They say the grass is always greener on the other side of the fence – for people on both sides of the fence. There is a similar truth in the automation profession. If you are a building automation guy, then industrial automation protocols are total black magic. If you are an industrial automation guy, then building automation protocols are the same black magic. (The only thing you both understand is Modbus.) Very few people have made a living working on both sides of the industrial and building automation chasm. Let’s pull back a few curtains and get a slightly better understanding of the differences and similarities in building and industrial automation.
What’s the Same?
All of the leading protocols in both building and industrial automation perform the same basic function. You monitor and control real world inputs and outputs that are a binary on or off, or represent a value such as speed, direction, time or position. The physical layers, transport mechanisms, error checks, data names and data encoding changes but, at the most basic layer, all the data being passed is the same. It doesn’t matter if you are a device in building or industrial automation, you are a set of binary and analog values.
This is the meat of the discussion. The major difference in the protocols is driven by the application requirements.
Industrial Automation Protocols offer a high level of determinism. For safety and quality reasons you can imagine how important every message in a complex application could be. Industrial protocols also can operate at very high speeds and are generally small networks with no more than few dozen nodes.
The industrial protocols are often described as IO protocols and encompass EtherNet/IP, Profinet, and EtherCAT. They move data but offer no built-in data context to go with it. Data context comes from separate reference documents like a schema, EDS or GSD. They forgo making data packets easy to work with, in favor of pure efficiency. You are able to move raw blocks of IO data back and forth at millisecond and sub-millisecond speeds. Speed and determinism dictate the form of the industrial protocols.
Building Automation Protocols, on the other hand, don’t need blinding speed. If a vent damper closes within a few seconds, that is perfectly acceptable. One of the major differences in building automation is the size of the networks. Entire university campuses are often built around a single central control system. The systems can encompass thousands of devices. That means the protocols are built to cross subnets and also to efficiently use network bandwidth. When you have thousands of nodes, a cyclic data exchange common in industrial automation protocols is not practical. Most devices operate with change-of-state data exchange and only periodically update with cyclic exchanges.
Because building automation networks can be so large, device and device data discovery is also common. Discovering and commissioning a device that may be connected a mile across a campus would otherwise be very challenging. A device, its data names, types and values can all be accessed over the network. The additional metadata would be too much overhead for industrial automation.
The building automation protocols live in a middle ground between IO and information protocols. You get data with a bit of meta information as well. The leading Ethernet based building automation protocol is BACnet/IP.
Will there be a Standard Protocol to Unite Them All?
Not likely. No protocol is currently being pitched that takes a green-field view of automation. The “uniting” protocols on the market today all take the pragmatic view that the current market is heavily invested in many different protocols. They understand that it would be extremely challenging to change that.
Instead of trying something totally new, the organizations behind these protocols (who should, in many cases, be competing) sign memorandums of understanding and set up Special Interest Groups to add volumes to each other’s protocol specifications that define how these different protocols should now become interoperable.
There is no fight to the death, leaving winners and less choice for the market. Instead everyone publicly gets in a technology-minded circle and sings Kumbaya.
What Could Shift the Paradigm?
The only evolving technology I believe has the ability to one day change automation would be something like Ethernet TSN (Time Sensitive Networking). A technology that vastly increases the bus utilization and efficiency of Ethernet networks could change the automation world.
This would allow any protocol, including higher-level information protocols, to offer the niche benefits of industrial and building automation protocols.
Traffic could be scheduled and given higher priority to meet the speed and determinism requirements of industrial automation. The efficient bus utilization would allow for extremely large networks, and eliminate the need of protocol-specific network routers common in building automation. Many of the application problems solved with building and industrial automation application layer protocols would no longer exist.
I don’t think this is a reality of the next decade, unseating the status quo takes decades in automation, but the technology is real today. It will be interesting to see how it shapes our future.
If you haven’t heard, the Allen-Bradley 1761-NET-AIC met its end. It was a good run but another step was taken to move customers away from legacy PLC Systems. The problem is many of you aren’t ready to say goodbye to the NET-AIC. You want to milk every last drop of function out of your legacy PLCs, and the NET-AIC is a vital tool to do just that.
Real Time Automation is here to save your legacy systems from obsolescence with the 515RTAAIC.
The biggest surprise in this data was the average age of automation equipment. We have a jaded view given the amount of work we do maintaining legacy equipment. Still the numbers surprised me.
47% of equipment being under 5 years of implies there have been a lot of new projects the last 5 years.
81% of the equipment our readers are working on (and with) was likely implemented after the crash in 08.
What is the average age of your Automation Equipment?
What is your annual capital improvement budget for automation?
Our busiest time of year is:
Your automation business is:
Congratulations, you’ve just been drafted as a professional baseball player! What will your walk-up song be?