I am looking for opinions anyone has using the different flavors of Modbus TCP/IP remote I/O. I have as system that is currently using Momentum InterBus S I/O. This is on a Square D Sy/Max based system using an NR&D IBSM module in the Sy/Max rack. I am looking into the possibility of replacing the IBSM with a EPE5-T and using Modbus TCP/IP I/O. I have seen this type of I/O from several manufactures. (Beckhoff, Turck, Moxa, .....) I am just looking for an reviews for any of the Modbus TCP/IP I/O. I may make the most sense to use Schneider Electrics Momentum I/O with the Ethernet top hats. I issue I have with that is the managers will see it as the same thing they already have.(looks similar to the InterBus S I/O)
Modbus TCP/IP Remote I/O
- Thread starter niteman9
- Start date
tragically1969
Member
Schneider Advantys STB is a slice I/O and much more up to date than the Momentum, works fine.
LadderLogic
Member
I could be wrong but I don't think Modbus/TCP protocol is fast and deterministic enough to be used in an I/O system.
Saffa
Member
You may be wrong indeed. It really depends on what you need to do with your IO, but for many applications this approach is perfectly fine. I scan my remote IO every 60ms which is more than fast enough for my applications.
+1 for the STB remote IO. It's quite cost effective compared to remote rack IO from Schneider and when you eventually upgrade your PLC (to something like a Schneider M340 or M580) you can easily configure the STB island from within Unity.
When it becomes less effective is when you have lots of analogs as the analog density of the io cards isn't that high
+1 for the STB remote IO. It's quite cost effective compared to remote rack IO from Schneider and when you eventually upgrade your PLC (to something like a Schneider M340 or M580) you can easily configure the STB island from within Unity.
When it becomes less effective is when you have lots of analogs as the analog density of the io cards isn't that high
CapinWinky
Member
B&R has a Modbus TCP bus controller for their X20 line (if you're familiar with Schneider's TM5 IO, that is a private label of a small selection of X20), which will give you access to any and every kind of IO you could want. Speeds easily under 10ms.
http://www.br-automation.com/en/products/io-systems/x20-system/#bus-controllers
It programs with either Automation Studio (not free) or Fieldbus Designer (free download), and you can configure a rack without buying anything to see if will do what you want.
http://www.br-automation.com/en/products/io-systems/x20-system/#bus-controllers
It programs with either Automation Studio (not free) or Fieldbus Designer (free download), and you can configure a rack without buying anything to see if will do what you want.
AutomationCtrl
Member
Advantys OTB
Another option in the Advantys line from Schneider is another range called OTB.
It is more cost effective than STB but is applied more on OEM applications. There are options of Modbus TCP, CANopen and Modbus serial islands.
Another option in the Advantys line from Schneider is another range called OTB.
It is more cost effective than STB but is applied more on OEM applications. There are options of Modbus TCP, CANopen and Modbus serial islands.
tragically1969
Member
What a strange statement, care to quantify it ?
tragically1969
Member
STB works out more cost effective than the OTB in the UK for some reason......
CapinWinky
Member
What LL is talking about is a lot of applications require IO feedback to come in at short predictable intervals and TCP/IP cannot be guaranteed to arrive at any set time, so you'll be getting your IO feedback at varying intervals that are completely at the whim of standard TCP/IP network traffic. There is the option to run Modbus as UDP which will make it pretty much the same performance as Ethernet/IP IO.
A deterministic protocol would be something like Powerlink, EtherCAT, ProfiNET, Sercos3, etc. They have a set cycle time and you will get the data at exactly the same point every cycle (with jitter usually in the single digit microsecond range).
A deterministic protocol would be something like Powerlink, EtherCAT, ProfiNET, Sercos3, etc. They have a set cycle time and you will get the data at exactly the same point every cycle (with jitter usually in the single digit microsecond range).
V0N_hydro
Member
I dislike the Advantys line because of the fiddly plastic snap together backplane. Too often I have had a bad connection where modules join, or a module/backplane that didn't go in straight. Once they are together and working it is fine though.
I prefer using M340 everywhere. I believe the modules are cost-competitive with Advantys if you compare for the same density and you have the option of putting logic right on the rack, which can be helpful for programming what to do if the rack loses communications with the main CPU.
Then you also have the option of using true "remote IO" with CIP instead of "distributed IO" with modbus tcp.
I prefer using M340 everywhere. I believe the modules are cost-competitive with Advantys if you compare for the same density and you have the option of putting logic right on the rack, which can be helpful for programming what to do if the rack loses communications with the main CPU.
Then you also have the option of using true "remote IO" with CIP instead of "distributed IO" with modbus tcp.
tragically1969
Member
Not sure i agree with that in full, an M340 with Rack, PSU, CPU (Rthernet) and 64 digital in would be three to four times the price of the same STB setup, that gap will never close regardless of how many IO you throw at it.
I agree that M340 is better if you want sometng bit more robust, especially if you need some local processing and the screw terminals are better but remote I/O was the question.
V0N_hydro
Member
Apparently I've come to believe my own lies to justify moving away from Advantys.
For a remote IO rack with 32 DI, 8 AI it might cost $2400 for the M340 solution and $1600 for the Advantys.
The bigger your rack the smaller the difference as a % of the total though, so if you end up with a $5000 rack with Advantys or $6000 for M340 (say with 132 DI, 32 AI, 1 DO or so) it is only a 16% increase in cost for a dramatic increase in quality.
For a remote IO rack with 32 DI, 8 AI it might cost $2400 for the M340 solution and $1600 for the Advantys.
The bigger your rack the smaller the difference as a % of the total though, so if you end up with a $5000 rack with Advantys or $6000 for M340 (say with 132 DI, 32 AI, 1 DO or so) it is only a 16% increase in cost for a dramatic increase in quality.
tragically1969
Member
Hmmmmm your CAD pricing is way off UK, for STB in the UK you could knock a 1 off the front of the Advantys price (still working in CAD)
Similar Topics
Hello,
Does anyone have a recommendation for a Modbus TCP remote IO. It pains me to have to use one. but yeah.
Just standard remote IO with like...
Hi All,
Trying to help a client who has a project that was started with Schneider STB remote I/O, but they are still waiting on the STBNIP2212...
I have a PH meter that I am trying to bring its data into 1756-L81. I have downloaded the Rockwell MODBUS AOI kit, but I am not sure if I need to...
Hello gentlemen,
Im working on a small project on TIA Portal, about establishing a Modbus TCP connection between my ET200SP plc and a socomec...
If a device has Modbus RTU over serial and Modbus RTO over TCP and Modbus TCP then there is a difference between Modbus RTU over TCP vs Modbus TCP...