Practical Guidance for X-ray CT Scanning Electronics

X-ray CT is a powerful nondestructive inspection technique, but some electronics components can be sensitive to X-rays and other ionizing radiation.  Passive electronic components like resistors, capacitors, inductors, and the actual PCBs themselves are unaffected by exposure to ionizing radiation, but flash memory units, IMUs, and accelerometers are all susceptible to total ionizing dose (TID), which is a measure of the cumulative amount of ionizing radiation they have received. TID can cause electronic threshold shifts, leakage current, timing changes and functional failures. The total dose threshold varies by device, so it is best to determine the dose threshold for a given device via direct testing. When scanning electronics, it’s important to understand TID to perform a scan that both achieves a given inspection while minimizing the risk of affecting the device’s performance.  

Dose is dependent on multiple factors, including:

•  Scanner source energy
•  Duration of the scan
•  Number of times the device will be scanned
•  Distance between the part and the source
•  Amount of filtering used

Dose rate is highest at high magnifications (smaller source to object distance) and at high source energy and lowest at low magnification (larger source to object distance) and low source energy.  The charts below show absorption rates for silicon, the primary material in active electronic devices, and can be used to determine the rate of absorbed dose at different magnifications with different source energies.   

Fig 1. Absorbed dose rate for electronics on 190kV and 120kV Neptunes.

The below table provides general guidance on TID thresholds for common electronic devices.

Fig. 2 X-ray Inspection Considerations for Surface Mounted Flash ICs. Infineon Technologies 001-98522 Rev. *C [ref].

Guidance for scanning electronics:

• In practice, most scans of electronic devices will use some level of filtering to mitigate the artifacts that arise from multi-material assemblies.  Filtering reduces the amount of X-ray light that is used for a scan, reducing the dose, but longer scan times will typically be required when using filtering.  Refer to the charts above to understand the impact of the level of filtering on your scan.
• While increasing magnification increases dose, it also increases fine feature resolution and reduces the chances of having to repeat a scan if the first scan does not resolve the target features.
• Once you have positioned the part, set a scan duration using Auto Scan to find the other scan parameters.  Setting scan time in conjunction with the charts above will let you control the TID that the device receives. 

Will my device be safe? A worked example

As can be seen from the table above, TID thresholds for many devices vary dramatically.  While not always possible, the highest confidence method to assess the impact of X-rays on an electronic device is to find the TID limit experimentally.  See this worked example to learn how to determine the TID threshold experimentally. 

Device: Flash Drive

Scan Conditions:

• Scanner Configuration: 190 kV
• Source to Object Distance: 200 mm
• Filter: 0.5 mm Copper
• Scan Duration: 3 Hours

Procedure:

1. Use the dose absorbance charts above to determine dose at the scan conditions.  In this example, the scan will be conducted on a 190kV Neptune at a source to object distance of 200mm with 0.5mm of Copper filtering.  This corresponds to an absorbed dose rate of approximately 2 krad/hr.  A 3 hour scan would result in 3 hr * 2krad/hr = 6krad of dose. 
   
   
   Fig 3. Using the 190kV absorption rate plot to identify the absorption dose rate at the intended scan settings.
2. Scan the device in 1 krad increments and subject the device to functional testing in between doses.  In this example , this would mean scanning the device in 30 min increments.
3. Continue until the device fails functional testing. The TID at which the device fails testing is an experimentally-derived dose limit to which a margin of safety should be applied.  This margin can be itself experimentally determined by repeating the above procedure to build statistics about TID for the device. 

As always, if you have any questions about your specific application, please contact us at support@lumafield.com!