Raspberry Pi - industrial enclosure

I actually just used one of the Raspberry Pi 3 boards as the driver for a plant information display. (Not a control system I know, I don't think I'd ever consider it for any kind of machine control). The little unit is cost effective when you compare it to putting out a PC in a messy industrial environment, maintaining a windows install/licensing/etc.

For what it does, serves up a custom webpage from a Micrologix 1400, I think I got good value for my money and time.
 
I actually just used one of the Raspberry Pi 3 boards as the driver for a plant information display. (Not a control system I know, I don't think I'd ever consider it for any kind of machine control). The little unit is cost effective when you compare it to putting out a PC in a messy industrial environment, maintaining a windows install/licensing/etc.

For what it does, serves up a custom webpage from a Micrologix 1400, I think I got good value for my money and time.

I agree. For something like HMI or casual data collection, it'd probably be a good fit.

But there's no way I'd rely on it to run anything.
 
Also, I just want to say that using the Raspberry Pi's higher clock speed as an argument for why it's better than a PLC is like arguing that because a Ferrari can go 200 miles per hour, it's the better than your Toyota for going to the grocery store and taking your family on vacation.
 
PLC from the likes Rockwell and Siemens consist of a PLC platform running on top of micro-controller; you just don't really see the micro-controller unless you look in deep. In this respect, running Codesys on a Pi is no different except for the fact that it is made clear that the PI version is not intended for field deployment.

I'd argue running an open-source OS, on a consumer grade microSD card is a substantial deviation.

As for longevity and supportability: I think that Codesys and IEC61131 are going to be with us for many years to come and you can use standard I/O modules on Modbus etc.

You make the assumption that Codesys will be supported long-term on the rPI.

The chances are very high that you will already have similar micro-controllers embedded in devices in your home already: Wireless Router? Satellite TV? Media streamer?

All of which are obsolete after 2-3 years, and only temporary until something new comes along. rPI will continue to evolve and today's rPI won't be the same as 1,3,5,10 years from now. While the rPI is popular, it will evolve and change based on consumer desires, not industry needs.

...is like arguing that because a Ferrari can go 200 miles per hour, it's the better than your Toyota for going to the grocery store and taking your family on vacation.

When you put it like this, I really really really want the Ferrari. I really do.
 
When I started this thread I wasn't really expecting the kind of responses that it has attracted. I'm not advocating that people should replace their industrial PLC's with a Rasberry Pi and nor do 3S. I merely thought that forum members might be interrested to know of the new enclosure from Phoenix - I know I certainly was (but not to replace a S71500 PLC or a logix 5000 system).

I'd argue running an open-source OS, on a consumer grade microSD card is a substantial deviation.

Last time I checked, Step 7 CPU's run on a Linux platform -not sure about others.

You make the assumption that Codesys will be supported long-term on the rPI.

No, I don't. I make the assumption that Codesys will still be supported and as you can use off the shelf I/O modules then one presumes they too would be supported. When 3S released the Codesys control for PI I believe that the intent was that people could develop a system cheaply and then buy an industrial strength 3S Codesys CPU for deployment.

All of which are obsolete after 2-3 years, and only temporary until something new comes along. rPI will continue to evolve and today's rPI won't be the same as 1,3,5,10 years from now. While the rPI is popular, it will evolve and change based on consumer desires, not industry needs.

Very true. If in 5 years my homemade garden sprinkler system doesn't work, I'll make a new one.

When you put it like this, I really really really want the Ferrari. I really do.

I want the Ferrari too! Then I could sell it and buy something that I actually like.

Nick
 
Last time I checked, Step 7 CPU's run on a Linux platform -not sure about others.

I'd never heard that about the S7 PLCs, but it wouldn't surprise me if it were true for their newest lines. At the least, I've been told it's true of the Siemens CNCs and motion controllers.

I've heard Logix PLCs are built on VXworks running on a standard Intel CPU. It isn't exactly open source, but its used just about everywhere in the embedded world. It definitely isn't some custom chip/firmware like PLCs probably were 30 years ago.
 
