In the field of precision manufacturing, initial electrode lugs, as the core components of high-end
products such as lithium batteries, electronic components, etc., have a direct impact on the
performance and reliability of the end-products due to the machining accuracy. As one of the
core indexes to measure the processing quality of the initiator sheet, its detection accuracy
should be controlled at micron level or even sub-micron level. However, in the actual production
process, from equipment selection to process control, from environmental interference to data
analysis, perpendicularity detection faces multiple technical challenges. In this paper, we will
analyze the six major technical difficulties in the verticality detection of the initiator chip, and
provide systematic solutions for the industry.
Physical Limit Breakthrough in Micron-level Precision Inspection
The thickness of the starter piece is usually 0.05-0.2mm, and the tolerance of perpendicularity is generally
within ±0.005mm, which puts forward strict requirements on inspection equipment and technology:
Sensor resolution bottleneck: the traditional laser displacement sensors are susceptible to the interference
of surface roughness under 0.1μm resolution, and need to adopt the confocal white-light interferometry
technology to improve the signal-to-noise ratio;
Reference Plane Calibration error: the flatness of the inspection platform needs to be better than
0.001mm/m², and a natural granite base is used to reduce the impact of thermal deformation with a
constant temperature system (±0.1°C);
Multi-point synchronized measurement challenges: through the array sensor layout (such as 8-point
ring probe), combined with data fusion algorithms to compensate for single-point measurement deviation.
An experimental data shows that the use of multi-spectral interferometer can improve the verticality
detection repeatability accuracy to 0.3μm, which is 60% higher than the traditional method.
Environmental interference suppression under complex working conditions
Production site vibration, temperature and humidity changes and other factors significantly affect the stability of the test
results, the need to build multiple anti-interference system:
vibration isolation system: the use of active air-float vibration isolation device, the external vibration transmission rate
down to less than 5%, and effective isolation of mechanical vibration in the frequency band of 10-200 Hz;
optical compensation technology: in the optical detection equipment integrated real-time temperature drift compensation
module, through the reference optical path differential calculation to eliminate the ambient environment; optical
compensation technology: integrated real-time temperature drift compensation module, through the reference optical
path differential calculation to eliminate the ambient environment. Reference optical path differential calculation to
eliminate environmental disturbances;
Electromagnetic shielding design: three-level electromagnetic shielding (30dB/100kHz-1GHz) is implemented in the
detection area to avoid interference with sensor signals by frequency conversion equipment.
After testing, in the absence of protective measures, the common 5μm amplitude vibration in the workshop will lead
to 32% deviation in the detection results, while the perfect environmental control can suppress the error to within 3%.
Measurement Adaptability Optimization of Shaped Structural Parts
As the demand for shaped batteries in new energy vehicles grows, the inspection of L-type and curved starting pole
pieces is facing new challenges:
Non-contact Measurement Upgrading: Structured-light 3D scanning technology is adopted to realize the
perpendicularity resolution of 0.05mm feature dimensions on the complex curved surfaces;
Flexible Fixture Design: Variable-topology clamping mechanism is developed, and the angle self-adaptive
adjustment of 0.01° level is realized through the hybrid pneumatic-magnetic drive; data fitting. data fitting
algorithm innovation: the use of NURBS surface reconstruction technology, discrete point cloud data into a
continuous spatial model, perpendicularity calculation error is less than 0.8%.
Real-time problems of high-speed online inspection
The efficiency of traditional offline inspection is difficult to meet the needs of modern production lines,
and it is necessary to break through three major technological barriers:
Detection beat matching: the development of high-speed CMOS cameras (5000fps) and FPGA parallel
processing architecture, the single-piece inspection time is compressed to 0.8 seconds;
Dynamic Error Compensation: based on the Kalman filtering algorithm, real-time correction of image
blurring caused by conveyor belt jitter and positioning deviation; Edge Computing Deployment: the
use of NURBS surface reconstruction technology to convert discrete point cloud data into continuous
spatial models. Dynamic error compensation: based on Kalman filter algorithm, real-time correction
of image blurring and positioning deviation caused by conveyor jitter;
Edge computing deployment: integration of embedded AI chip in the inspection station, realizing
millisecond-level closed-loop control of measurement-analysis-sorting.
The practice of an intelligent production line shows that the online inspection system reduces the
product defective rate from 1.2% to 0.15%, while saving 45% of the cost of manual re-inspection.
Adaptation of Inspection Parameters for Multi-material System
The differences in physical properties of copper foil, aluminum foil, and composite materials require strong
adaptability of the inspection system:
Intelligent Adjustment of Reflectivity: Automatically adjust the intensity of the light source and the
parameters of the aperture for different materials, such as copper (reflectivity of 60%), aluminum
(85%), and so on;
Precise Control of the Contact Force: Adopt the piezoelectric ceramic driver to realize the closed-loop
adjustment of the force measurement of 0.01N level in the contact measurement;
Database support: the establishment of an expert system containing 200 + material parameters, a
key to call the optimal detection program.
Closed-loop linkage between inspection data and process optimization
The value of perpendicularity inspection is not only in quality determination, but also in guiding process improvement:
Factor Traceability Model: Establishing the quantitative relationship between 30+ parameters, such as stamping
speed, die clearance, etc., and perpendicularity deviation through big data analysis;
Adaptive Compensation System: Feedback the inspection results in real time to the machining equipment, and
correct the die inclination angle automatically (accuracy of 0.001°);
Digital Twin Application: build virtual production line to simulate the effect of different process combinations,
predict the verticality pass rate improvement program.
Technology Evolution Trends and Industry Suggestions
Verticality inspection technology will break through in three directions in the future:
Quantum measurement technology: using the characteristics of electron spin to achieve nanometer
precision inspection;
Intelligent process: autonomous monitoring of verticality from raw materials to finished products;
Improvement of the standardization system: promote the development of a globally unified standard
for the verticality inspection of starter laminations.
Enterprises should focus on:
selecting modularized inspection equipment to adapt to product iteration;
building in-depth integration of inspection data with MES/ERP system;
cultivating a composite technical team spanning mechanics, optics and algorithms.