Grade A or Grade B: how to recognise original EVE cells and avoid paying for rejects

If you are building a home energy storage system on LiFePO4 cells, you have almost certainly noticed something odd: the same 314 Ah EVE MB31 cell costs noticeably more from one seller than “the same” cell on a marketplace. The description is identical, the photos look alike, and the price difference is tens of dollars per cell. In most cases the explanation is simple: the cheaper cells are not Grade A but Grade B, or even used cells with new stickers.

In this article we look at what the Grade A and Grade B labels really mean for EVE cells, where the “cheap” cells on the market come from, how to check a cell before and after buying (by QR code, appearance and measurements), and whether it ever makes sense to buy Grade B.

What Grade A and Grade B are, and why it is not marketing

The EVE Energy factory produces millions of cells, and each one goes through final inspection: actual capacity, internal resistance and self-discharge are measured, and the case geometry is checked. Cells are sorted into categories based on the results.

Grade A: cells that passed factory inspection

Grade A is a cell that fully meets the manufacturer's specification: capacity no lower than rated (314 Ah for the MB31), internal resistance within the datasheet range, normal self-discharge, a case without defects. These are the cells EVE supplies to official partners and battery system manufacturers. Grade A is covered by the factory specification for cycle life, which for modern LFP cells is thousands of cycles.

Grade B: factory rejects

Grade B cells come off the same line but fail one or more inspection criteria. The most typical reasons:

  • Capacity below rating. A “314 Ah” cell actually delivers, say, 295–305 Ah. It formally works, but you pay for capacity that is not there.
  • Higher internal resistance. The cell heats up more under load, its voltage sags faster and it holds current less well. In a pack, such a cell unbalances the whole battery.
  • Higher self-discharge. A sign of internal micro-defects. These are the cells that drift away from their neighbours in voltage over time, keeping the BMS constantly busy balancing.
  • Cosmetic defects: dents, scratches on the case, terminal defects. This is the safest B category, but it still means the cell did not pass full inspection.

It is important to understand: Grade B is not a “slightly worse” cell with a warranty. It is a product for which the manufacturer explicitly disclaims responsibility. EVE does not give it a full specification and does not guarantee its cycle life.

A third category: used and “repackaged” cells

A separate story is cells that have already served in a system (for example in telecom cabinets or electric vehicles), were removed, “refreshed” and sold as new. Formally these are not even Grade B: they are second-hand goods. Re-engraved QR codes and stickers over the original markings are most common among these cells.

Where cheap cells on marketplaces come from

Factory rejects are not destroyed: they are legally sold on the Chinese domestic market at a much lower price. Some of these cells then reach traders who export them as “new EVE cells”. Formally they are not lying: the cells really are new and really are EVE. They are simply the part of production the factory itself did not release for sale as full-grade.

So when you see an MB31 cell on a marketplace 20–30% below the market price of Grade A, it is not that “the seller found cheaper logistics”. It is almost always Grade B or used. Every market player knows the price of Grade A cells in large volumes, and there are no miracles: nobody sells full-grade cells at a loss.

How to check an EVE cell: step by step

The check has three levels: QR code, visual inspection and measurements. Let's go through each.

Step 1. The QR code on the case

Every EVE cell has a QR code laser-engraved on its case. It is not a sticker but an engraving on the case itself: it cannot be re-stuck, only ground off or re-engraved.

What to know about the structure of the code:

  • The full code of an EVE cell consists of 24 characters and is usually split into two parts: a longer first part (the serial part) and a shorter second one. The encoding follows the Chinese national standard for battery marking.
  • The fourth character of the code indicates the product type: “C” for cell, “P” for pack, “M” for module. If the fourth character on an individual cell is not “C”, that is a reason to doubt its origin.
  • From the code you can decode the manufacturer, type and capacity of the cell and, most importantly, the production date. There are free online decoders for LiFePO4 cell QR codes (for example the Gobel Power decoder) where you simply enter the scanned string.
  • A note on the date: the date in the code corresponds to the production stage before the first charge, so the cell may actually be “finished” a few weeks later than the date in the code. This is normal and not a sign of a fake.

And most importantly: on official Grade B EVE cells, the factory itself partly destroys the QR code : the first part of the code is scraped off at the factory, leaving only the second. This is the factory's direct way of saying “this cell did not pass inspection”. So if you see a cell with part of the QR code ground off, worn away or unreadable, it is very likely Grade B, whatever the seller calls it.

EVE does not publicly document the exact rules for decoding each character, so do not draw conclusions from every individual character; but decoders reliably read the production date, type and capacity, and that is enough for a check.

Step 2. External signs

Inspect the cell carefully, preferably in good light:

  • Condition of the QR code. The code should be sharp, complete and readable by a smartphone scanner. A worn, ground-off or re-engraved code, or a code with part of it destroyed, is red flag number one.
  • Stickers over the markings. Any sticker covering the original laser marking on the case is a reason to ask what is underneath. On genuine Grade A cells the manufacturer has nothing to hide.
  • Terminals. Look at the threads and contact pads. Tightening marks, scuffs, traces of thermal paste or flux, welding marks or bus bar marks mean the cell has already been mounted in a pack. A new cell from the factory has clean, even terminals with no tool marks.
  • Case. Dents, deep scratches, swelling, signs of overheating. Light scuffs from transport are acceptable; deformation is not.
  • QR codes within a batch. If you buy several cells, the codes should be from the same or adjacent batches. A set of cells with production dates spread over six months is a sign that the seller collects them from different sources.

