In-Line Characterisation for Improving Cell Quality
In-line characterisation for cell quality is becoming a priority as battery makers move from batch production to gigafactory scale, where even small defect rates translate into large financial losses. This case study follows a pouch cell manufacturer deciding which quality control technology to buy in order to raise yield, produce more grade A cells, and cut scrap through an early predictive approach.
The Scale-Up Quality Problem
The company at the centre of the case produces electrodes, assembles cells, and runs formation, and it wants to scale from hundreds of cells per week to mass production. The challenge is that scrap can exceed 30 percent during the ramp-up phase, and each scrapped cell adds cost to every good cell shipped. Cells are graded at end of line: grade A cells ship as good product, grade B cells are managed through concession, and grade C cells fail acceptance criteria and become scrap. Grading rests on visual inspection, dimensional and mass checks, insulation evaluation, and electrical tests. The goal is to reduce waste and lift yield toward 98 percent while keeping capital expenditure and time to market under control. The core insight is that catching defects late, at the finishing stage, is far more expensive than catching them early.
Where Inspection Falls Short Today
Much quality testing still relies on manual, offline methods and human interpretation. The most mature in-line techniques are concentrated at the cell finishing steps and mostly assess electrical performance. Common methods include the hi-pot dielectric test for insulation integrity, open-circuit voltage, alternating current internal resistance at 1 kHz, direct current internal resistance by pulse test, initial capacity, and self-discharge. Each has limitations. Open-circuit voltage is an unreliable standalone indicator because a low reading can signal either high self-discharge or an error in setting the state of charge. Direct current internal resistance measurements can deviate under high current, and cells with higher resistance heat unevenly. Self-discharge testing is accurate but slow, sometimes taking days. Crucially, the electrode preparation stage, which is fundamental to the whole pipeline, has almost no in-line inspection. Cell assembly is starting to see vision systems, hi-pot tests, and in-line computed tomography, but speed and cost still limit adoption.
Building the Business Case for Early Detection
The team modelled a hypothetical 15 gigawatt-hour gigafactory and set a target of cutting scrap to 2 percent. At an assumed cost of 100 euros per kilowatt-hour, that reduction represents savings of roughly 30 million euros. To reach it, they scouted the most promising technologies capable of being upscaled to industrial production, focusing on 3D computed tomography scanning and real-time impedance spectroscopy. The logic is that inspecting 100 percent of samples earlier in the process, rather than waiting for end of line, prevents defective material from carrying cost through every subsequent step. Moving inspection upstream into electrode preparation is where the largest untapped quality gains sit, because defects introduced there currently go undetected until much later.
What It Means for the Industry
As gigafactories multiply, quality control shifts from a finishing-line checkpoint to a distributed, predictive capability embedded across the line. Reducing scrap is not only a cost issue; it also reduces the volume of production waste that would otherwise need recycling, which strengthens the environmental case for the factory. The case argues that the winning investment is technology that can be deployed in-line, run at production speed, and detect problems early enough to prevent waste rather than merely sort it. That reframes quality control as a lever on capital efficiency and time to market, not just a gate at the end.
Key Takeaways
Scrap can exceed 30 percent during ramp-up, so early defect detection directly protects margin.
Cells are graded A, B, or C at end of line, and effective in-line inspection reduces the share that fails.
Most mature in-line methods sit at finishing and test electrical performance, leaving electrode preparation largely uninspected.
Common electrical tests each carry limitations, and self-discharge testing is accurate but slow.
Cutting scrap to 2 percent in a 15 gigawatt-hour plant could save around 30 million euros.
3D computed tomography and real-time impedance spectroscopy are the most promising upscalable technologies.
Moving inspection upstream and covering 100 percent of samples is the route to higher yield at lower capital cost.
Disclaimer: This case study was developed and presented by BatteryMBA participants as part of the Case Study Track. Views, analysis and recommendations are the authors' own. BatteryMBA does not take responsibility for the accuracy or completeness of the content and it should not be relied upon as investment, engineering or legal advice.
This is the public summary, the full case study lives inside the programme
Every BatteryMBA cohort runs the Case Study Track: small teams build the full recommendation, backed by a written document and a live presentation, supported by the BatteryMBA team. Full case study documents are not shared outside the programme. programme.