If you've got a batch of used 18650s, a charger that shows capacity, or a pile of cells you're not sure about, this guide covers every practical method for testing lithium-ion battery capacity and health. You'll learn how to get accurate capacity numbers, grade cells by internal resistance, and decide which cells are worth keeping.

What "battery capacity" actually means

Rated capacity is what the manufacturer claims at a specific discharge rate (usually 0.2C) to a cutoff voltage (typically 2.5V or 2.75V for 18650s). A cell rated at 3000 mAh delivers 3000 mAh under those exact conditions. Your test conditions may differ, so your number will too.

A cell at 80% of rated capacity is generally considered acceptable for reuse in a pack. Below 70%, it's marginal. Below 60%, retire it.

Method 1: Capacity tester (the right tool)

A dedicated capacity tester like the OPUS BT-C3100, Liitokala Engineer Lii-500, or SkyRC MC3000 discharges the cell at a controlled rate and measures how many mAh came out before it hit the cutoff voltage. This is the most accurate and repeatable method.

How to do it

  1. Charge the cell to full first (4.2V for NMC/NCA, 3.65V for LFP).
  2. Set the discharge current. Use 0.5A for standard testing. Lower currents (0.2A) give slightly higher numbers. Pick one rate and stick to it across your whole batch.
  3. Set the discharge cutoff. 2.8V is a safe choice for NMC 18650s. Going lower adds a few mAh but stresses the cell.
  4. Run the discharge cycle. The tester shows mAh when done.
  5. Record the result against the rated capacity.

Run each cell at least twice if you want high confidence. The first cycle sometimes reads low on cells that haven't been used in a while.

Method 2: Charge counter (less accurate, faster)

Some chargers report how many mAh went into a cell during charging. This is a rough capacity proxy, not a measurement. Charge acceptance is typically 95 to 100% of true capacity, and the number varies by charger, temperature, and how far the cell was discharged before charging. Use it for screening, not grading.

Method 3: Load discharge with a multimeter (DIY)

If you don't have a capacity tester, you can discharge a cell through a known load resistor and track the time.

  1. Measure the resistor value precisely (use a multimeter, not the label).
  2. Connect the fully charged cell across the resistor.
  3. Monitor voltage every few minutes until it hits your cutoff.
  4. Calculate capacity: integrate current over time, or use the average current times the total time in hours.

This works but it's tedious and less accurate than a tester. It also doesn't protect against over-discharge if you walk away.

Measuring internal resistance

Internal resistance (IR) tells you how much the cell resists current flow. High IR means voltage sag under load and heat generation. It's the fastest way to screen a large batch.

A good 18650 from a reputable manufacturer typically reads 50 to 150 milliohms on a DC IR test. Above 300 milliohms, the cell is degraded. Above 500 milliohms, don't use it in a pack.

You can measure IR with a dedicated meter like the YR1030 or with a multimeter using the load resistor method (see our internal resistance guide for the full procedure). Capacity testers with IR measurement (OPUS BT-C3100, SkyRC MC3000) measure it during the cycle.

Grading cells from a batch

When you're building a pack from salvaged or surplus cells, grading is how you match cells so one group doesn't drag down the others.

Grade A

90% or more of rated capacity, IR under 150 milliohms for standard 18650s. Use these in matched groups.

Grade B

75 to 89% capacity, IR under 250 milliohms. Fine for lower-demand applications or mixing within a group if you're careful about matching within the group.

Grade C

Below 75% or high IR. Use for low-drain applications (flashlights, remote controls) or discard.

Matching within a group

In a parallel group, cells should be within 50 mAh of each other and within 20 milliohms of each other. Tighter matching reduces current imbalance when charging and discharging.

Temperature effects

Cold cells test low. If you're testing salvaged cells from a cold environment, let them sit at room temperature for an hour first. Capacity at 10°C can read 10 to 15% lower than at 25°C for standard NMC cells.

When to retire a cell

Retire any cell that:

  • Tests below 60% of rated capacity
  • Has IR above 500 milliohms
  • Shows a flat spot in the discharge curve (indicates lithium plating or dead sections)
  • Gets warm during a 0.5A discharge (should barely get above ambient)
  • Has visible swelling or damage

Tools for the job

You don't need to spend much to get accurate results. A Liitokala Lii-500 costs around $25 and measures capacity and IR for 4 cells at once. For larger batches, the OPUS BT-C3100 or SkyRC MC3000 step up in accuracy and features. You can check the Cell Database to compare rated specs against what you're measuring.

FAQ

How accurate are cheap capacity testers?
Within 2 to 5% for testers like the Liitokala Lii-500 at a fixed discharge rate. That's accurate enough for grading. Lab-grade equipment gets within 1%, but you don't need that for battery pack work.
Does testing damage the battery?
A single full discharge/charge cycle as part of testing causes negligible degradation. Deep discharge below 2.5V does harm cells, which is why using a tester with a set cutoff matters.
Can I test LFP cells the same way?
Yes, but adjust cutoffs: charge to 3.65V, discharge to 2.5V. LFP capacity reads lower at the same discharge rate compared to NMC because the voltage plateau is lower.
How many cells can I test at once?
Depends on your tester. Most 4-bay testers run all 4 simultaneously but may limit total current across all bays. Check your tester's manual.