Smart sensors, edge computing and AI are practical tools that boost uptime and cut costs for the future.
By Josh Cosford, contributing editor
Fluid power is an interesting industry. On the one hand, enormous mills are using fifty-year-old hydraulics while producing hundreds of thousands of tons of steel per year. On the other hand, new factories are popping up with connected technology providing real-time data and analytics on every parameter from filter health to millisecond-accurate pressure curves.
Smart technologies revolve around information and its decentralization. Despite the proven hydraulic technology that may still allow an old steel mill to turn a profit, smart technologies are simply becoming too good to ignore. So, if you’re a stakeholder within any industrial environment operating hydraulics, keep reading to see how Industry 4.0 is changing our Fluid Power World (pun intended).
modern manufacturing relies on connected technology to ensure uptime and reliability while increasing efficiency.
The Industrial Internet of Things (iiot) describes the interconnected nature of machinery and components used within the industrial space. Machines now may connect to the internet, much like your TV, thermostat and, for some of you, the fridge. I get it; the whole smart fridge thing didn’t pan out as we all hoped. We expected everyone’s food supply to automatically replenish when Maytag sent orders to Karen through Uber Eats so she could leave organic Kefir or Flamin’ Hot Cheetos at your doorstep.
Keep in mind that a typical family home with 3.7 persons can run more dynamically than an automotive assembly plant producing 1.3 cars per minute. Let’s be honest — your home won’t shut down because you ran out of oat milk. However, in the high-stakes, high-pace world of automotive, an hour of downtime equates to millions in potential lost revenue. An unexpected failure of a material handling robot is simply unacceptable, so the increasing use of integrated and connected equipment not only provides insight into predictive maintenance but also enhances efficiency and safety while reducing energy consumption.
More than just buzzwords
IIoT moves past “network-enabled” components that provide insight into only a few machines relaying data to the maintenance or production manager’s desktop computer. Using sensors, edge computing, cloud connectivity, and sophisticated analytics, we can monitor and optimize every aspect of machine performance in real time.
All these buzzwords sound great, but we need to dig deep into actual examples to elucidate the power of modern, smart technology for industrial hydraulics. Specifically for hydraulics, we can monitor performance variables like pressure, flow, temperature, vibration and contamination to predict failures, analyze productivity, and plan maintenance.
Artificial intelligence can help provide solutions to problems or highlight patterns through the analysis of massive data.
Smart sensors and controllers are the backbone of every smart industrial hydraulics machine leveraging Industry 4.0 technology. Unlike traditional analog sensors, smart sensors can perform self-diagnostics, detect anomalies, and communicate rich data via digital protocols such as IO-Link, CANopen, or Ethernet. Smart sensors help hydraulic systems become more autonomous and efficient by providing actionable information, not just raw signals, to controllers or cloud-based systems.
I understand the apparent contradiction between a more autonomous machine and one that is part of the Industrial Internet of Things, so let me explain the difference. An old injection molding machine was certainly autonomous but used no smart technology, providing no capacity to monitor, predict or adapt. Using some of the previously mentioned sensors, modern machines can identify and adjust in real time and use advanced, integrated multi-core computers with AI capability to improve efficiency and reliability.
Edge computing has risen to prevalence with the above in mind while still communicating with other machines or the cloud. This hybrid model provides your operation with advanced technologies that your smartphone has been providing for years, alongside automated performance improvements on the fly. Again, I feel like it’s all too easy to throw around Industry 4.0 jargon to sound cool, but without specific examples, it’s all very arbitrary.
A case in point: injection molding machines
Let’s use the injection molding machine to illustrate how smart technologies apply to real-world applications, especially since these machines are hydraulic-heavy. Any good IMM starts with a mold clamping system to keep the dies closed during the injection molding process. Most molds use a hydraulic cylinder and a toggle to exert hundreds or thousands of tons of force against the platen and dies. In some large machines, you may even see four hydraulic cylinders, one each at every corner.
Injection molding machines rely heavily on hydraulics, while pressure sensors ensure even clamping throughout.
Using a pressure sensor on the piston side of the clamping cylinder(s) or load cells in each tie bar, the machine ensures even clamping force on each corner. The smart controller can detect misalignment, general wear or hydraulic imbalance to make pressure corrections to maintain quality. For example, if even a slight reduction in clamping force is detected, compensating with higher pump pressure or flow to prevent even a single short-short or flash non-conformance provides 100% first-pass yield. An egg tray in your smart fridge is of no use if it only holds eleven eggs.
A standard deviation within the above example may be processed automatically without operator or maintenance interaction, but should internal adjustments fall outside of a predetermined range, other connected sensors can provide data to tell a broader story. A thermal imaging camera now detects the clamping cylinder exhibiting localized heat across the tube, especially during clamping. Artificial intelligence now diagnoses the original clamping force issue as leaking piston seals, which reduced clamping force. The small increase in pressure compensated for the original minor leak, but as the seals further degraded, bypassing fluid resulted in more heat.
