Arduino/Raspberry Pi PLC's...

[Paullys50]

I am always surprised for the support rPI and Arduino seem to get by professionals for use in the automation world. Sure it's exciting to see a powerful platform at the price point it's at, but let's not lose sight of the fact that it is meant for enthusiasts, students and people looking to learn. Doesn't matter how fast it is, or how cheap it is. It's not an industrial product by any stretch.

To me it's the equivalent of saying an erector set is appropriate for industry use.

 

[leem2209]

How would one prove CE compliance for RPi or Arduino? If you're working on a hobby project or something similar you don't need to. But if you're talking about machinery in a production environment you probably don't like to be held accountable if an accident occurs. Real PLCs have all the approvals needed to comply with the rules. All you have to care for is that your programming also meets the set standards.

BTW Don't be surprised if some day (maybe even today) one or the other PLC appears to be powered by the same processor found in RPi or Arduino. But then again, the PLC manufacturer will provide all the approvals you need...

PLC's have evolved and in the future so could devices similar to that Arduino PLC. And if it's proven to pass all the required tests, then it could become a new competition. It may change the way current PLC manufacturers run their business and they may well have to in order to stay ahead of the competition.

 

[Peter Nachtwey]

I see this argument over and over again. We make motion controllers. We use micro controllers. We don't use an ARM chip like in the Raspberry PI but it would do for a motion controller or a PLC. The Arduino isn't powerful enough. So what magic does do do that makes the ARM chip suitable for a motion controller or PLC? Not much really. The problem isn't usually with the CPU but how it is implemented.
The big difference is that a PLC has firmware that lets the programmer get the job done with few chances of crashing the system. The other item is robust I/O.

We actually have records for every product we ship because we use automated testing now. We have been since the late 1990s when Rockwell insisted on it. We know how things fail and what fails more often.

People blow up I/O due to miss wiring.
The DC to DC converters fail. Sometime we get a bad batch. Other times the bad I/O wiring damages the DC to DC converters.
People connect AC to the 24 volt supply.
Sometimes we get a bad batch of rams. They may take years to fail but we can trace back to the supplier and manufacturer. They may give us $1 per chip but after a few years the warrantee has expired. Tough luck for us and our customers. Samsung screw us and our customers over big time in the early 2000. We don't buy from them any more.
Sometimes the opto isolators fail. The hobby boards usually let the hobbyist connect their own I/O and it is rarely isolated whereas most PLC and our I/O is optically isolated. This is a BIG difference.
The PLC and our motion controller come with firmware that essentially start your project out one the opponents 1 yard line. You just have to go the last yard without fumbling or getting a penalty. Hobby controllers have firmware but unless they have PLC specific firmware the end user must do a lot and the gives the end users plenty of chances to fail.

My point is that there is nothing wrong with the micro controllers on the Arduino or Raspberry PI. It is how they are implement on a circuit board. You can't expect optimally isolated I/O on a cheap hobby board. You can't expect firmware that provides 99% of what you need for you project on a cheap hobby board.

Be realistic.

 

[icky812]

Peter as usual has pretty much nailed it.

PLC's are evolving, just possibly not as fast as the consumer market.

Take into consideration the amount of testing that a PLC must endure to ensure that it is safe right out of the gate. Now add in the UL, CE, CSA, Class 1 Div2, etc. certifications that come along with the PLC.

Then there are the years of testing and development that keep the PLC from crashing every few minutes.

Each of those things adds to the cost of a PLC. But, each of those things adds to the safety and longevity of a PLC platform.

Consumer devices do not go through this same sort of ordeal. If you are lucky you get UL and possibly CE on a consumer device. That is it.

How about support on consumer devices? Can you just call up the manufacturer and ask a question any time you want and expect to get an answer within a fairly short time frame?

Lastly as several have pointed out, what about longevity? Can you expect that consumer device to be nearly identical in 10, 15 or even 30 years? AutomationDirect still sells PLCs that were introduced in the 1980's that look and perform identically to the original. What consumer computing device can you find that with? If you want something more modern, we have that too. I'm not sure you can find any consumer device manufacturer that can make that claim.

There is a lot more to a PLC than bits and bytes and a pretty enclosure. Working for an automation supplier get's your eyes opened as to how much there is that goes on behind the scenes. To customers it seems like PLC's are developed in a day. In reality it takes years to get one to market mainly due to all of the testing and certifications that have to happen.

 

[shooter]

In any PLC is a microcontroller, and that can be a ARM Mega controller, the rest is level converters, like optocouplers or relais. And a internal power supply mking 5 Volts, even the oldest are running on 5 Volts.
and yes i do use PLC and arduino systems like the Nano and the mega.
All the big companies try to keep in business. And they will, as almost everybody started on school with a 6800 or Z80 system. and hey old units have this type of processor in it.
i remember i used a Z80 8255 to make traffic lights in machinelanguage mnemonics and paper and a book, now i use scratch and internet. it is about 100 times faster development.