How a Cell Is Made — and Where “Grade” Comes From
Eleven process stages, three of which decide whether a cell ships as Grade A, B or C — and why yield is the whole economics of a cell plant.
Battery Fundamentals · Part 1 — Anatomy and Instrumentation · Chapter 4 · 15 min read
There is no international standard behind the letters A, B and C on a cell. They are trade terms, and they are decided at three specific points near the end of a manufacturing line. Knowing which points, and what is being measured there, is the difference between buying a discount and buying a defect.
11
Process stages
<1 % RH
Dry room for filling
Days–weeks
Ageing before grading
70 vs 95 %
Yield that decides viability
4.1 — Why cell making is a process industry, not an assembly industry
The intuitive model of manufacturing is assembly: parts arrive, they are put together, and the quality of the result is the quality of the parts plus the care of the assembly. Cell manufacturing is not like that. It is closer to semiconductor fabrication than to carmaking.
What comes out of a cell line depends on coating uniformity measured in microns, drying profiles, calendering pressure, moisture control, and the absence of metal particles too small to see. None of those are parts you can buy. They are process conditions you have to learn to hold, and learning them takes calendar time that capital cannot compress.
Technical framing
This is the structural reason cell manufacturing has defeated well-funded newcomers. Two lines built to the same drawing, from the same equipment vendors, running the same materials, can produce cells with completely different lives — because the variables that matter are not in the drawing.
4.2 — Stage 1 — making the electrode
- 1Mixing. Active material powder, conductive carbon and a polymer binder are mixed into a slurry. Dispersion quality here sets electrical contact throughout the electrode; a badly mixed slurry produces a cell with higher resistance that no later step can fix.
- 2Coating. The slurry is spread onto foil — aluminium for the cathode, copper for the anode — on both sides, to tolerances measured in microns. Coating weight variation translates directly into capacity variation between cells, which is the origin of the batch spread you will later have to match around.
- 3Drying. Solvent is driven off in long ovens. Too fast and the binder migrates to the surface, weakening adhesion; too slow and the line is uneconomic.
- 4Calendering. The coated foil is pressed between rollers to a controlled density and porosity. Denser means more energy per litre; too dense and the electrolyte cannot penetrate, which kills rate capability. This single parameter is much of the difference between a high-energy and a high-power cell from the same materials.
- 5Slitting and notching. Cutting to width and forming the tabs. Burrs left at this step are a classic cause of latent internal shorts, because a metal sliver is exactly what a 20 µm separator cannot survive.
4.3 — Stage 2 — assembly and filling
- 1Winding or stacking. Cylindrical and many prismatic cells are wound into a jelly-roll; pouch and some prismatic cells are built as stacked layers. Stacking gives better space utilisation and more uniform current distribution; winding is faster and cheaper.
- 2Assembly and tab welding. The electrode assembly goes into its can or pouch and the tabs are welded to the terminals. Weld quality here is invisible from outside and shows up years later as resistance growth.
- 3Electrolyte filling and sealing. Done in a dry room held below one per cent relative humidity, because LiPF₆ reacts with water to form hydrofluoric acid. A line that cannot hold its dry room cannot make good cells, no matter what else it does correctly.
Important
Everything to this point is reversible in the sense that a bad batch can be scrapped. From the next stage onward, the cell has been electrically activated and its character is fixed.
4.4 — Stage 3 — formation, degassing and ageing
4.4.1 — Formation
The cell receives its first controlled charge and discharge cycles. This is when the SEI film forms on the anode — the layer described in Chapter 3 — and the profile used determines how good that film is.
Formation is slow and it ties up expensive equipment, so it is a standing temptation to shorten. A cell formed too quickly gets a less stable SEI, which means faster capacity fade and worse gas generation over its life. Nothing about it is visible on delivery.
4.4.2 — Degassing
Formation generates gas. Pouch and many prismatic cells are opened, the gas is evacuated, and the cell is resealed. A cell that was not degassed properly will swell in service, and a swollen pouch cell in a compressed module transmits that force into everything around it.