As the engineers get younger and ladder logic (which is antiquated and hated by recent graduates) is phased out, there will be more C/Java/Javascript applications written that will control IO.

The big names in PLCs will fight this for years with their proprietary hardware and software but it is coming. The arduino and Pi applications will just have to be deployed on industrialized hardware, which exists.

IOT is the future of industrial automation and there is no room for proprietary and antiquated technologies in the IOT market.
 
As the engineers get younger and ladder logic (which is antiquated and hated by recent graduates) is phased out, there will be more C/Java/Javascript applications written that will control IO.

The big names in PLCs will fight this for years with their proprietary hardware and software but it is coming. The arduino and Pi applications will just have to be deployed on industrialized hardware, which exists.

IOT is the future of industrial automation and there is no room for proprietary and antiquated technologies in the IOT market.

You can't beat ladder when it comes to ease of understanding by a wide range of skill types. (Not including Function Block or Sequential Flow Chart styles, I mean compared to more "code" based programming languages).

I don't think Ladder and "IOT" have to be mutually exclusive (or that they are even to be used to describe the same set of things. Apples and Oranges type deal). And I don't think Ladder has to be considered Proprietary or Antiquated.
 
Last edited:
Did someone say longevity?

directlogic_305_plcs.jpg


Still making and supporting them in 2016 while offering updated CPUs that are backwards compatible. We have a few of the GE I/O modules floating around out there in machines on the floor, chugging along to this day.
 
...When 3S released the Codesys control for PI I believe that the intent was that people could develop a system cheaply and then buy an industrial strength 3S Codesys CPU for deployment.

I think the true intent was to create a dirt-cheap platform so they could get Codesys into higher-educational institutions and influence the next generation of Controls Engineers. Part of why Rockwell is so big in the USA is you can find their PLCs in classrooms all across the country, expose students to a platform and guess what they will lean towards once they graduate.

It certainly helps that current engineers have a cheap platform where then can invest in education for themselves, something I have done with the rPI and Codesys. It's also a nice platform for "Proof-of-Concept" exercises before investing into a project.

My comment was probably more motivated due to Shooter's post, engineer's who just see the $$ savings and willingly look past the intent of Codesys on a rPI are blind and won't be doing anyone any favor by putting it into a production environment doing actual control.
 
I do agree fully with paully, yes a simple Controller like RPi or arduino is very good for training, and learning codesys,
On the other hand a BIG PLC also just runs on the same type of controller.
and yes i do agree ladder is easy understandable, however all new PLC run on compiled programs so the program is way to fast for a ladder.
 
however all new PLC run on compiled programs so the program is way to fast for a ladder.

I'm curious what you mean here. Maybe I'm just misunderstanding, but how could code be "too fast" for ladder? Once you compile it, doesn't the code run at the speed it runs at?
 
I meant, you cant see what happens anymore, in old days you could almost hear the relais clicking and seeing on the screen.
 
I meant, you cant see what happens anymore, in old days you could almost hear the relais clicking and seeing on the screen.

You still can. That has little to do with the Plc processor speed and more to do with the speed of the process itself. How long an input or output is on, not how fast it is scanned.

I have a brand new top of the line GuardLogix system and you can see the ladder status no problem.
 

Similar Topics

Good Day, Though I have nearly 0 experience with the raspberry pi, I hear alot and read alot about it and its functionality. I know this has...
Replies
11
Views
3,140
Has anyone ever tried to integrate a Raspberry Pi to a Fanuc robot? I have this idea that I want to try out but I need to write integers from a...
Replies
4
Views
1,557
Hello, I have been getting more and more requests for Real Time Clock (RTC) sensitive automation from my clients. I know that there is a Modbus...
Replies
13
Views
2,758
Hello! I have seen some stuff with people using a raspberry pi running some sort of open PLC software to run some test examples. Has anyone...
Replies
25
Views
6,554
I have a customer who has 20 of our systems using PanelView Plus 7 Performance Series A terminals. With how hard it is to find printer nowadays...
Replies
4
Views
1,625
Back
Top Bottom