In 1918, N. W. Akimoff, a Russian immigrant, invented the first-ever dynamic balancing machine in the United States and helped pave the way in the last Industrial Revolution. Akimoff made it possible for the industry to balance machine rotors such as turbines, pumps, fans, or rolls and was duly lauded for his ingenuity in the world of mechanics. His balance machine, while a breakthrough in its time, is still used today when coupled with the latest technologies.
After establishing Vibration Specialty Corporation (VSC), Akimoff solved machinery problems for high-profile clients such as Thomas Edison and Henry Ford. VSC’s original service center balanced Edison’s turbines and generators and Ford’s Model T crankshafts. Since then, VSC has consolidated its position as a trusted leader in vibration analysis and predictive maintenance. While developing and leveraging the latest technologies, VSC provides onsite and remote monitoring services in machine diagnostics, balancing, and alignment for industries such as industrial gases, pulp and paper, marine, and power generation manufacturing.
In this conversation with CIO Applications, Jeffrey P. McGuckin, president of VSC, divulges details of VSC’s new solution offering—the V-Stream—and explains how his company uses state-of-the-art instrumentation to remotely monitor machines and prevent their failures.
Mr. Akimoff’s breakthrough in the realm of balancing machines opened the floodgates for a number of industries and sowed the seeds for the industrial revolution. The technology was so critical that our employees were exempt from serving in the armed forces during World War II and, instead, were asked to create balancing machines for the U.S. Navy and other military ventures. Since then, we’ve constantly been creating and designing new ideas, to prepare for the next industrial revolution. We believe our latest technology called the V-Stream will facilitate that.
The V-Stream allows us to gather vast amounts of vibration data and other pertinent information on machinery health conditions, to help us diagnose and analyze machinery health, much better than anyone ever did before. In essence, V-Stream enables us to gather about 1,000 times more data than anyone has gotten in the past or in today’s standard practices. Right now, our goal is to expand this new technology and partner with players in the Artificial Intelligence spectrum. Due to the vast amount of data gathered by V-Stream, the technology is tailor-made for AI.
The biggest pain point is that most companies lack the personnel on staff to perform maintenance work. When their machinery fails, they are unable to foresee the breakdowns.
That’s why they require remote monitoring for their machinery, so they can anticipate system failures in advance. Also, in many factories, the maintenance staff hires contractors, as they lack the personnel and expertise in machinery maintenance. As a result, these organizations approach us, so they can eliminate possibilities of downtime due to machinery failure. It can be a 3 to 7-day downtime before a machine can be up and running again.Another challenge besetting this industry is the methodology used to diagnose machinery. Our industry gathers data on machinery by taking a vibration snapshot (a picture), which cannot always reliably predict a problem in advance so that precautionary measures may be taken. This is where V-Stream will change the game.
Instead of taking a discrete snapshot in time, our technology captures a long, gap-free, raw data stream of 10 seconds or more. Since we capture the raw data, we can continually reprocess the data and pull out undetectable faults that we couldn’t find before. We can study the data from infinite perspectives, and continually extract more mechanical issues. For example, if a customer says, ‘our bearing failed six days ago,’ we can revert back to raw data that we had collected, and start mining for information. This way, we can discover problems with the machine that might have been initially missed.
It depends on the customer and requirement. Most of our customers buy our solution to collect their own data and monitor their machinery on at least a quarterly basis. They have two options to avail themselves with our remote monitoring services—either online where we are constantly connected to their equipment, or they can choose periodic monitoring which is less expensive. We also have customers who prefer a one-time diagnostic service, which allows us to capture their data and instantly show them the health of their machines. Often times, a one-time service turns into a long-term relationship because our customers are amazed at the diagnostic analytics we provide on their machinery.
The new technology allows us to manipulate data and discover problems after the fact, whereas our competitors are unable to do that. They have to rely on snapshots that have pre-set data collection parameters and lack access to any other data after collection is completed unless they perform a resurvey. We can unearth every conceivable problem with V-Stream. In many ways, we’re adopting a reverse approach, wherein we gather raw data and allow ourselves to manipulate it in any way and at any time. Through V-Stream, we are changing the way industries do machinery maintenance. We are facilitating preventive maintenance, allowing organizations to spot problems and perform simple tasks such as greasing a machine, in order to resolve a fledgling problem. Instead of trying to fix a machine in its dying seconds, they can prevent it from failing altogether.
Until now, nobody has had precise information with regard to machinery diagnostics. We try to be as precise as possible with everything we do and try to attain the highest possible standards.
A customer had installed a new motor only 24 hours prior to our data collection. After we collected the data, we noticed a strange signature in one of the two motor bearings. When we asked the client to check the motor, they discovered that the motor bearing wasn’t lubricated enough. As expected, the other motor bearing—the one in which we didn’t spot abnormalities—was properly lubricated and able to function optimally. After they completed the repairs, we analyzed the data again and the anomaly disappeared in the previously bad bearing. After that incident, the client didn’t run into problems again with this motor. This is a classic case of how predictive maintenance using V-Stream enables us to find the ‘needle in the haystack’.
As of now, we’re using V-Stream internally to solve problems for a few clients. However, we haven’t yet given them the access to this technology, but we plan to do that by July. We are also getting investors onboard and seeking technology partners to drive us forward. We are working to integrate all our gathered data on the cloud and train AI to mine it far better than any human could.