PLCs or microcontrollers?

Inside

Member
Join Date
May 2015
Location
San Antonio, TX
Posts
1
Good afternoon!

Small bit of background: My company manufactures small assemblies with integrated solenoids and pressure transducers. To make sure that the assemblies are properly assembled and that the solenoids & transducers work, we stick them into a test stand which pressurizes cavities within the assembly, activates the solenoids, and checks the readouts on the transducers.

An example of a test:
1) Turn on an external solenoid (built into the test stand) to pressurize a cavity from a supply chamber.
2) Check that the transducer for the cavity reads the external pressure
3) Shut off the external solenoid; turn on a venting solenoid which vents the supply chamber.
4) Check that the pressure within the cavity doesn't drop.
5) If it drops, then display a list of possible causes: (ie: fittings leaking, check valves leaking, transducer seal is leaking, external solenoid is leaking, etc).
6) If it doesn't drop, vent the chamber then move on to the next test.

The problem with the current test stand is that we need to build several more of these test stands and the current one is poorly documented (okay, not at all documented), takes a custom board that someone designed 10 years ago (also no documentation), and is basically unmaintainable.

I've decided to look into using PLCs (specifically the Click series because they seem to be economical) to replace the test stand because (hopefully) we can use more standardized components and have everything be better documented.

My question is - is this not too complicated for PLCs? Everything that I've looked up regarding ladder logic is that it's not really suited for complicated logic with many conditions such as this. Or is the whole point that I use a laptop to handle all of the logic & operator interfacing to do all the actual logic and then just use the PLC to drive the relays and handle the analog inputs?

Or should I just use a microcontroller (I only have experience with consumer-grade microcontrollers like arduino + clones) with some relay boards to handle all of that logic? What are the pros & cons of each?
 
Everything that I've looked up regarding ladder logic is that it's not really suited for complicated logic with many conditions such as this.
Strange, everything you will find on this site points to the opposite direction, that PLCs are extremely well-suited for handling complicated logic with many conditions.

There are so many PLC models from which to choose that it is almost certain that one or more can do your job. If your test has many steps (hundreds), then you probably need a PLC with a built-in sequencer function. With only 6 steps (or a couple dozen) then almost any PLC can do the logic while standing on its head at the same time.

What you will need in addition to a PLC is an operator interface (a graphic screen with maybe some control buttons or a touch screen) showing the test progress and having control switches for the operator to start, stop, reset, restart, or go to a certain step in the sequence. The PLC does all the logic and turns the control devices on/off and monitors the pressure, and displays the pressure on the operator screen if desired.
 
Last edited:
1. >If it drops, then display a list of possible causes: (ie: fittings leaking, check valves leaking, transducer seal is leaking, external solenoid is leaking, etc).

That's the function of an HMI, not a PLC.

PLCs and their HMIs that display readable data are usually different products that 'talk' to each other.

2. >Check that the transducer for the cavity reads the external pressure

That sounds like youR PLCs I/O will need two analog inputs, one to read do what you've written, the other to read the external pressure so that the first has something to compare to.

I'm not familiar with the brand you mentioned but take analog inputs into consideration when you shop.
 
Yes ladder logic is suited to simple tasks, and high level languages such as C are more suited to complicated things. However what you want to do is not complicated. It can easily be done in ladder logic. I'm fluent in ladder logic (a little Structured Text) and many high level languages (C#,Python, C, C++, JAVA). Unless you are very familiar with high level languages I definitely do not recommend you go that direction.

I've never worked with the click series before. I've done a lot of work with the Koyo DL05 and DL06s. This could easily be done with them and they are very simplistic.

If money is a major factor you can look at Maple System HMI they are very good for the money.

If money is not a factor I'd go with the Wago 750 series they call them fieldbus controller but they are full fledged PLCs. they support all the IEC 61131-3 languages.
 
Arduino is good enough for this as it is not commercial, but mere a testrig.
So you would need some transistor outputs for the solenoids, a few for some leds indicating the problem on a board.
and an input pressure sensor , depending on the pressure used you can even use a waterhose with some black water in it and a photosensor, you can detect mBars with ease.

and even the smallest PLC can be used , the only point is how to see the pressure change? as many of these small units have no analog input.

You could use a differential pressure so you can see even very small changes.
 
There is virtually no automation task you can imagine that is beyond the capabilities of a modern PLC.

I changed to PLCs from micro-controllers and single board computers in high level languages for my specialized process control systems many years ago. I have never regretted that change, which was driven by many factors:

- PLC programming languages, particularly ladder logic, were simpler to learn than high level languages

- Many more people were available with the ability to work in PLC programming than high level languages

- PLC programming packages came with many monitoring and diagnostic tools built in, making debugging and trouble shooting much simpler and faster

- Operator interface configuration and integration was much simpler than with high level languages

- PC based SCADA and monitoring was more readily available off the shelf

- Customer acceptance with a PLC was virtually guaranteed

In today's world I would only use a micro-controller for a high volume embedded control application where the economies could (maybe) offset the higher cost of engineering and programming with lower component cost.
 
Last edited:
I agree that higher level languages might be better suited for complicated tasks, but when somebody tells me that something is too complicated for ladder then my answer is very simple:

"you need one or two things to change:
  1. buy a better (bigger/faster/...) PLC AND/OR
  2. hire a better programmer"
And sorry if I'm talking to the programmer in such a case.
 
The Click is an excellent choice. Go for the one with analog input. Software is free. Just one unit being tested, or multiple? That determines your HMI size. If it's just a single item being tested, then you would have Start, Stop for input buttons, and Pass, Fail, and pressure reading. Could do that on a 4" display.
Buy a 24DC supply, breaker, and some terminal blocks. Snap it all on a DIN rail.
You could build the entire system in a few hours.
Consider same thing with a microprocessor. Relay board, Touchscreen module, 5 volt DC for microprocessor, 24DC for valves. Enclosures for each. Mounting, drilling. Many items are not rated for industrial temperatures, or electrical noise.

If several test stands is under 50 or so, then PLC is the best choice.
 
I can think of a few. Motion control for one. Another is position/force control.

Change "virtually" to "almost" if you wish. However, there are PLCs with motion control capability and modules and separate controllers that can be integrated with a PLC system. Certainly a dedicated application specific controller might be better for high end applications, but these tasks are done with a PLC too.
 

Similar Topics

I'm at a new job as the sole control engineer and trying to piece together where my predecessor left off. One machine I'm trying to get online...
Replies
2
Views
103
Or there is an alternative to this? I suppose there are many but I've only seen put/get Can you for example set up a new plc and connect it to an...
Replies
5
Views
306
I have a client who periodically experiences network communication issues. Sometimes when I VPN into the site, their SCADA systems will flash comm...
Replies
2
Views
182
Hi, Seeking consultation on an implementation matter, and have a question about Modicon Compact 984 communication through RS485: Three Modicon...
Replies
4
Views
225
Back
Top Bottom