Electrical methods for structural defect detection in battery cells
Systems, methods, and devices for characterizing a defect of a device-under-test (DUT). A first measurement is performed of a first quantity on the DUT prior to performing a first operation on the DUT, producing a first result. The first operation is performed on the DUT, and subsequently a second measurement of the first quantity is performed on the DUT, producing a second result. A defect class is characterized for the DUT from a plurality of defect classes based on a difference between the first and second results.
1 . A method for characterizing a defect of a device-under-test (DUT), the method comprising:
performing a first measurement of a first quantity on the DUT prior to performing a first operation on the DUT, wherein performing the first measurement produces a first result;
performing the first operation on the DUT;
after performing the first operation on the DUT, performing a second measurement of the first quantity on the DUT, wherein performing the second measurement produces a second result;
characterizing a first defect class for the DUT from a plurality of defect classes based at least in part on a difference between the first and second results, wherein the plurality of defect classes comprise defects in different regions of the DUT.
2 . The method of claim 1 ,
wherein the first defect class indicates a defect in a first region of the DUT, and
wherein the first and second measurements are not specific to the first region of the DUT.
3 . The method of claim 1 , further comprising:
characterizing a magnitude of the first defect based at least in part on the difference between the first and second results.
4 . The method of claim 1 , further comprising:
heating or cooling a first region of the DUT prior to performing the first measurement, the first operation, and the second measurement;
wherein characterizing the defect class is performed further based in part on the heating or cooling of the first region.
5 . The method of claim 1 ,
wherein characterizing the first defect class for the DUT utilizes a machine learning algorithm to map the difference between the first and second results to different ones of the plurality of defect classes.
6 . The method of claim 1 , further comprising:
determining the first and second measurements based at least in part on a structure of the DUT.
7 . The method of claim 1 ,
wherein the DUT comprises a battery comprising a plurality of conductive plates,
wherein the first and second measurements comprise applying an alternating current (AC) voltage with a frequency sweep to the DUT and measuring a first and second reactive impedance of the DUT, respectively, wherein the difference between the first and second results comprises a difference in the first and second reactive impedances of the DUT as a function of frequency;
wherein characterizing the first defect class for the DUT comprises:
identifying a first frequency that causes a dip in difference in the first and second reactive impedances of the DUT;
determining a thickness of the conductive plates based at least in part on the first frequency; and
determining the first defect class based at least in part on the thickness of the conductive plates.
8 . The method of claim 1 ,
wherein the DUT comprises a battery,
wherein the first and second measurements comprise applying an alternating current (AC) voltage with a frequency sweep to the DUT and measuring a first and second reactive impedance of the DUT, respectively, wherein the difference between the first and second results comprises a difference in the first and second reactive impedances of the DUT as a function of frequency;
wherein characterizing the first defect class for the DUT comprises:
identifying a first slope of the difference in the first and second reactive impedances of the DUT as the function of frequency; and
determining the first defect class to be a torn tab of the battery based at least in part on the first slope.
9 . The method of claim 1 , further comprising:
wherein characterizing the first defect class for the DUT comprises:
determining that a signal-to-noise ratio (SNR) for reactive impedance of the DUT increases with excitation frequency; and
determining that a bad tab weld occurred on the DUT based on the determination that the SNR for reactive impedance increases with excitation frequency.
10 . The method of claim 1 ,
wherein the first operation comprises one or more of:
a manufacturing step of the DUT;
controlled regional heating of the DUT;
a physical actuation of the DUT; or
an electrical actuation of the DUT.
11 . The method of claim 1 ,
wherein the plurality of defect classes comprises two or more of:
a weld defect;
a bad tab weld;
a torn tab;
a folded separator;
a dendrite;
an internal short;
internal contamination; and
wetting.
