Development of a New Dry Transducer Tester
I developed a tester that verifies the functionality of transducers designed to operate with water between them, but does so in a completely dry environment.
Until now, these transducers (typically supplied by manufacturers like Audiowell or CTS) arrived pre-tested, but as part of our incoming inspection process, we were required to test them again using a dedicated setup.
Traditionally, testing involved immersing the transducers in water on the electronic production line. As you can imagine, this presented a serious challenge, introducing water to components connected to an electronic PCBA is far from ideal and carries significant risks.
I began looking for a way to improve the testing process specifically, a method that wouldn’t require water at all.
My first step was to search for materials that shared similar properties to water, such as homogeneity and density. I was presented with various options, including specialty adhesives and rubbers. However, many of these were expensive, hard to obtain, or impractical for production use. I needed something accessible, reliable, and easy to integrate into a tester.
Then, I noticed a hot glue gun on my lab bench.
Curious, I checked the datasheet of the glue sticks we were already using and discovered that their material properties, particularly density and homogeneity – were surprisingly similar to water.
I thought to myself, why not give it a try?
My plan was to try casting the glue onto the transducer, in order to simulate a constant water medium.
For this purpose, I designed a fixture to the exact dimensions of the transducer, which would serve just like a mold for casting the glue.
The first prototype:
The transducer operates at 2MHz, and after several attempts, I understood that the ideal layer thickness is between 2 and 5mm.
Since the purpose of the test was to check if the transducers work after they are soldered to an electronic board, a kind of Go/No-go test.
I started testing the effect of coupling a transmitting transducer using an oscilloscope.
The results showed a better than expected signal, very similar to the results that would be obtained in a regular water medium.
It was possible to identify faulty transducers, like this one for example.
After successfully proving the concept, I turned my attention to designing a jig that is more suitable for a production line:
Up until now, the transducer was connected via a probe to the scope, which caused wire tearing and many malfunctions. I designed a new jig that will allow connection to a simple BNC cable to avoid the problems described
Inserting the transducer into the jig and applying hot glue.
BNC connector soldering
Securing the connector and closing the lid
Testing:
