A customer called into the EMPF Helpline to request root cause failure analysis of functionality issues with their life critical electronic devices…

An original equipment manufacturer (OEM) contacted the EMPF Helpline and requested guidance and direction to obtain root-cause failure analysis services for an electronic assembly classified as life critical. A teleconference was conducted with the client to discuss various investigative processes to achieve their requirements. Based on the agreements from the teleconference, a summary of investigative methodology was  issued for review and acceptance. The OEM selected the following technologies: electrical testing, chemical analysis, visual inspection, X-ray inspection, and temperature testing. The OEM submitted multiple samples of materials and assemblies, applicable drawings, bill-of-materials and copies of recent field failure reports and a summary statement of known product failures. The client confirmed functionality failures occurred when temperatures went below 40 degrees C.

Test Methods:
Visual and X-ray inspection were performed to determine any obvious defects or anomalies. The technical data sheets were examined and evaluated for Design for Manufacturability and process design and a number of experiments were performed to evaluate the comparative insulation resistance and dielectric properties of different revisions. Ionic cleanliness testing was performed to assess the overall relative cleanliness of the assemblies. Samples were immersed in a 75 percent alcohol, 25 percent deionized water solution heated to 35ºC. Ionic contaminants were extracted from the assemblies and measured using a conductivity meter.

Results:
Visual Inspection:
The surface-mount and through hole soldering process used for the assembly produced acceptable solder joints. However on some samples, debris and solder balls were found on component bodies. Microscopy used to inspect the assemblies revealed the quality of the printed circuit board etching differs between revision assemblies. When compared with the source Gerber files, poorer trace etch resulted in some reduction in trace spacing (Figure 2-1).

X-ray inspection:
X-ray inspection was also performed on one revision assemby and solder was found under several component bodies (Figure 2-2.) The contamination was a point of interest and was a process indicator even though the assemblies passed functional testing at room temperature.

Gerberfile comparison:
A comparison of the Gerber data did reveal some minor differences between the board revisions including the increase in the clearance from some traces to the ground fill.

Design of Experiments:
A series of experiments was performed with the different revision assemblies.  Units were tested over temperature per customer’s procedure. The performance of each sample was monitored using the client supplied data logger and a digital storage oscilloscope.  Tests included procedures to:

• Baseline the performance of each unit before making any changes to the assemblies;
• Establish the effect of the solder contamination on performance of the revision assembly with functionality issue;
  
• Determine the effect of contamination on the performance of an assembly;
  
• Determine the effect of component tolerance on inputs with different assemblies, e.g. differential amplifier inputs.

Dielectric Withstanding Voltage and Insulation Resistance Testing were performed and revealed that they are not likely to affect the circuit performance.  In addition, bulk ionic test results from the Ionograph indicated the ionic residue level observed on the different revision assemblies exceeded IPC standards specifications.

Conclusions and Recommendations:
The tests and circuit analysis performed on each revision of the PCB revealed several critical issues such as the presence of solder balls and solder contamination under component packages; higher ionic residue level on the suspect boards assemblies, and component tolerance test value discrepancies. 

These issues collectively contributed to the intermittent functionality and in some cases performance degradation.  A sampling test program to screen for these characteristics could provide some control on the quality of the printed circuit boards. 

To address these problems, the EMPF recommended updating the PCB artwork to incorporate coupons that allow for calibrated resistance measurements.  Also, the EMPF recommended screening of circuit assemblies for contamination on a lot sampling basis and also running a selected list of acceptance tests based  on IPC-6012: Dielectric Withstanding Voltage, Electrical Continuity and Insulation Resistance Testing, Moisture and Insulation Resistance Testing. 

Using experienced electrical, materials, process and
manufacturing engineers and scientists, the EMPF performs detailed investigations in the areas of materials analysis, electrical characterization, and device integrity. For more information, please contact the EMPF helpline at (610)362-1200 or via email at helpline@empf.org.