12 . A controller, comprising:
a non-transitory computer-readable memory medium; and
a processor coupled to memory medium, wherein the processor is configured to execute program instructions stored on the memory medium to cause the controller to:
perform a first measurement of a first quantity on a device-under-test (DUT) prior to performing a first operation on the DUT, wherein performing the first measurement produces a first result;
perform the first operation on the DUT;
after performing the first operation on the DUT, perform a second measurement of the first quantity on the DUT, wherein performing the second measurement produces a second result; and
characterize a first defect class for the DUT from a plurality of defect classes based at least in part on a difference between the first and second results, wherein the plurality of defect classes comprise defects in different regions of the DUT.
13 . The controller of claim 12 ,
wherein the first defect class indicates a defect in a first region of the DUT, and
wherein the first and second measurements are not specific to the first region of the DUT.
14 . The controller of claim 12 , wherein the program instructions are further executable to cause the controller to:
heat or cool a first region of the DUT prior to performing the first measurement, the first operation, and the second measurement;
wherein characterizing the defect class is performed further based in part on the heating or cooling of the first region.
15 . The controller of claim 12 ,
wherein the DUT comprises a battery comprising a plurality of conductive plates,
wherein the first and second measurements comprise applying an alternating current (AC) voltage with a frequency sweep to the DUT and measuring a first and second reactive impedance of the DUT, respectively, wherein the difference between the first and second results comprises a difference in the first and second reactive impedances of the DUT as a function of frequency;
wherein, in characterizing the first defect class for the DUT, the program instructions are further executable to cause the controller to:
identify a first frequency that causes a dip in difference in the first and second reactive impedances of the DUT;
determine a thickness of the conductive plates based at least in part on the first frequency; and
determine the first defect class based at least in part on the thickness of the conductive plates.
16 . The controller of claim 12 ,
wherein the DUT comprises a battery,
wherein the first and second measurements comprise applying an alternating current (AC) voltage with a frequency sweep to the DUT and measuring a first and second reactive impedance of the DUT, respectively, wherein the difference between the first and second results comprises a difference in the first and second reactive impedances of the DUT as a function of frequency;
wherein, in characterizing the first defect class for the DUT, the program instructions are further executable to cause the controller to:
identify a first slope of the difference in the first and second reactive impedances of the DUT as the function of frequency; and
determine the first defect class to be a torn tab of the battery based at least in part on the first slope.
17 . The controller of claim 12 ,
wherein the processor comprises one of:
a parallel multi-processor system;
a field programmable gate array (FPGA); or
an application specific integrated circuit (ASIC).
18 . A non-transitory computer-readable memory medium storing program instructions which, when executed by at least one processor, cause the at least one processor to:
perform a first measurement of a first quantity on a device-under-test (DUT) prior to performing a first operation on the DUT, wherein performing the first measurement produces a first result;
perform the first operation on the DUT;
after performing the first operation on the DUT, perform a second measurement of the first quantity on the DUT, wherein performing the second measurement produces a second result; and
characterize a first defect class for the DUT from a plurality of defect classes based at least in part on a difference between the first and second results, wherein the plurality of defect classes comprise defects in different regions of the DUT.
19 . The non-transitory computer-readable memory medium of claim 18 ,
wherein the first defect class indicates a defect in a first region of the DUT, and
wherein the first and second measurements are not specific to the first region of the DUT.
20 . The non-transitory computer-readable memory medium of claim 18 ,
wherein the DUT comprises a battery,
wherein the first and second measurements comprise applying an alternating current (AC) voltage with a frequency sweep to the DUT and measuring a first and second reactive impedance of the DUT, respectively, wherein the difference between the first and second results comprises a difference in the first and second reactive impedances of the DUT as a function of frequency;
wherein, in characterizing the first defect class for the DUT, the program instructions are further executable to cause the at least one processor to:
identify a first slope of the difference in the first and second reactive impedances of the DUT as the function of frequency; and
determine the first defect class to be a torn tab of the battery based at least in part on the first slope.