When I think of Modbus and the future, I am reminded of that old joke that everyone wants to go to Heaven but no one wants to die. I am reminded of that joke because I often think that Modbus is obsolete and won’t last, but then I realize that Modbus will never die.
Yes, the Internet and Web Services are dominating our thinking and our new development. Everything has to be connected to the web. Everything has to connect to the enterprise. Every miniscule bit of data must be manipulated, analyzed and archived.
I agree that this is won’t argue that this isn’t the direction of the future, but I don’t think that means that every photo eye, proximity switch and valve is going to be web-enabled. There are some things that are too low in the automation hierarchy to be of consequence. Who wants to manage enterprise communication and the bandwidth needed to move data from a thousand small devices in a manufacturing plant? How about a million devices in a refinery?
The traditional automation hierarchy will change, but that is going to happen fairly slowly. The Programmable Controller is going to evolve, but it will evolve slowly. Until it does, we will have an automation hierarchy that is substantially similar to what we have today. DeviceNet, PROFIBUS DP, BACnet, EtherNet/IP, ProfiNet IO and the other I/O protocols are not going away. Just as the networks that feed IO to Programmable Controllers like Io-Link aren’t going away.
There is always going to be a place in this automation structure for small, single-purpose devices that support a protocol like Modbus RTU or Modbus TCP. The advantages of using Modbus technology haven’t changed, and it’s still going to be the right choice for years to come:
1.It’s small. You can easily fit it into the smallest of processors. No big RAM or ROM requirements.
2.It’s cheap. You can build it yourself or you can buy it for a very small charge.
3.It works with all processors. Anything that has a UART can run Modbus.
4.It’s simple to deploy. It’s just daisy-chained RS485. Wire from one device to the next to the next or a standard Ethernet network.
5.It has very simple data typing, memory map and operating functions so that most competent programmers can implement it within a week or so.
6.There’s a lot of it out there! You can’t go wrong with a Modbus interface. Everybody understands it and can usually find a way to move the data to where it needs to go.
When I think about this, I come to the conclusion that MODBUS WILL LIVE FOREVER. A big reason for that is the guys who build a lot of automation devices are not network communication experts. They are domain experts in valves, glue controllers, pumps, chillers, drives or whatever. They need a way to talk to their device. Something simple that isn’t going to cost too much and that they can get implemented quickly. That answer for the last 40 years has been Modbus, and I see no reason for it to change now.
Hail Modbus! Modbus Forever!
PS: There a few free, limited copies of my new book, Modbus in the 21st Century. Contact me to grab one before they’re all gone. Email firstname.lastname@example.org.