fr en
< Back to articles

RENAULT Group - Using X-ray CT for failure analysis in on-board electronics

by Roland on 19 May 2021 at 16h46

The technical progress expected in the automotive industry in the coming years is important and very challenging, particularly with the development of electric vehicles. Renault Group has made innovation one of the keys to its success with its Guyancourt Technocentre, one of the largest automotive research and development center in Europe.

Vehicles are integrating a multitude of new functionalities, so much so that we now speak of "intelligent vehicles". These new features, such as gesture recognition, virtual assistance, eye-tracking, or attention control, will add to the current progress in accident prevention. These functionalities are made possible by the development of electronics and the integration of numerous sensors in modern vehicles.
Thus, understanding failure mechanisms will ensure quality levels consistent with large-scale manufacturing and distribution.

X-ray micro-tomography at Renault Group

The analysis of electronic parts follows the same pattern as the analysis of mechanical parts. In other words, it always starts with level 1 analysis: analyses using non-destructive means.
In the case of electronic parts, the main technologies used are radiography and micro-focus X-ray Computed Tomography. This allows us to observe the internal features of parts and components to detect defects and better understand the failure mechanisms in a non-destructive way. This allows when authorized, more intrusive analyses, targeting the samples very precisely so as not to destroy the defects and the signatures’ failures.
As electronics are more and more integrated, the challenge is to visualize smaller and smaller defects, and therefore to have technologies that can reach very high resolutions.

The challenges encountered 

The following analyses were performed using the EasyTom 160 nano-tomography equipment designed by RX Solutions for maximum flexibility. Here is one of the challenges encountered on water sensor with cracked diodes (the other three challenges are in the application note at the end of this article).

The high-resolution CT scans were able to unambiguously show the cracks in the glass. A mechanical cutting approach would have been inconclusive because the cuts could have potentially created these cracks in the brittle glass. The cracks are a few microns in size. 

Case n°1: Water sensor - cracked diodes

The sensor is made of a card encapsulated in black resin. The part is inspected as a whole: dimensions 100x80x60mm



Failure: Sensor not working.
Analysis Challenge: To identify cracks in a very small glass material within a large part.
Root cause: Mechanical breakage of the diode glass due to mechanical stress.
Hypothesis: The mechanical stresses induced by the resin (during its polymerization for example) may have put the PCB in slight bending due to its design, which caused the breakage of the 2 diodes located in the area most sensitive to this bending.




Positioning of the sensor in the tomography system: 
Limited angle acquisition allowing to obtain a strong magnification despite the 10cm long branch

Results on cracked diodes 

Image 1 
Location of the incriminated diodes in the water detector
Image 2&3 
EasyTom 160 -Voxel size: 2µm
Whole part acquisition 100x80x60 mm
Cracking of the diode glass
CT is a powerful technology that can be used for a wide variety of applications, ranging from material analysis to assembly inspection to dimensional measurements.
In the context of failure analysis for embedded electronics, high-resolution tomography allows inspection of internal components without risking further damage to the offending part (which could lead to damage or loss of elements relevant to the conclusion).
CT systems such as the EasyTom from RX Solutions allows to work easily at several scales, going from a global observation of the component to a very high resolution targeted analysis, despite the large size of the part to be analyzed, which makes it a tool of choice for failure analysis in on-board electronics.