4.4.3 — Ageing
Cells then rest at controlled temperature for days to weeks, and their open circuit voltage is measured repeatedly. A healthy cell holds its voltage. A cell with a microscopic internal short — typically from a metal particle that got in during electrode manufacture — leaks charge, and its OCV falls measurably.
Important
This rate of voltage decay is the K-value, quoted in millivolts per day, and it is the single most important screening measurement in cell manufacturing. It is what catches the defect that otherwise turns up three months later as one permanently low cell in a customer’s pack.
4.5 — Stage 4 — grading, and where A, B and C are actually decided
Finally every cell is measured — capacity, open circuit voltage, internal resistance, K-value — and binned. This is where the grade letters come from.
What the grade letters mean in practice — note there is no standard behind them
| Grade | What it means | What you get | Where it belongs |
|---|---|---|---|
| Grade A | Passed every factory test inside specification. | Capacity, IR and self-discharge all within band; no cosmetic defects; full batch traceability; warranty. | Anything, including vehicle packs. |
| Grade B | Out of specification on at least one parameter, but functional. | Typically capacity 3–8 % below rating, or IR above limit, or wide dispersion, or cosmetic damage. Usually no warranty and often no traceable batch data. | Low-stakes, non-vehicle, non-financed applications — if at all. |
| Grade C | Significantly out of specification. Sometimes recovered from scrap; sometimes with known self-discharge faults. | A cell with a defect, sold as a discount. | Not a vehicle pack. Not a financed asset. |
| Refurbished / second-life | Harvested from used packs, re-tested and re-graded. | Legitimate when honestly labelled and matched. Highly variable. | Stationary storage with honest labelling. Not vehicles. |
| “Grade A+” | A marketing invention. There is no such bin. | Whatever the seller decides it means. | Ask which specific parameter band it refers to. If there is no answer, there is no grade. |
Important
The distinction that matters is not “strong cell versus weak cell”. A cell that failed the K-value screen is not a slightly weaker cell — it is a cell with an internal short, and it will fail in the field. Capacity dispersion is a performance problem; self-discharge is a defect.
In plain English
Grade B cells cost 25 to 40 per cent less and pass a bench test perfectly on day one. They fail at month six to eighteen — inside warranty, in the customer’s hands, at your cost. On a financed pack, a cell failure is not just a warranty claim. It is a defaulted instalment, an angry customer and a lender asking questions.
4.6 — Yield is the whole business
Everything above collapses into one number. Two lines built to identical designs, one running at 70 per cent yield and one at 95 per cent, cost roughly the same to build and operate and have completely different economics — the low-yield line is producing a third of its output as Grade B and C, which it then has to sell into the discount market to recover anything at all.
This has two consequences worth carrying around.
- •The discount cell market exists because of yield problems. Grade B cells are not a product line; they are the output of a line that missed. A market flush with cheap Grade B cells is telling you something about the quality of production upstream.
- •Scale and yield are the same problem. Yield improves with accumulated production experience, and experience requires volume, which requires surviving the low-yield years. This is why sub-scale cell plants struggle everywhere, and it is the mechanism behind the manufacturing economics we cover in India’s battery manufacturing catch-up and Europe’s failure to build one.
4.7 — What to ask a cell supplier
The process above generates records. Asking for them is not an imposition, and the manner of the refusal tells you as much as the data would.
- •The batch test report for your specific cells — capacity, OCV, internal resistance and K-value, per cell or at minimum as a distribution. Not a generic model report.
- •The K-value acceptance limit. Ask what it is in mV/day and what percentage of production it rejects. A supplier who does not screen self-discharge will not have an answer, and that is the answer.
- •The factory, not just the brand. Large manufacturers run multiple plants with genuinely different quality records. “A CATL cell” is not a specification.