Until now, the machine ran “at the edge,” and no warnings or human interaction were required. The machine subsequently signals the maintenance team via their mobile app that they should replace the cylinder within the next week using the spare cylinder the connected ERP software knows is in stock. The maintenance concern is pressing but not critical, as the machine controller is still providing extra pressure and flow to stabilize the clamping force in the interim.
When taking full advantage of today’s advanced ERP platforms, the maintenance can be scheduled, and the software can automatically make adjustments for the production downtime. The work order for the cylinder replacement is created, and jobs are assigned to the maintenance team, who scan their barcode at the beginning and end of the repair. The seal kit used for the repair is then automatically removed from inventory while an RFQ or PO is auto-generated and emailed to the supplier for fulfillment.
Sensors provide the insights
Of course, such technologies are available for any and every machine. Even if you wish to take advantage of smart technology without breaking the bank or retrofitting an entire machine, individual sensors can provide insight into critical parameters for the purpose of improving efficiency, reducing unexpected and costly downtime or saving costs on redundant maintenance.
using a variety of sensors allows operators and maintenance personnel to better predict machine failures.
Smart pressure sensors, for example, are so much more than a way to monitor pressure remotely; they can solve problems. A Bluetooth pressure sensor provides monitoring through mobile and desktop apps while providing real-time measurements and trends, compiles data in various formats, uses configurable notifications and alerts, all the while offering a cloud-based app for any number of team members to view and analyze.
Imagine installing such a smart sensor on the nitrogen side of an accumulator. It’s rare to see pressure gauges monitor accumulator precharge to begin with, so such an upgrade is already an obvious choice. Precharge inevitably drops as nitrogen leaks through the permeable rubber bladder, so degrading machine performance may go unnoticed for an extended period. The app can be programmed to alert the maintenance team to add more nitrogen as required or to replace the bladder entirely should the sensor observe a catastrophic rupture in the bladder. Some accumulators are used to reduce pressure spikes or dampen vibrations, so more is on the line than simply actuator speed.
Keeping track of heat
I’ve written other articles about how temperature is the unsung hero of troubleshooting hydraulic machinery. Although, before this crazy inflation, I’d even bet dollars to fifty-cent donuts that nearly every hydraulic failure is concomitantly symptomatic of wasted heat. Using temperature sensors in strategic locations will provide valuable troubleshooting insights. I would prioritize the pump outlet, case drain, reservoir, and return line, although optional sensors installed into all work port lines would be helpful as well.
When running standard production, the software logs temperature at each location to provide baseline data that are understood as typical and cyclical. It will be easy to see natural fluctuations plotted against ambient temperatures and various production loads. Then, as any sensor location relays anomalous temperature rise, it’s easy to pre-emptively troubleshoot and sidestep a potential failure, preventing expensive downtime while reducing the costs of traditional, interval-based maintenance planning.
For example, let’s say a modest temperature increase is observed inside the reservoir, a moderate rise in the return line and a clear hot point in the outlet work line of a hydraulic motor. What could be happening inside of a motor to create heat that is transmitted downstream to the valve bank and reservoir but has no effect on the pump or case drain? Your fluid power professional can quickly predict there are worn internals that are allowing flow to bypass before achieving useful work, and you’d be right.
The above example doesn’t need to be a result of a dramatic temperature rise, and the machine could be otherwise performing perfectly. However, with months or years of “normal” data combined with alerts programmed to trigger when such anomalies are detected, you can avoid any unexpected downtime and related production interruptions.
AI-integration is the next step
Of course, edge computing and small-scale smart sensors can help any smart industrial hydraulic machine currently in operation, but where exponential improvements are made in conjunction with this technology is through artificial intelligence. A factory using varied and plentiful smart sensors all connected through one of various digital communication protocols and linked in the cloud can provide factories with absolute magic when integrated with AI.
Artificial intelligence has been shown to discover serendipitous solutions to old problems or highlight previously unknown patterns through the analysis of massive data, and recent advancements are not leaving the industrial space forsaken.
A recent real-life example has shown an unexpected benefit to Industry 4.0 within the automotive industry, where a well-known manufacturer employed these strategies. Using temperature, vibration, pressure and flow sensors combined with electric motor current signatures and operator usage patterns, this specific example has all but obsoleted the preventive maintenance methodology so popular for decades.
They were able to show that assumptions about wear, stress and component age requiring the proactive replacement of electric motors, pumps and actuators were entirely false. In fact, they discovered that the machine ran more efficiently when left past the expected service life of the components in question. Rather than arbitrarily replacing a pump that was expected to be past its useful life, keeping it in place until moments before failure increased production uptime, reduced maintenance costs, and expanded factory profit.
I understand that the six- or seven-digit cost of capital projects to replace an aging, profitable steel mill won’t make sense to everyone. But the benefits of smart technologies, Industry 4.0 and AI are much too relevant to turn a blind eye to. These systems will help any type of factory, from plastics to food & beverage, or from aerospace to pulp & paper. Especially in our industry where skilled labor shortages are increasing, technology is finally nearing the point where maintenance and troubleshooting problems can be solved even as our experts retire. And they’ll need that time in retirement to figure out how to use their smart fridge.
Filed Under: Components Oil Coolers, Engineering Basics, IoT, Sensors, Sensors & Gauges, Trending
