TB-6348 · REV B · Technical newsheet

Machine Vision & InspectionDevice profile

ZEISS INRADIA 3D Brings In-Line CT to Ten Parts Per Minute

ZEISS rates its INRADIA 3D in-line CT system at ten components per minute on prismatic battery cells, using a moving source-detector gantry, dual robots and AI pass/fail analysis.

By Nathan Brooks3 min read641 words

Features

  • ZEISS INRADIA 3D inspects up to 10 prismatic battery cells per minute in-line.
  • Source and detector rotate around a stationary part, inverting conventional CT design.
  • AI models in ZEISS INSPECT X-Ray deliver automated OK/NOK decisions on 3D volume data.
  • Two robots alternate part feeding and removal to minimize wait times.
  • Automatic calibration, temperature compensation and reference measurements support continuous operation.
ZEISS Launches INRADIA 3D In-Line CT Production Inspection System
Device photoZEISS Launches INRADIA 3D In-Line CT Production Inspection System — u_oly6hcjdky / pixabay

ZEISS has set an in-line computed tomography throughput mark of up to ten components per minute with its new INRADIA 3D system, aimed at prismatic battery cell inspection on high-speed production lines. The figure is the headline specification of the machine, which the company positions as a fully automated, non-destructive 3D inspection station embedded directly in mass production rather than in a metrology lab.

How does the scanner reach production-cycle speed?

The architecture inverts the conventional CT arrangement. Instead of rotating the part between a fixed X-ray source and detector, INRADIA 3D moves the source and detector around a stationary component. ZEISS states this design allows rapid acquisition of three-dimensional volume data while preserving high image quality and repeatability — a claim about system behavior under continuous operation, with no accuracy class or resolution figures published in the launch material.

Two robots working alternately handle part feeding and removal. The vendor credits the coordination between scanning motion and component handling with minimizing wait times and sustaining inspection speed at line rate.

Dr. Valentina Aloisi, Head of Product Management X-Ray at ZEISS, framed the engineering problem in cycle-time terms: "The key challenge in in-line computed tomography is to provide precise three-dimensional information within the available production cycle time. With ZEISS INRADIA 3D, we have consistently tailored the entire system architecture to these requirements. This enables manufacturers to inspect complex quality features in the production line."

What does the AI analysis chain decide?

Speed alone does not qualify a system for unattended production use; the evaluation of CT volume data must also run without an operator. ZEISS pairs the hardware with ZEISS INSPECT X-Ray and Volume Inspection AI, a software chain that links data acquisition, volume reconstruction and analysis in one workflow. Application-specific AI models trained to recognize defined features and deviations in the 3D volume deliver automated OK/NOK pass/fail decisions.

ZEISS says the reproducible evaluation reduces manual intervention, improves traceability, and returns results in real time with the production cycle. The company has not published detection rates, false-reject figures or model validation data, so the reliability of the automated verdicts remains a vendor claim pending field results.

Which inspection tasks does it cover?

The lead application is quality assurance for prismatic battery cells. The system automatically inspects:

  • electrode overhangs,
  • the number of electrodes,
  • folded or protruding electrodes.

The 3D method detects these deviations inside sealed cells without destroying them — the core advantage over cut-open sample checks, since every cell on the line can be examined rather than a statistical subset.

Further named applications include soldered and welded joints in power electronics and geometric inspection of complex connectors. All three application families share the same requirement: internal or hidden geometry that optical or tactile methods cannot reach at production speed.

How does it integrate and hold calibration?

INRADIA 3D adapts to different production requirements in component handling, inspection parameters and automation interfaces, and integrates into existing lines and Manufacturing Execution Systems (MES). Inspection results therefore feed directly into downstream production processes rather than stopping at a quality report.

For measurement stability during continuous operation, the system carries automatic calibration functions, temperature compensation and reference measurements. ZEISS describes these as the mechanisms keeping measurement performance stable over shifts, though no calibration interval or environmental operating envelope appears in the launch documentation.

ZEISS sells the system as a single-vendor package of hardware, software and services, with manufacturers expected to use the resulting quality data to monitor and optimize their manufacturing processes beyond simple part rejection.

The open question for adopters is whether ten parts per minute and AI-based OK/NOK decisions hold up against their own cycle-time and traceability requirements — and whether in-line CT at this throughput finally displaces destructive sample testing in battery cell plants.

via zeiss.com (Original)

Filed under

  • zeiss-inradia-3d
  • computed-tomography
  • battery-cell-inspection
  • non-destructive-testing
  • ai-quality-inspection
Share this article:

More from Nathan Brooks

Nathan Brooks

Show full bio

Staff writer covering people, launches and day-to-day practice in Instruments, test and measurement at Testbench Report.

105 articles

Application notes

« Previous articleNext article »