As recently as two years ago, when Okuma America’s Brad Klippstein wanted to squeeze some operational data out of a legacy machine tool, he had to connect dozens of wires to the control cabinet and add a PLC. Once he was done, data flowed from the machine tool to the PLC and then on to a hub or server where it’s stored.
“We have things like Ethernet and Wi-Fi and all of this technology we can use nowadays,” said Klippstein, product specialist supervisor. “But equipment from 30-plus years ago did not have that capability, so you had to physically wire a device to the control cabinet to extract all of that data. Now we don’t even have to have a hard connection [with new machines]. You can do things wirelessly. That’s the transition over the past 30-40 years.”
Part of Klippstein’s job is to help customers take discrete steps into the connected, digitized and data-driven world of Industry 4.0—for new and legacy machine tools.
These benefits can include:
Whether transmitted through wires or the air, data collected from factory devices can help a shop owner make smarter business decisions.
With analysis of the data in hand, workers will have increased visibility of shop floor operations, which can help them solve problems, enhance processes and boost productivity.
With one recent advancement, that shop owner may not even need someone like Klippstein anymore for his legacy machine tools.
In 2019, Okuma came out with NET BOX suite-C Quick as a lower-cost alternative to an engineered solution to collect data from older machines. The NET BOX CQ, for short, connects directly to a signal tower and then to a hub or device used to pull in signals.
NET BOX CQ uses three optical sensors—one each for the green, yellow and red lights on the tower—and converts the information they gather to a format that can be read by a computer system. To standardize data collected by any of its software, Okuma uses MT Connect, an open-source, commonly used communications protocol.
“It’s easy to install, and you don’t need an Okuma technician to do it,” Klippstein said of NET BOX CQ.
To start, the ability to see whether a machine is running is a great metric to track because a shop owner usually wants to push every piece of equipment to maximum production. He wants his overall equipment effectiveness (OEE) to be in the 65-85 percent range because a machine is only making money when it is running.
“So, just by pulling that information—green, yellow, red—I’m able to track and trend that over time and at least see when stoppages have occurred and for how long,” Klippstein said. “And then I’d take the next step and say, ‘Why has this occurred and why does it continue to occur every day at 2:30?’ So, now I can trend things. At least that gets me a step in the right direction.”
Christopher Rezny, central regional manager for factory automation provider Fastems, said while MT Connect and Okuma’s software are very good vehicles for data collection, the limited amount of data available from a legacy machine isn’t their only downside compared with modern machines.
“Interfaces from older, legacy machines back to other internal software being used, say an ERP system as an example, are also either very limited or non-existent, too,” he said. “There can also be issues pertaining to the feedback mechanisms from these older machines concerning safety protocols for the user.”
Despite the limitations Rezny described, there’s plenty of data and machines for companies like Okuma and Excellerant Manufacturing to work with.
“The majority of shops have legacy equipment,” said Excellerant Manufacturing President John Carpenter. “Most people come to us because of our history with legacy equipment—knowing those cabinets inside and out—because we’ve literally connected tens of thousands of CNC machines through the years and we’ve accumulated what I would call trade secrets on what’s the best way to connect this legacy equipment.”
The first objective when called to a new customer is to understand what it is he or she wants. Is that person just trying to bring data up from the factory floor so everyone’s aware of what’s going on?
“It’s all about what problem you are trying to solve,” Carpenter said. “And then let’s give you the basics to start with. Most factories haven’t fully appreciated the data that’s there. They’re crawling right now. I will walk into shops and feel sorry for them because they haven’t even gotten to the crawl stage.”
Carpenter recalled one customer where management wanted basic transparency—data from what was happening on the shop floor. And shop floor workers were pushing back against collecting the data out of fear of Big Brother watching them.
Carpenter left the issue for shop management to handle. But he recognized the workers didn’t have a choice. He knew shop leaders were looking for data to measure production and improve processes, not as a Big Brother tactic. Unfortunately, management hadn’t prepared its employees for the move.
“It opens up so many eyes at so many levels but your people have to be ready for the change,” he said.
Once the basics are being tracked—Are we cutting? If we’re cutting, is it at 100 percent? And if it’s not, are we in a reduced production mode?—and the data is being fed into Excellerant’s real-time, machine data and communications platform, it’s possible to add myriad sensors and transducers to get more data.
Sensors can be incorporated to measure temperature, vibration, pressure, flow, concentration and more.
“You have to determine what’s important,” Carpenter said. “We can collect all kinds of data from a point on a shop floor. There’s a vast amount of data when you get into vibration analysis and tool breaks. The list goes on and on.”
There’s always a way to collect more data, but sometimes it becomes cost prohibitive, he said. The shop owner needs to determine how much he wants to spend on a sensor to get his data up to the next level.
“Typically, when you wrap this around some kind of OEE project, it’s a quick ROI,” Carpenter said.
“This is obviously an industry-wide problem, but the questions we’re getting usually fall into two categories,” said Cory Weber, senior industrial IoT architect at Bosch Rexroth. “It’s: How do I increase the quality of my production using IoT or using IIoT? Or it’s: How do I increase my efficiency?
“The other question that gets thrown in there is: How do I decrease my production costs?”
In addition to help answering these production-related questions, Weber estimates 90 percent of customers want real-time visualization and machine tool monitoring, on-site and remotely.
“It’s the first step, right?” he said. “What’s my [automated line or machine tool] doing?”
Bosch Rexroth’s solution for squeezing data out of legacy machines is an IoT gateway called PR-21, which is a small, single-board computer that’s hardened for industrial environments. It also has an IIoT platform, ctrlX CORE, to pull and process data and move it into storage on a local server or in the cloud.
One of the more common applications Weber has seen for using collected data is that of preventive maintenance.
Smart, data-driven tweaks in a maintenance schedule are one way to keep machine tools humming while decreasing production costs by reducing waste.
“I know that even in our own facilities we’ve done things like switch over to hours of use as opposed to [calendar] scheduled maintenance,” Weber said.
For example, Bosch Rexroth has a hydraulic test area where workers used to change the fluid every two weeks. After they put continuous monitoring on the machine they were able to revise the fluid-replacement schedule to one based on use time vs. calendar time.
“When you’re talking 50-60 gallons of hydraulic fluid or more, that can get pricey,” he said.
Other types of fluids have set parameters that must be met for their optimal use, whether based on concentration, heat exposure or other factors.
“If you can test on a regular basis, you can change your maintenance over to as-needed and you can save a lot of money and protect your quality,” Weber said. “It’s an attractive selling point and relatively low-hanging fruit.”
Weber picked up domain knowledge for manufacturing while he was growing up: His dad owns a small manufacturing business. His education, though, is in information technology. Both manufacturing domain knowledge and data smarts are needed to benefit from implementing Industry 4.0, he said.
“You need someone who understands the actual process,” he said. “And somebody who understands the process can look at the data and say, ‘This doesn’t look like what it’s supposed to’.”
Once the data are in hand, the possibilities are endless, depending on how much the user wants to process it.
If a shop owner is fortunate enough to have someone with a real data science background, he can even venture into artificial intelligence.
“It’s a matter of collecting enough data to make it valuable and then, and I can’t stress this enough, knowing what questions you want to ask of the data,” he said.
Weber’s data-related advice, of course, applies to both new and legacy machines. Data mining for smart operations applies to both.
Connect With Us