Step 3. Measurements

Even a cell that looks perfect needs checking with instruments:

  • Storage voltage. LiFePO4 cells leave the factory partly charged, usually at about 3.28–3.30 V. A voltage below 3.2 V on a “new” cell is a warning sign: either it was stored for a long time and incorrectly, or it has higher self-discharge. A cell below 2.5 V is effectively damaged by deep discharge.
  • Voltage spread within a batch. Cells from one batch should have very close voltages, within a few millivolts. If one cell clearly stands out, that is the one that will cause problems in the pack first.
  • Internal resistance (ACIR). Measured with a tester such as the YR1035+ at 1 kHz. For large prismatic EVE cells (280–314 Ah), typical values are about 0.15–0.25 mΩ. A value noticeably higher than the neighbouring cells in the batch indicates a problem cell.
  • Real capacity test. The most reliable, if longest, check: a full charge, then a controlled discharge at a standard current (for example 0.5C), recording the amp-hours delivered. A Grade A MB31 cell should deliver at least 314 Ah; in practice good cells usually show slightly more than rated. If the cell delivers 290–300 Ah, that is exactly the “reduced capacity” of Grade B.
  • Self-discharge test. Charge, record the voltage, leave for 2–4 weeks and measure again. A noticeable voltage drop compared with other cells in the batch indicates an internal defect.

Why the production date is not a minor detail

LiFePO4 chemistry is known for its longevity, but that applies to correct use, not careless storage. A cell that has sat in a trader's warehouse for two years, perhaps fully charged and in the heat, degrades before its first working cycle: internal resistance rises and some capacity is lost. And none of this is visible from the outside.

So the freshness of a batch is a real quality factor. Ideally no more than 6–12 months pass between production and purchase. Cells older than a year or eighteen months are a reason to ask for a discount and always to test capacity. And that is why an honest seller readily shows photos of the QR codes and states the production date before payment, not after.

Is it ever worth buying Grade B?

The honest answer: sometimes yes, but not for a home storage battery.

Grade B can make sense for non-critical tasks: a garage power bank, powering tools, experiments, educational projects, where a failed cell is a minor inconvenience rather than the loss of a system. If you knowingly buy honestly labelled Grade B at a correspondingly low price and understand the risks, that is a legitimate choice.

But a 15 kWh home storage battery is 16 cells connected in series that have to live together for 10–15 years. In such a pack, the weakest cell determines the behaviour of the whole battery: it reaches the voltage limit first on charge and discharge, cutting the system's real capacity, constantly loads the BMS balancer and degrades faster than its neighbours. After a year or two you have a battery running at 70–80% of its rated capacity, and a choice: replace the cell (and find one with matching characteristics) or live with it.

Do the maths: the price difference between Grade A and Grade B for a set of 16 cells is usually 10–20% of the cost of the set. Losing even 10% of capacity and halving the cycle life eats up that saving many times over. For a system that has to work every day for a decade, Grade B is bad maths.

How we approach this at SKAT

We work with direct supply and sell only Grade A cells with intact, readable QR codes. Every batch goes through incoming inspection: storage voltage, internal resistance and a sample capacity test. Before buying you can request photos of the QR codes and the production date of specific cells; we consider this normal practice, not a service.

In the catalogue:

Frequently asked questions

Can Grade A and Grade B cells be mixed in one pack?

No, that is the worst scenario. In a series pack all cells carry the same current, and the weakest one determines the capacity and behaviour of the whole battery. A Grade B cell with higher resistance or self-discharge will keep drifting out of balance, the BMS will cut charge and discharge because of it, and the uneven load will speed up its degradation. A pack should consist of cells of the same grade, the same batch and with close actual characteristics.

What voltage should a new EVE cell show?

The typical factory shipping voltage is about 3.28–3.30 V (a partial charge for safe transport and storage). Values below 3.2 V should raise concern, and a cell below 2.5 V has suffered a deep discharge and is considered damaged. The spread within a batch also matters: cells from one delivery should show practically the same voltage, within a few millivolts.

What if the seller refuses to show the QR codes before purchase?

Take that as your answer. A photo of the QR code and the production date are minimal transparency that costs a seller with honest goods nothing. A refusal, or answers like “all cells are the same” or “photos after payment”, means the seller has something to hide, most often ground-off Grade B codes or used cells. Simply look for another supplier.

What production date is normal for a new cell?

Ideally up to 6 months from the production date; up to a year is acceptable if stored correctly. Cells older than 12–18 months should be bought only at a discount and after a real capacity test: long storage, especially charged and in warm conditions, eats into capacity and cycle life before the cell is even used. Remember too that the date in the QR code may precede the cell's actual completion by a few weeks; this is a feature of factory marking, not a deception.

When you buy cells for a system that will run for years, you pay not only for amp-hours but for predictability. Grade A with confirmed origin, a readable QR code and a recent production date is the foundation on which a home storage battery really delivers its rated life.