Charged Reviews

Battery leaderboard

Charged tests

Sorted by share of advertised rating, not ranked. Notes

  1. Humsienk

    105.8% of 100 Ah

    Share of advertised rating

    Ah delivered at 20 A
    105.8 Ah
    Wh delivered at 20 A
    5,465 Wh
    View details

    Model: Humsienk 48V 100Ah 3U server rack LiFePO4 battery, matched by Charged Reviews

  2. ECO-WORTHY Cubix Pro

    104.5% of 100 Ah

    Share of advertised rating

    Ah delivered at 20 A
    104.5 Ah
    Wh delivered at 20 A
    5,401 Wh
    View details

    Model: ECO-WORTHY Cubix100 Pro, matched by Charged Reviews

  3. Vatrer Classic

    102.8% of 100 Ah

    Share of advertised rating

    Ah delivered at 20 A
    102.8 Ah
    Wh delivered at 20 A
    5,305 Wh

    No confirmed match yet.

  4. Vatrer New

    101.8% of 100 Ah

    Share of advertised rating

    Ah delivered at 20 A
    101.8 Ah
    Wh delivered at 20 A
    5,253 Wh

    No confirmed match yet.

  5. ECO-WORTHY Cubix 100

    98.6% of 100 Ah

    Share of advertised rating

    Ah delivered at 20 A
    98.6 Ah
    Wh delivered at 20 A
    5,080 Wh
    View details

    Model: ECO-WORTHY Cubix100 48V 100Ah Server Rack Battery (ECO-LFP4810002), matched by Charged Reviews

  1. WirenTech

    97.7% of 100 Ah

    Share of advertised rating Not compared

    Ah delivered at 20 A
    97.7 Ah
    Wh delivered at 20 A
    5,065 Wh
    View details

    Model: Wirentech WT5100 51.2V 100Ah LiFePO4 Rack Mounted Battery, matched by Charged Reviews

Voltage during discharge

At 20 A, from full until the run ended
Move along a line, or tap and drag it, for its readings.

Volts

Notes

Measured by Charged Reviews on one unit each. Not a certification. Brands and stores don't pay for testing, results or rank. Names are from our test log unless a row says it's matched. Where it isn't, the model isn't recorded, so results may not apply to other versions. Ratings come from the maker's document where a row cites one. Otherwise they're as advertised when bought, per Charged Reviews, unchecked.

Where a row links to our catalog, we don't show test results on that page. Some of those pages have links that earn us a commission.

WirenTech: 97.7% of 100 Ah, not compared. Measured 97.7 Ah, listed apart. The last reading under load was 48.9 V at 20 A. About 2 seconds later it read 37.6 V at about 0 A, and the run ended. The other batteries' last readings under load were 40.7 to 41.5 V, so this figure may understate what this unit can deliver. The cause isn't established: it could be the battery's protection, a connection or the test equipment. Not compared with the others until it's re-tested or the cause is known.

A flatter line holds its voltage longer. Voltage at 25, 50 and 75% of this run's discharge is measured against that unit's own delivered Ah.

How we test batteries

We charge and discharge each battery the same way on our bench, and test each battery as shipped.

Test
20 A from full
Tested
Nov 2 and 3, 2026
Readings
Every 10 seconds

Test method 1.

Capacity at 20 A (0.2C of 100 Ah), room set to 75 °F

Share of advertised rating is Ah delivered ÷ the advertised rating. The bar's line marks 100%. Ah is what labels usually print. Wh is the energy it delivered. Average voltage at 20 A is Wh ÷ Ah delivered, read at the tester, and isn't ranked.

We also tested these batteries at 100 A. Those results aren't published yet because we're still checking them.

Some links earn us a commission; that never affects what we test or how results rank.

  1. UnitWe tested each battery as shipped.
  2. Charge20 A constant current, then 58.4 V constant voltage, until current falls to 1 A.
  3. Rest1 second.
  4. Discharge20 A. The tester was set to stop at 40 V. Our test logs don't show whether a battery's own protection acted first.
  5. RecordVolts, amps, Ah and Wh every 10 seconds.

Capacity ran in two sessions, Nov 2 and Nov 3, with the same program. Room set to 75 °F (24 °C), with a mini split blowing air toward the batteries during tests. Not logged in the test files. Voltages are read at the tester, so they may include losses in the cables.

What changed

Nov 5, 2026: We removed NewtiPower from this table because it's advertised at 135 Ah, so 20 A wasn't 0.2C for it. That isn't about how it performed. We also removed the run-time estimate.

Test method 1, Nov 2026: first results published. A later method gets a new number, and results from different methods aren't compared.

The maker's rating

Makers rate capacity under their own conditions, which can differ from ours. Where a row cites a maker's document, the rating comes from it. Otherwise it's what the battery was sold as, per Charged Reviews. We haven't added makers' printed conditions yet.

Sorted, not ranked

We rank only when we can state the measurement's accuracy. Until then, the table is sorted, not ranked.

Corrections and maker replies

Every result has a link to report a problem with it. We check the raw data and retest when needed. An earlier result stays in the history, marked with the retest date, and a correction is dated.

The Evidence score

Test results are kept apart from the Evidence score, which rates how well a product is documented.

Browse batteries in our catalog