Browse Prior Art Database

CRACK DETECTION IN MULTILAYER CERAMIC CAPACITORS USING ACOUSTIC EMISSION SENSING

IP.com Disclosure Number: IPCOM000248415D
Publication Date: 2016-Nov-25
Document File: 10 page(s) / 2M

Publishing Venue

The IP.com Prior Art Database

Related People

Caroline Andersson: AUTHOR [+5]

Abstract

During manufacturing, handling and testing of PCBs, mechanical stress can result in small cracks in multilayer ceramic capacitors (MLCCs) which with state-of-the-art measurement techniques cannot be detected. After the PCBA containing product such as a converter has been delivered to the customer, the crack might turn from a latent failure into an electrical short resulting in failure of the whole converter. In this invention we propose a method to detect cracked MLCCs in a non-destructive way by measuring the acoustic emission of MLCCs with a point contact sensor. A test pulse is driven into the measured MLCCs and the resulting acoustic emission measured with the point contact sensor shows whether the measured capacitor has mechanical cracks or not, as shown by the proof of concept experimental results in this document.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 31% of the total text.

Page 01 of 10

Page 1 of 10

CRACK DETECTION IN MULTILAYER CERAMIC CAPACITORS USING ACOUSTIC EMISSION SENSING

Abstract

During manufacturing, handling and testing of PCBs, mechanical stress can result in small cracks in multilayer ceramic capacitors (MLCCs) which with state-of-the-art measurement techniques cannot be detected. After the PCBA containing product such as a converter has been delivered to the customer, the crack might turn from a latent failure into an electrical short resulting in failure of the whole converter.

In this invention we propose a method to detect cracked MLCCs in a non-destructive way by measuring the acoustic emission of MLCCs with a point contact sensor. A test pulse is driven into the measured MLCCs and the resulting acoustic emission measured with the point contact sensor shows whether the measured capacitor has mechanical cracks or not, as shown by the proof of concept experimental results in this document.

1 Background of the Invention

Experiment using acoustic emission (AE) of multilayer ceramic capacitors (MLCCs) have been performed, and the difference in signals of new and cracked MLCCs have been investigated. The experiments, of which parts are presented here, show that an acoustic based method is feasible for crack detection in MLCCs. The method has not yet been tested in a product.


1.1 Problem to be solved

In all kind of electronic hardware capacitors are employed to store energy, e.g. for stabi- lizing DC voltage levels or for filtering. During manufacturing and handling of the PCBs, mechanical stress might result in small cracks in ceramic capacitors already put onto the PCB. Initially, the crack cannot be detected and/or might result in a failure, often in form of an open connection. After some time in operation, often when the converter has already been delivered to the customer, the crack will often turn from an open connec- tion into an electrical short. This has a huge impact because all parallel capacitors will

© Copyright [2016] ABB. All rights reserved.



FI-1604901


Page 02 of 10

Page 2 of 10

be shorted, the voltage at this point of the circuit will drop to zero, and the converter will stop working. The proposed invention could be used as a screening test methodology where the cracked MLCCs can be screened out and replaced, so that the failures in the field due to cracked MLCCs could be avoided. Because MLCCs are one of the most common devices in the electronics industry, also many other industries could benefit from the proposed invention.


1.2 How is the solution intended to work

The PCBAs containing MLCCs would be introduced to a functional test setup where during powering of the MLCCs, a point contact sensor would measure the acoustic emission of the individual MLCCs. The signal spectra can either be compared with a signal spectra from a golden MLCC sample representing what the spectra should theo- retically look like. Or selected parts of the spectra corresponding to the cracks in an...