What is holding us back from seeing a broader embrace of IoT, especially in pneumatics? We asked Mark Densley, head of product management for controls at Aventics Corp., to share his views.
What is holding us back from full implementation of the Internet of Things in pneumatics?
Densley: If you are an end-user responsible for maintaining the machine and keeping it running, the path is very clear. You can have an IoT system provide data from the machine, and ensure that the systems or the components on that machine are working within their normal range. In that regard, the end-users are the ones pushing for IoT implementation.
I think the disconnect is often with the machine builders. There are multiple customers pushing them for IoT solutions so the questions the OEMs are asking are, “How do you do implement that? Who has the product to do that? And what should we be monitoring?”
Pneumatics has had different diagnostic features for some time, so we can sense today whether the power is too low or too high, or whether there is a short circuit in an I/O system that caused some sort of failure. And now we are taking it a step further with our sensors and the ability to use our I/O system to look at the things that are driving the actuators. But I think the thing that people are struggling with is, “I have all this data, but what does it mean? How do we turn that data into useful information?” That is the big stumbling block today.
Component manufacturers like Aventics can work with OEMs to come up with an IoT architecture to make sure that appropriate sensors are in place and are sensing the right things to help keep a machine running full-time.
So, you’re saying the data is already there being collected, but underutilized?
Densley: Yes, so now we can focus on what you can do with that data to convert it into useful information to react to, or even predict failure. For example, the Aventics ST4-2P programmable sensor has been in our product line for years. It measures travel distance and velocity for pneumatic cylinders. We can sense the velocity of the piston in the cylinder. But we can also use that data coming back from the sensors to monitor the cushions and shock absorbers inside the cylinder, to ensure they are performing within a certain specification window. If you interpolate that data, you can come up with a way to ensure the cylinder is performing as it should be, and quickly determine when it is not.
In other words, the capacity to analyze that data and act on it is what you think is being neglected?
Densley: Yes, exactly. You can see the data today in your PLC and with a gateway like the Aventics Smart Pneumatics Monitor, which lets you analyze the data to create usable information to figure out what is going on with the machine.
Let’s look at energy consumption, for example. We can measure the air pressure going into the system. We can measure the flow and correlate that to an event and conclude that when you turn a given valve on you get a certain flow, which means if that flow deviates the next time the valve shifts, there must be something wrong with the system. Maybe the tube has come out or maybe the seals are wearing in the cylinder. You know something is off, but what you do with that information and how you use it is the differentiator as machine builders seek to implement ways to leverage IoT capabilities.
Whose responsibility is it to address that problem once it is discovered?
Densley: Both the OEM and the component manufacturer have the responsibility. Our IoT approach is not “IoT in a Box.” There is some consultative nature to it. OEMs and end-users typically would like to have some additional monitoring to make sure that their machine processes are being controlled correctly. We can work with the OEM to come up with an IoT architecture to make sure that the appropriate sensors are in place and are sensing the right things to help keep that machine running full-time.
It sounds as if the relationship between the various component suppliers and the OEMs is becoming more important than ever.
Densley: I believe so. The important thing is that the component manufacturers outfitting the machine need to talk to one another and understand how the pneumatics come into play, so we can have a true partnership when we talk to customers about IoT. When you get down to the field level, you are going to have pneumatics, drives, controllers, and I/O systems. The key is knowing how the data is passing between those items and making sure that the data gets to wherever it needs to go, whether that is locally on a web server, or out to a cloud system.
Sounds like potential users will need to develop a strategy to address where that information is stored, how it is handled, and how it is disseminated.
Densley: Yes, definitely. Going back to the energy consumption example: the best way to realize energy savings is to turn off the compressor. It sounds very simple. Although we don’t make compressors and we don’t make control systems for compressors, with smart pneumatics we can tell the compressor control system what the demand for air is over the next 12 hours. And from that, the controls can turn off the compressor accordingly, because it may not need all that air. That is just one example of components and systems communicating with each other to affect energy savings.
Are potential users familiar with this higher level of analysis and data sharing?
Densley: I think they are slowly starting to get familiar with the concept. More importantly, it depends on who you are speaking with in the machine-design process. In the past, when we worked with machine designers we discussed things like power and the air connection. Now the discussion needs to involve factors such as network connections, security, and VPN connections. It is more important than ever to include the IT department in the design architecture and discuss what those connections look like and how the data is moving. The security issues are important also, especially to end-users, so you need to involve IT experts early in the design to understand what the data is, how much data, where the data is going, and how it is going to connect. There wasn’t a need to do that in the past.
When the IoT is fully implemented, what does that look like? How will all these things work together to improve the situation for a manufacturer?
Densley: One vision is to enable a highly autonomous maintenance process. To us that means the pneumatic circuit, or ultimately the machine, has 100% uptime and never fails. Obviously, components wear out: valves and cylinders go through their specified life cycles and need to be replaced. But the data is there to predict failure before it happens, so that is where predictive maintenance comes in.
One vision for the use of IoT and pneumatics is to use the appropriate data to enable a highly autonomous maintenance process where the pneumatic circuit, or ultimately the machine, has 100% uptime and never fails.
For example, one application is predicting the performance degradation of things like pneumatic valves. Based on a life-cycle benchmark and B-10 life-cycle credits, we can use the data to track when a given valve will reach 75 million cycles and need to be replaced. The operator can receive a message about predictive failure and replace the valve before production is disrupted.
Using data from sensors, we can also predict when a shock absorber on the end of the actuator is deteriorating by sensing an increase in the cycle speed, even by a few milliseconds. This would trigger an alarm or even send an automatic email to the component supplier that maintenance needs to be completed or that the component is ready to be replaced. The system could even generate an order and a new component would be shipped to the customer automatically so the installation could be scheduled to minimize any production downtime. The root of it is to use the data to provide this information to ensure that machines never fail, that they are self-diagnosing, and that end-users know what is occurring and the machine just keeps on working.