Six Things to Check Before You Wire a Deposit to a Battery Factory
Forty pallets arrived with correct labels and clean cases, and the capacity test came back at 82 Ah on a 100 Ah block. The factory said the tolerance was normal. Here is what a certificate actually proves, what to read on the test report, and the clauses that end the argument.

A buyer in Valencia wired 30% against a proforma in March. The container arrived in May, forty pallets, labels correct, cases clean. Then the sample test came back: 82 Ah on a 100 Ah block, measured at the ten-hour rate, all four samples between 81 and 83.
The factory's reply was three lines. Capacity tolerance is normal. The batteries are new. Please send a video.
Nothing in that reply is technically a lie, and that is the problem. "100 Ah" is not a number until somebody says at what rate, at what temperature, down to what end voltage, on which cycle. A supplier who will not put those four things in writing is not necessarily cheating you. They are simply keeping the option open, and you are the one holding the risk.
So here is what to actually check, in the order that costs you the least money.
A certificate proves that a document exists
This is the uncomfortable part. Almost every quotation you receive will have a folder attached: CE, ISO 9001, MSDS, maybe a test report. Buyers treat the folder as evidence. It is not evidence. It is a claim with a logo on it.
| Document | What it actually proves | What it does not prove |
|---|---|---|
| CE declaration | The product is claimed to meet the directives named on the paper | That the model you ordered was the one tested, or that a notified body looked at anything |
| ISO 9001 | The factory has a documented quality process | That the process produces good batteries, or that the certificate covers this product line |
| MSDS | The hazards and transport classification of a chemistry | The quality, capacity or age of what is in the box |
| UN38.3 test summary | Cells and packs passed T1–T8 for transport | Capacity, cycle life, or that the cells in your order came from that batch |
| A capacity test report | What one sample did, once, under stated conditions | Consistency across the batch — unless the sample size is stated |
None of this means the documents are worthless. It means you have to read the scope line, not the logo. A CE document that names EMC and nothing else is not a safety certificate, and it was never meant to be.
We will use our own paperwork as the example, because it is easier to be blunt about your own. We hold two CE documents: one for the stationary VRLA range, one for the LiFePO4 range. The lithium one covers EMC. When a buyer asks whether that is a full safety certification, the answer is no, and we say so in the first email rather than the tenth. If a supplier cannot tell you the exact scope of their own certificate in one sentence, that is your first data point.
The report behind the certificate matters more than the certificate
Ask for the test report, not the certificate. Then read four things on it: the model, the standard, the discharge rate, and the end voltage. If any of those is missing, the number on the front page is decoration.
| Field on the report | Why it changes the answer |
|---|---|
| Exact model and capacity | A report for the 200 Ah block says nothing about the 100 Ah one |
| Standard, with the part number | IEC 60896-21 and IEC 61427 test very different things |
| Discharge rate (C10, C20, C5) | The same block is 100 Ah at C20 and about 85 Ah at C5 |
| End voltage per cell | 1.75 V/cell and 1.80 V/cell are different products from the same box |
| Ambient temperature during test | Capacity is referenced to 25 °C; a 35 °C test looks better than it is |
| Cycle number | First-cycle capacity is always the best number you will ever see |
| Sample size and spread | One sample proves one battery; four samples with a spread prove a process |
That last row is the one buyers skip and the one that decides whether you have a supplier or a lottery. A report showing four samples at 101, 102, 100 and 103 Ah tells you the process is controlled. A report showing one sample at 104 Ah tells you nothing at all except that they picked a good one.
Three questions that separate a trader from a factory
You can usually settle this in one call, without a flight.
Which line makes this model, and how many lines do you have? A trader will describe the product. A factory will describe the line, the shift pattern, and which line is running your model this month.
Who charges the formation current, and at what rate? Formation charging is the step that decides whether the plates are properly converted. Anyone who has stood on a factory floor knows the answer without looking it up. Anyone reselling does not.
Can I have the last three shipments' capacity test data for this model? Not a certificate. Not a summary. The actual sheets. A factory that tests routinely has them in a folder. A trader will offer you a discount instead.
None of these questions are hostile. Ask them the way you would ask a mechanic whether they replaced the filter, and you will get a more useful answer than any audit checklist.
Tests you can run before the container ships
You do not need a lab. You need a scale, a multimeter, an internal resistance meter, and one afternoon.
| Test | Tool | Pass line | What it catches |
|---|---|---|---|
| Weight vs datasheet | A scale, ±10 g | Within 2% of the stated weight | Less lead than the design calls for — the oldest trick in the industry |
| Open circuit voltage | Multimeter, after 24 h rest | 12.80–13.00 V for a 12 V AGM at 25 °C | Undercharged or aged stock being shipped as new |
| Voltage spread across a batch | Multimeter, 20 blocks | Within 30 mV of each other | Mixed ages, mixed production dates |
| Internal resistance | Resistance meter | Within 10% of the batch median | Loose welds, dry joints, inconsistent plate pressure |
| One capacity test at C10 | Load bank or a known resistive load | ≥95% of rating on the first cycle | The number the invoice is actually about |
| Date code | Eyes and a torch | Within 3 months of shipment | Stock that has been sitting since last season |
The capacity test is the only one that costs you real time, and it is the only one that settles an argument. Do it on one block per 500, chosen at random from the middle of the pallet, never from the top layer where the good ones live.
If a supplier refuses to let your inspector pull from the middle, you have learned something worth more than the test.
Put the acceptance criteria in the purchase order
Everything above is a conversation until it is a clause. These are the clauses that actually get used, in the order they get used.
| Clause | The exact wording that works |
|---|---|
| Capacity | "Capacity at C10, 25 °C, to 1.75 V/cell, shall be not less than 95% of rated on the first cycle; verified on 3 samples per container." |
| Consistency | "Voltage spread within any string shall not exceed 30 mV at delivery." |
| Age | "Date of manufacture shall be within 90 days of the shipment date; date code legible on the case." |
| Weight | "Unit weight shall be within 2% of the approved sample." |
| Documents | "UN38.3 test summary and MSDS for the exact model, issued within 24 months." |
| Remedy | "Batch failing any of the above to be replaced or credited at the seller's cost, including freight." |
The remedy line is the one that makes the other five real. Without it you have a specification, not an agreement.
And when something does go wrong, the evidence you need is the same evidence you should have collected at receiving: photographs of the date code, the weight, the voltage of every block in the string, and the resistance reading with the meter model visible. A claim with those four things gets settled in a week. A claim that says "the batteries are bad" gets a request for a video and then silence.
We are on the wrong side of this argument often enough to have a view. The buyers who get fast resolutions are not the ones who negotiated hardest on price. They are the ones who wrote down what they would accept before they paid, and then measured it when the container opened. It is a boring habit that saves an unreasonable amount of money. If you want the template we send to new distributors, it is on the [OEM page](/oem), and the warranty wording that goes with it is on the [quality page](/quality).
Looking for a reliable battery supplier?
Send us your requirements — model, quantity and target market. Quote within 24 hours.
Get a Quote