Unless you’re living in a cave I’m sure you have heard about the Internet of Things(IoT) or maybe the Industrial Internet of Things (IIoT). We’ve seen a lot more interest in a closely related type of project I like to call distributed data logging. You might be asking yourself what is distributed data logging and what does in involve? I’d like to give you my thoughts on what it is and how it is different than traditional data logging. Also, some ideas on how you might build a distributed data logging system and what the advantages are.
What is Distributed Data Logging?
First off, what is distributed data logging? In my mind, it is collecting data from several sensors or pieces of equipment that are spread out. It could be temperature and humidity sensors spread around a warehouse or other buildings. It could be reading the temperature of a number of processing furnaces in a manufacturing facility. Or, it could be monitoring multiple pumps spread around a wastewater treatment plane.
Distributed data logging is different than a traditional data logger. In the past, there was usually one or a few loggers and these were wired directly to the sensor or equipment. It could be that the logger was just something like a simple single channel USB voltage logger.
In this case, you might have a few or even dozens of loggers spread around. And, when it was time to collect data, you had to bring them back to the office or take a laptop or data collector out to the logger. Or, you might have a multichannel universal input logger that was reading data from a bunch of sensors that were spread around. Here, the challenge was to bring all the wiring back to the logger and not lose accuracy or get noise in the readings.
How Do I Use Distributed Data Logging?
For distributed data logging, there will still be a central logger that records all the measurements. But, the measurements are made using “smart” modules that sit close to the sensor or equipment being monitored. These smart modules digitize the data and send it via one of the standard communications protocols back to the logger. Performing the measurement close to the sensor eliminates or at least greatly reduces the chance of measurement errors. The logger can record the data, process it to make immediate decisions, and generate alarms if something is out of whack.
To send the data there are many standard communication protocols, both old and new. For example, we have used Modbus quite successfully for many projects. Modbus comes in versions that run on serial interfaces like RS-232 or RS-485. However, many of the newer devices use Modbus TCP which run on standard Ethernet networks. These could be wired networks using standard Cat5/6 connections. Or it could be Modbus running over WiFi wireless networks.
What’s New in Distributed Data Logging?
There has been a lot of hype lately about a new protocol – OPC UA/Industry 4.0 – particularly in Europe. It is supposed to make it easier to put together systems from different vendors by creating a standard client-server architecture for data exchange.
If any of you have ever tried to put together a Modbus based system you probably know the pain of working with equipment from different vendors:
- Big-endian vs. little-endian.
- Integer vs float data types.
- Zero or one based register sets.
OPC is supposed to remove all this pain by establishing an interoperability standard. Fortunately, my experience so far is that it lives up to the hype. Once you get the client and server talking to each other, the rest is just deciding what data you want.
Distributed Data Logging vs IoT
Now, you might be asking yourself how all of this is different than IoT? Well, I see IoT devices pushing data to a server somewhere in the cloud. If you want to interact with the data, you must log into the cloud server. This is not necessarily a bad thing, but there are a few disadvantages.
One of the big ones is security. We talk to customers all the time that just don’t want to let the data out of there facility. Telling them that the data is being sent to a server farm in California just won’t do. I am not sure what is so critical about their freezer temperature, but they insist it can’t leave their building. And, this is not to mention that the next “Big One” could take them off-line for weeks. They want the security of having the data stored on a device in their office where they can get to it any time. And, they never have to worry about a competitor obtaining critical information about their unique manufacturing process.
Read on for Part II – Putting together a distributed data logging system.