The severe conditions faced by machinery in today’s manufacturing environment, coupled with the demand that they operate longer without downtime, has led to the need for better tools to monitor their condition. Sensors have emerged as a useful way to provide prompt analytical data in real time.

A previous TLT article discussed a bio-sensor containing a specific bacterium developed to detect quantitatively the presence of specific compounds in wastewater streams (1). Two components that the bio-sensor were able to detect were arsenic and naphthalene down to a detection limit in the range of 10 ppm.

Minimizing wear remains a major objective for end-users to ensure optimum machinery performance. But identifying and using a reliable, real-time technique has proved challenging. Dr. Sameh Dardona, an associate director of research and innovation at the United Technologies Research Center (UTRC) in East Hartford, Conn., says, “The most frequent approach for monitoring wear is done through optical means, which literally means inspection of the machinery by an individual. This process can be time consuming, labor intensive and expensive. The cost can increase if the specific machine must be shut down, leading to a loss of manufacturing time.”

Wear sensors are available and used, but most are big and not flexible enough to be useful for specific applications. Anson Ma, associate professor of chemical and biomolecular engineering at the University of Connecticut in Storrs, Conn., says, “A better type of wear sensor is needed to obtain a clear picture of the health of a component present in a machine. There are many applications where such components as coatings and bearings need to be monitored in real time to keep a specific machine running. Ideally a wear sensor embedded in a specific machinery component would be in the best position to monitor the wear in real time.”

One technique that is valuable for developing wear sensors in machinery is 3D printing. Dardona says, “We believe 3D printing is advantageous because no foreign materials need to be present in the machinery, and the fine features of a sensor can be printed on the microscale. Only a small amount of material is needed in a sensor, which does not interfere with the performance of what can be a delicate component.”

Ma, Dardona and their collaborators have developed a technique for embedding wear sensors into machinery components using 3D printing.

Direct writing
The researchers used a technique called direct writing (DW) to produce wear sensors that can be placed into machines. DW is almost like actual writing but instead of using conventional ink, the team applied fine lines of a semisolid metal to create continuous metal filaments on a substrate. Both the direct write method and the smart sensor components were developed in the direct write lab at UTRC. 

The metal used by the researchers in demonstrating this technique was silver, which is very conductive. Ideally the best procedure for using this sensor is to embed it into a machine component or a coating. 

Figure 1 shows an illustration of a sensor produced by DW. 


Figure 1. A wear sensor produced through a 3D printing technique known as direct writing can be embedded into machinery components, leading to the potential for more quickly detecting changes in wear in real time. (Figure courtesy of the University of Connecticut.)

Says Ma: “DW is a controlled deposition process where semisolid metal with the consistency of toothpaste is applied through extrusion, which controls the pressure and location of the filaments. As an additive manufacturing technique, DW deposited the metal ink where it was needed in a highly efficient manner. No etching takes place, which meant that no waste was produced.”

Dardona indicated that the DW technique can use any type of metal ink. “We used silver in our initial work but semi-conductive and resistive inks can also be applied in the same fashion,” he says. 

The researchers evaluated five different commercially available inks that contained silver. Dardona says, “Our objective was to evaluate the flow properties of each ink. Due to their toothpaste consistency, the key properties were their rheology and their ability to draw lines on the substrate. One thing we did not want was broken lines, which is an indication of an incomplete circuit.”

The size of the nozzle was at least 15 times larger than the largest ink particle size to minimize any possibility of clogging and ensure consistent printing. Once the ink was deposited, the researchers evaluated how each material wetted the substrate. Ink viscosity also is important because if it is too low the ink flows everywhere and loses its shape. 

Ma summarized the three key properties for the ink. He says, “We evaluated the flow properties to determine the right consistency for the ink, the shape of the ink once it is deposited and the particle size of the ink to make sure it was small enough not to clog the nozzle.”

To select the proper ink the researchers developed a process flow model based on the rheology of the inks. Ma says, “As part of this process, we analyzed the rheology of the ink to make sure it would sufficiently stretch to draw thin lines. The rate of deposition was assessed, but we determined that too high a speed led to broken lines.”

The researchers’ hope is that these sensors will be integrated into machines and become effective in detecting even minor increases in wear to enable any problems to be identified and corrected quickly.

Additional information on this research can be found in a recent article (2) or by contacting Ma at anson.ma@uconn.edu or Dardona at dardona@utrc.utc.com. 

REFERENCES
1. Canter, N. (2013), “Bacteria-containing biosensor,” TLT, 69 (1), pp. 14-15.
2. Shen, A., Caldwell, D., Ma, A. and Dardona, S. (2018), “Direct write fabrication of high-density parallel silver interconnects,” Additive Manufacturing, pp. 343-350.