Retrofitting is often the superior option when digitalizing a test bench. However, there are several
things to consider when doing so, because ultimately digitalization must be worth the outlay
Digitalization is on everybody’s minds and is becoming an instrumental part of the business strategy of many companies. Automotive manufacturers have devised some terrific ideas for digitalization of new test benches. However, they often require substantial investment. So what should be done with the existing systems if they are still operational?
At this point it makes sense to think about retrofitting. But what systems are suitable for this? They will usually be very expensive or large test benches that have a lot of media infrastructure and technology in the background. These test benches are often unique pieces in the company and therefore indispensable.
Let’s assume we have found a suitable test bench that we want to digitalize. Now what? Predictive maintenance sounds like a good option to ensure very high availability. To achieve that, a lot of sensors need to be installed and the entire system must be monitored.
But that is not enough. The test bench is now totally transparent but it doesn’t yet provide any foresight. Systems such as pumps, fans and other drive motors can be monitored by means of vibration, speed, current consumption and temperature. With a little effort this information can, for example, be used to predict damage by employing fast Fourier transform (FFT) analysis.
At this stage, far too much money has probably been spent for too little benefit. Digitalization must pay off financially, so think smart and digitalize test units not completely but cleverly. I always recommend analyzing in advance what the most common reasons for failure are and what should be additionally monitored. Financially it sometimes makes more sense to protect the device under test rather than the test bench, because the cost of an early-stage prototype is often enormous.
Another option is to automate the test sequence so that several test cycles run automatically and are recorded accordingly with the measuring system. The next step is automatic DUT change, which can be challenging due to the complexity of devices. Here it makes sense to look at end-of-line test benches because effective processes have already been established in this area.
Recently a plant manager I hold in very high regard told me that when automating existing production lines the process must always be analyzed and improved first. Otherwise you may have a digitalized process but not a very good one. We can transfer the same theory to test benches: always optimize the test process before you digitalize it.