- •The date code. Cells more than six to twelve months old have already spent calendar life you are paying for. Check it on arrival, against the purchase order.
- •The storage state of charge on arrival. Should be around 30 to 50 per cent. Cells arriving at 0 per cent have been over-discharged in transit; cells arriving at 100 per cent have been ageing hard on a ship.
Then verify a sample yourself. The incoming inspection procedure is in Chapter 15, and the matching criteria that decide whether a batch can become a pack at all are in Chapter 11.
Quick check: test yourself
1.A supplier offers cells at 30 % below market and says they are “Grade A, just an older batch”. What do you check?
Show answer
2.Why is a cell that fails the K-value screen categorically different from one that is 5 % under capacity?
Show answer
3.Why does shortening the formation step save money and cost life?
Show answer
Chapter summary
- ✓Cell manufacturing is a process industry closer to semiconductor fabrication than to assembly. The variables that decide quality — coating uniformity, drying, calendering, moisture, particle contamination — are process conditions, not purchasable parts.
- ✓Formation is where the SEI is built and the cell’s character is fixed. Shortening it improves throughput and shortens life, invisibly.
- ✓Ageing and the K-value screen catch cells with internal micro-shorts. This is the measurement that prevents the “one permanently low cell” failure three months into service.
- ✓Grade A, B and C are trade terms with no standard behind them, decided at the final binning step. Grade A+ does not exist as a bin.
- ✓A cell that failed the self-discharge screen is defective, not weak. Capacity dispersion is a performance problem; self-discharge is a defect, and it does not belong in a vehicle pack.
- ✓Yield is the whole economics of a cell plant, and the discount cell market exists because of yield misses upstream.
- ✓Ask for the batch test report, the K-value acceptance limit, the factory, the date code and the arrival SOC. How a supplier declines is itself information.
Frequently asked questions
What do Grade A, Grade B and Grade C cells mean?+
They are trade terms with no international standard behind them, decided at the final binning step. Grade A passed every factory test inside specification, with full traceability and warranty. Grade B is out of specification on at least one parameter but functional — typically capacity 3 to 8 per cent low, or resistance above limit, usually without warranty or batch data. Grade C is significantly out of specification, sometimes recovered from scrap, and does not belong in a vehicle pack.
Is a Grade B cell just a slightly weaker cell?+
Not necessarily, and the distinction matters enormously. Capacity dispersion is a performance limitation you can design around by matching. But a cell that failed the self-discharge screen has an internal micro-short from particle contamination — that is a defect that grows, and it will present three to six months later as one permanently low cell dragging a whole series string.
What is the K-value of a battery cell?+
The rate at which a cell’s open circuit voltage falls during the ageing stage of manufacturing, quoted in millivolts per day. A healthy cell holds its voltage; a cell with a microscopic internal short leaks charge and drifts down. It is the single most important screening measurement in cell manufacturing. Ask your supplier for their acceptance limit — if they do not have one, they are not screening for this.
What is formation in battery manufacturing?+
The first controlled charge and discharge cycles a cell receives, during which the SEI film forms on the anode. The profile used determines how stable that film is. Formation is slow and ties up expensive equipment, so shortening it is a standing temptation — a cell formed too quickly gets a less stable SEI, which means faster capacity fade and more gas generation over its life, none of it visible on delivery.
Why do sub-scale battery factories lose money?+
Because yield is the whole business. Two lines built to identical designs, one at 70 per cent yield and one at 95, cost roughly the same to build and operate and have completely different economics — the low-yield line produces a third of its output as Grade B and C. Yield improves with accumulated production experience, which requires volume, which requires surviving the low-yield years.
Reviewed by
Battery Fundamentals is an original educational series on lithium battery technology. Threshold tables, cycle-life curves and worked examples use representative values drawn from published product specifications and widely-observed industry patterns, not measured data from a specific product. Always consult the current manufacturer datasheet before making design, purchasing, warranty or certification decisions.