Wireless and autonomous embedded digital strain sensor

Use case

Thanks to the polymer matrix integrating part of the electronic, the bonding of this strain sensing element is extremely simplified. One has just to quickly clean the surface of the test object to remove traces of grease or other contaminants, spread the suitable adhesive, e.g. a cyanoacrylate, and finally place the sensing element on it. Compared to traditional metal foil strain gauge there is no need of a precise handling of the sensor with tweezers or to weld wires. With this embedded strain sensor, we obtain a plug and play sensing element with an ease of use never achieved before. These sensors do not require any modification of the object to instrument. This is a key feature to deploy this solution on already existing structures. The complete installation of the digital strain sensor can be easily performed by an operator without any prior training in less than a minute. The result is a drastic decrease of installation cost and time to instrument an object with a strain sensing element. The communication device just has to be eventually plug into the connector of the sensing element.

The digital strain gauges allow reading data related to several physical phenomena: pressure, force, strain… Nowadays companies desire to control their processes by minimizing their stocks or optimizing the maintenance done by the operators. The technology involved in the digital gauges allows to collect data and to fill these needs. We will introduce one kind of use case that our technology is able to perform.

For instance, our connected gauges can be used to detect the presence of an object and to measure its weight. One use case highlighting this application is a shelf. To do so, four gauges were installed on each corner of the shelf (Fig. 2). Installing the gauges this way allows to map it and then to obtain data for all the surface. They were directly glued to the shelf. All the gauges were connected to a device and communicated all the data: time and strain.

Matchmaker+ Anbieter zum Thema

zu Matchmaker+

Data acquisition and treatment

We have designed our tests to determine the weight of an object in the centre of the shelf. To do so, a tare with known weights was done to calibrate our system. The weights were put at the centre of the rectangular shelf in a way to get the strain read by all the gauges. The tare was done with a series of weights ranging from 300 g to 1 kg. The weights generate a bending of the shelf. By using several weights, we determined the response of the 4 strain sensors as a function of the mass and the position.

The position of the weight also has a direct impact on the strain that the gauges read. The strain due to the weight of an object depends on the dimension of the shelf, its material and the position of the object on the shelf. After having done this tare and measuring the distance between the applied weights and all the gauges, the strains are related to a specific weight and position. We then can use these data to calculate an equation relating strain, weight and position for all the gauges.

These equations are then used in an algorithm in a way to determine the weight of the object and its position on the shelf regards the strain read by the gauges. We have tested our algorithm using the strain obtained for the gauges when a known weight is put at the centre. By only using the read strain, we can get the weight of the object, as shown on the following Fig. 3.

We have used different calibration weights placed successively at the centre of the shelf, which is 38.5 cm away from the gauges. We then had for each weight a couple of strains. The algorithm computes and determines that the force related to the strain. The errors involved by the algorithm are below 7 %. One has to remember that the objective was to be able to roughly determine the weight and the position of an object on the shelf. The objective here was not to perform metrology, therefore the precision obtained is largely sufficient.

There is difference between the algorithm and the reality because the approximation of the strain is done by averaging the strain stage. However, the observed difference between the calibration weights mass and the measured mass in this study was not significant, and could be even more reduced by doing an exhaustive calibration of our system. By increasing the number of positions to do the tare, we can quickly map the shelf and know the behaviour of the gauges all over it. We are then able to determine the weight of several objects put on the shelf. Mechanical simulation of the test object (using finite element for example) could also provide the necessary data to calibrate the parameters of the algorithm in order to sense the mass and position of objects on the shelf.

1. Wireless and autonomous embedded digital strain sensor
2. Electronic, communication, power consumption
3. Use case
4. Conclusion