1660318859268 Resizedimage201212nathancropped

A new look at leak testing

Dec. 29, 2016
Manufacturers need to pull more data out of their leak test to meet increasingly stringent standards.
The leak test is a critical and oftentimes challenging quality-assurance process for manufacturers of everything from automobiles to printer cartridges. And, unsurprisingly, the leak test is being updated by elements of digital transformation. We chatted with Nathan
Sheaff, founder and CEO of Sciemetric Instruments, to learn more. Take a look…

Smart Industry: How is digital transformation affecting the world of leak testing? 

Nathan: Manufacturers need to pull more data out of their leak test to meet increasingly stringent standards for quality-assurance and to maintain yield, not only to make sure they’re not passing a faulty part, but also to address any problems with the test equipment itself, whether it’s a broken solenoid, a sensor that’s out of place or an interruption to the test station’s air supply. Their best option is to equip the test stations with smart sensors and instrumentation to capture the complete digital-process signature of every test cycle. This delivers the depth of data necessary to achieve more reliable pass/fail results, trace the root cause of a problem faster and drive continuous improvement. Software is key. And, yes, you can improve the electronics. But it’s the software that enables you to achieve better pressure regulation, or to measure with greater accuracy to smaller increments. It’s software that allows that good old fashioned pressure gauge to be more than it is. We use software and smart sensors to continuously monitor 20 internal characteristics to ensure our leak-test systems are optimized and running properly, in addition to monitoring the test cycle itself.

Smart Industry: What digital-transformation elements come into play here?

Nathan: The other demand of IIoT is how you manage all that data, from every process and test station on the line. Manufacturers can no longer afford to leave data trapped in silos. It must be stored in a central database where it can be correlated, studied and visualized. Leak test is an operational requirement, a test on the factory floor. But when you start collecting and saving the data, leak-testing moves into the IT realm. You gain the data trail and insight for traceability in the plant, by machine and by product. By ensuring that data you capture from any process or test station flows seamlessly into the IT realm without being left in a silo, you can perform the good analytics for continuous improvement, to make the whole factory operate better. This is a process of digital transformation, to convert all those analog signals—like pressure, temperature, flow and test pressure—into digital thumbprints that we can analyze in new and innovative ways. The manufacturer can easily run simulations in a "digital sandbox” for continuous improvement of not just leak, but every process on the line.

Smart Industry: Why is there great opportunity with making leak testing "smart"?

Nathan: Many of the big manufacturers cite leak as a top warranty concern. It’s often a multimillion-dollar problem that can leave them with a black eye. The pressure is on to achieve a higher benchmark of quality and provide assurance that their products have met spec and passed their test benchmarks. Leak test has to get smarter as these competitive pressures force manufacturers to put greater demands on it. By capturing and analyzing process signatures, manufacturers can optimize and shorten their leak test to reduce the number of test stations and reduce warranty claims.

Smart Industry: What are unique challenges with leak testing in this respect? 

Nathan: Leak never has, and never will be, a cut-and-dried test. There are so many variables and potential points of failure that can skew the accuracy and consistency of the test—variations in the air supply, the thermal effects of compressed air, poor fittings and seals for the coupling to the test station, the impact of ambient air and part-temperature variations, etc. These have always been the challenges with leak and this hasn’t changed. But by making the test “smarter” with sensors/software/analytics, quality engineers can manage these variables far faster and more reliably than they ever could before. In this digital age, it’s also easy to overlook that the challenge is often mechanical. The more complex your test station set up—the more hoses, valves and connections—the greater the odds the test will suffer from poor reliability and repeatedly. Achieving a higher standard of leak testing often starts with simplifying the mechanical complexity of your test stations. But it takes data and manufacturing analytics to help understand the effects of these changes and improvements.

Want more? Check out the white paper “Design Solutions for the Internet of Things—Enabling Decisions at the Edge” right here.