakdowns. (a few programming bugs, yes )

LLoyd

Where are you located? I do this regularly with pneumatic systems.

snipped-for-privacy@a42g2000vbl.googlegroups.com:

One of the best methods I have found to create a ladder logic program from a "new to me" machine is to create a series of "IF ... THEN ..." sentences on a piece of paper. For example:

IF "Spindle Start" button is pushed AND "Spindle is not running" THEN "turn on spindle motor" IF "Spindle is Running" AND "Spindle Stop" button is pushed THEN "turn off spindle motor"

These types of statements will export very easily to ladder logic diagrams. Keep in mind, these are the basics of PLC programming. There is a world of additional functions that modern PLCs offer... timers/leading-edge events/trailing-edge events/serial IO

_kevin

snipped-for-privacy@a42g2000vbl.googlegroups.com:

I realize those were just examples, but also as examples do it really matter whether the spindle is running or not when you push the buttons? You have to understand which inputs to test and which don't matter for each action.

jsw

- snipped-for-privacy@a42g2000vbl.googlegroups.com:

Jim: You bring about a good point. What matters for each action, and reducing the testing of insignificant inputs, giving you speed during the PLC scan process.

In the examples, I generically associated a "turn off spindle motor" action to a set of generic inputs. This ouptut action may be as simple as "turn off spindle motor contactor", or a very complex process (sub-routine) incorporating VFD braking processes and a restart lockout process.

But still a valid point that you make. _kevin

I tend to use a steping form of logic as in equ 10 then this rung is active and when it is satisfied there is a move command to move the stepper control value to step 20 or so. Gotta leave room for the unexpected additions to logic.

My looking at the stepper register you know where you are and by a write to that value you can move the sequence back to anywhere you want.

Complicated systems, you use a series of steppers and treat sections of the system as small machines with input and output conditions.

Wes

-- "Additionally as a security officer, I carry a gun to protect government officials but my life isn't worth protecting at home in their eyes." Dick Anthony Heller

All the relays in most of the machines are 110vac as are the valves, switches and limits. It's worked for years but we see a lot of advantages to use program tweaks in a PLC. And, all the switches would last longer using a low voltage.

I'm in Cleveland.

when it is

That takes a little more planning up front, but for machines with synchronous motions it makes debug much easier, and is easy to follow when you come back to make a change six months later. I haven't touched a Mitsubishi PLC in a long time, but they had a family of functions they called "step-ladder" that made that style of programming very convenient.

One of my engineer's brother's next-door-neighbor's cousin's buddy does this all the time and sells the hardware too. I'll just look over his shoulder and go for the osmosis method.

I just suggested one of these somewhere elsa on the group. They are pretty cheap and are programmed in a sort of Basic like language

On Tue, 15 Jun 2010 03:26:42 -0400, "Buerste" wrote the following:

Yeah, shouldn't take more than a couple minutes.

Ladder is probably the right language for your job, but I've programmed plc's in function block diagrams, structured text (like basic or pascal), and sequential function charts. I used ladder the least.

Pete Keillor