A two percent capacity gap between two cells from the same batch looks like a materials problem. More often it is a measurement problem, and the first suspects are the fixtures, the sense wiring and the current range. Before you blame the chemistry, it is worth checking the test setup.
Contact resistance is the usual suspect
If the cycler measures voltage across the same two wires that carry the current, the voltage drop in the leads and fixtures is added to the cell voltage. At high currents this shifts the charge cutoff and discharge cutoff, so the cell reaches its limits early or late. A four-wire (Kelvin) connection, with separate sense leads, removes most of that error. Check that the fixtures, crocodile clips or coin-cell holders make clean, firm contact. Worn springs and oxidised clips are common sources of error that look like bad cells.
Accuracy is usually quoted as a percentage of full scale
This one catches many labs. Suppose a channel is rated at 6 A with an accuracy of 0.05 percent of full scale. That is about 3 mA of possible error. If you are testing a coin cell at a current of 2 mA, the error is larger than the signal. For small cells, low rates or self-discharge style measurements you need a channel with a lower current range, or multiple ranges on the same channel. When you compare a Neware Battery Tester, or any battery cycler, always look at the accuracy at the current you will actually use, not the headline maximum.
Temperature, rest steps and cutoffs
A battery cell’s capacity moves with temperature. A cycler sitting next to an air conditioner, or cells in a chamber with uneven airflow, will show scatter that has nothing to do with the cells. Put cells in a temperature chamber, give them time to settle before the test starts, and keep the placement consistent.
Also look at the program itself. If the constant-voltage step ends at a current cutoff, small differences in measurement at that point change how much extra capacity is counted. Rest times between steps matter too, because a cell that has not relaxed gives a different voltage than one that has. Two programs that look identical on paper can behave differently if one has a short rest.
What to look for when choosing a system
If you are shopping for a battery testing system manufacturer in China, start from your cells, not from the catalogue. Write down the chemistry, capacity range, maximum current, voltage window and the tests you will run: formation, cycle life, rate capability, pulse tests. Then match them against the datasheet. For a series like the Neware BTS4000, the datasheet is where you confirm which voltage and current ranges, and which channel counts, fit your cells. Do not rely on a summary.
Ask how the channels are controlled. Independent channels let you start, stop and change programs on one cell without affecting the others, which matters in a busy lab. Ask about data logging, how often it records and under what conditions, and whether it can export the raw data without manual cleaning. Large cycle-life tests produce a lot of data, and software that is slow with it is a daily annoyance.
Finally, ask about calibration. Any battery cycler drifts a little over time, so find out how it is calibrated, how often, and whether you can verify it yourself with a precision reference cell or meter. A manufacturer that explains this clearly is usually one that takes measurement seriously.
Whichever system you choose, keep one reference cell that never leaves the lab, and run it on every channel every few weeks. If its capacity starts to wander, the cycler is telling you something long before your real cells do.
Battery cycler and battery testing system details are on the Neware homepage: https://www.newarebts.net