Why use remote I/O?

Let me preface this by saying, I completely understand the RIO "bricks" that many vendors are making that use M12 connectors and are washdown rated / dust-tight, etc... Those have an easily defined use case.

I'm asking why anyone would use DIN rail mounted RIO modules from well... anyone, when small PLCs are cheaper. To put it another way, what makes RIO modules so expensive?

Take this module from TURCK:
https://www.alliedelec.com/product/turck/fen20-4dip-4dxp/70348217/

Why would I pay $300 for a module with 4 inputs and 4 outputs when I can buy a full blown PLC with an Ethernet port, and analog and digital IO for about half? Even the new Stride I/O offerings from traditionally low cost AutomationDirect seem out of wack.

Using PLCs as RIO seems to be the better choice; cheaper, and you get some local "brains" if you need to have some local logic as well.

Thoughts?


-rpoet

Cost
The labor to install the equipment
The average labor cost is about $80. Per hour plus the cost of all the materials it takes a lot of time to run the hardware and pull the wires. While if the remote I/O is a few hundred feet away the while the sensors and output device will cost the same no matter where they are terminated in the main panel or in a remote panel. Reducing the size of the main panel is a cost savings in both material cost the labor
If you have a large number of IO over a large area the cost savings could add up quickly. It’s easier to handle a number of small panels then one large panel and if you have the armor point they don’t even need a panel just mount and connect.
The other thing to consider is if you have a large number of IO you may not be able to put them all in a single local rack so you have to add another rack or remote IO

GaryS said:

Cost
The labor to install the equipment
The average labor cost is about $80. Per hour plus the cost of all the materials it takes a lot of time to run the hardware and pull the wires. While if the remote I/O is a few hundred feet away the while the sensors and output device will cost the same no matter where they are terminated in the main panel or in a remote panel. Reducing the size of the main panel is a cost savings in both material cost the labor
If you have a large number of IO over a large area the cost savings could add up quickly. It’s easier to handle a number of small panels then one large panel and if you have the armor point they don’t even need a panel just mount and connect.
The other thing to consider is if you have a large number of IO you may not be able to put them all in a single local rack so you have to add another rack or remote IO

Click to expand...

I understand and agree with the need for remote IO. It certainly has advantages, especially with large machines. One Ethernet cable is much easier to run than dozens of (or more) individual conductors.

What I'm asking is why RIO is so expensive, compared with a roughly comparable full-blown PLC with a similar point count, used as remote IO.

I can get a CLICK PLC with an Ethernet port for about what I'd pay for a Stride RIO module, and I get a basic HS counter, local logic capabilities, etc... What purpose does RIO serve if it costs more and does less?



-rpoet

Each remote IO has as many or more components in it as some small PLC's have
CPU, communications, memory, opto isolated inputs and outputs power supply
they charge the price you are willing to pay.

I use Omron Ethernet I/O all the time. 16 inputs about $80 AU - 16 outputs not much more. 8 analogue inputs around $340 AU and 4 RTD inputs about the same. I even use them in the switch board alongside the processor. I very rarely use small PLCs - minimum I/O is usually 100 and 6 analogues or something like that - then RTDs for heating control. I had a look at the Click with the Ethernet port and it is cheap - so are the I/O but do not suit most of my jobs.

Generally, if you have a failed remote I/O module, you just replace it off the shelf (and set an address if necessary). If you have a failed remote CPU, you have to get out the software to download it.

Cost isn't necessarily an incentive.

If you just spent £2k to 4k on a main PLC, you don't really want a network of smaller PLCs as remote IO Stations to save a few £100 here and there. At least not if you're sane.

Having say 10 remote IO stations, all of the same, with no programming involved makes a spare parts list easier to maintain and it means replacing failed components much easier. It can be done by the onsite staff quite easily and doesn't require a laptop.

janner_10 said:

Cost isn't necessarily an incentive

Click to expand...

Exactly, and sometimes you have no option. Try to sell a Koyo Click to a Nestle site and see how far you get. Additionally, cost is subjective. As an OEM I'm not likely to pay what most other people are paying for different various products.

Are there any smaller plc's / shoebox plc's can integrate into a safety network?
Safety I/O has been my biggest drawcard towards RIO

rpoet said:

I'm asking why anyone would use DIN rail mounted RIO modules from well... anyone, when small PLCs are cheaper.
-rpoet

Click to expand...

Just imagine trying to support a large system that is using small PLCs scattered through out. I'm wrapping up a project with 15 control panels. 1 PLC and ~20 remote PointIO and AS-Interface gateways. If this were small PLCs everywhere that's 15-20 PLC programs to support. That's a nightmare, and controlling via plc to plc comms isn't ideal at all. No reason to hop through a plc to control IO.

Years ago when I was green an OEM supplied a system using WinPLCs from Automation Direct. 7 of them on 1 system! Plus the HMI had control logic! This thing had a total of 8 PLC programs, it's no wonder why the OEM struggled for months on end getting it up and running. I got pulled into it, and at the time, being green thought I was just stupid for not being able to comprehend what the **** was going on.

I wish I could go back and see what happened to that one.

As Paullys50 points out, there is value in having all the control functions encapsulated in a single point. The designer doesn't need to map out where specific control functions are located and passing information from one functional section to another is MUCH easier when it is all locate in a single plc.

Then as many others have said, you have the maintenance point. Swapping out faulty I/O without having to program anything carries a certain value.

The "why does it cost so much" argument comes down to packaging and target use I think. You use the example of a Click with built-in I/O and an Ethernet port. You can pack more stuff into and share more resources in a unitary plc than you can a comm adapter with add-on I/O. Also, most I/O systems are intended to be protocol independent. I can swap between protocols just by changing the comm adapter. Finally, in most cases, you wouldn't use remote I/O on most systems that could support use a small plc for control. So, as janner_10 said, the cost of remote I/O relative to the control system as a whole is relatively small. That means the engineering cost of designing a system with distributed programmable devices seems more than using a purpose-built I/O package. Remote I/O is sold at a premium to reflect that.

Keith

I've always used a ton of remote IO. So much easier in every way than with long runs of cabling.
But my prolific use might have something to do with me only paying $100 for a remote IO station with integrated 4/4 I/O and a 2-port EtherCAT switch. I don't know why yours are gold plated to warrant a $300 price tag.
Customer want's another four buttons added to that small panel way over there? Let me just purchace a $20 input card and throw it at my already installed coupler.

If I need any remote I/O I will use a PLC.

I get 4 analog inputs, 1 analog output, 7 relay outs, 8 digital inputs for $189.00. With ethernet!

If you need I/O that is distant but needs to be tightly integrated and highly responsive to the main controller, then Remote I/O is the only sensible solution.

I use Profinet..
I use a couple of these on my furnaces:
16 I/O that can be either input or output...
https://pdb2.turck.de/en/DE/products/0000000a0003f69c0004003a


And use 1 of these in a control panel that has lights and PBs on the furnace:
Also 16 I/O that can be either input or output...
https://pdb2.turck.de/us/DE/products/00000030000128680002003a

So I taking care of 90% of my field I/O wiring using a simple Profinet cable, that bounces across these modules
I just have to run the 24VDC and a Profinet cable to the modules.
That is a lot of wiring that does not have to go from the control panel to the equipment.