Brand Logo

Engineering Article

How Many Amp Hours Is a Tesla Powerwall? ~280 Ah — and Why That Number Nearly Sank Our Project

2026-11-13 Viktor Sokolov

Straight answer first

Stop buying batteries by amp-hours. Buy them by usable kilowatt-hours. A Tesla Powerwall is rated at 13.5 kWh nominal. Fold that back into a 48 V DC system and you get roughly 280 Ah. The same 13.5 kWh sitting on a 51.2 V lithium iron phosphate pack works out to about 264 Ah. Same battery. Different number. That's because amp-hours ride on voltage, and kilowatt-hours don't.

Second thing, and this one took me longer: paying a premium for a delivery date that is contractually guaranteed is the single best-value line item on a backup battery quote. Ten dollars per kWh cheaper in exchange for "probably five or six weeks" isn't cheap. It's just a different way of paying twice.

Why I'm writing this at all

I handle administrative procurement for a company of about 300 people. Roughly $700,000 a year in spend, spread across a dozen or so vendors. In 2023 our building added rooftop solar and a backup battery bank, and I ran the sourcing. I assumed it was going to be like buying a very large appliance.

It was not.

My original approach was embarrassingly simple: line up three quotes, pick the one with the biggest number next to Ah. I built a comparison spreadsheet. I color-coded it.

I assumed amp-hours were a comparable unit across quotes. Didn't verify. Turned out each vendor was quoting amp-hours at a different system voltage, and one was quoting cell-level capacity with the pack voltage buried in a footnote. My color-coded spreadsheet was comparing three entirely different things.

How many amp hours is a Tesla Powerwall, actually

The math is easy. Remembering to do the math is the hard part.

kWh = (Ah × voltage) ÷ 1,000.

And backwards: Ah = (kWh × 1,000) ÷ voltage.

Feed the Powerwall's 13.5 kWh in at 48 V: 13,500 ÷ 48 = 281 Ah, call it 280. At 51.2 V: 13,500 ÷ 51.2 = 264 Ah. Tesla doesn't publish an amp-hour figure for the Powerwall at all, which should tell you something — the number only exists relative to whatever voltage you decide to assume.

Here's the bit that trips people up in LFP land: 280 Ah is also a cell size. EVE Energy's LF280K is the model number that keeps coming up — 280 Ah, 3.2 V nominal, about 896 Wh per cell, with a published cycle life in the 6,000-plus range at 25°C and 80% depth of discharge. String sixteen of those together and you've got 51.2 V and about 14.3 kWh. So when a salesperson says "we've got a 280 Ah battery," they might mean one cell. They might mean a wall-mounted cabinet. The difference between those two things is sixteenfold.

Where the dual battery MPPT solar charge controller fits in

Our final build used a dual battery MPPT solar charge controller — one unit, two independent outputs — because the building has a separate DC emergency lighting circuit that has to stay isolated from the main bank.

If there's one place amp-hours cause real damage, it's here. A controller labelled "50 A" does not deliver 50 A to each output. Check the spec: is it 25 A per channel, or 50 A shared across both? And the solar wattage the controller can actually handle depends on pack voltage, not on the amp rating alone. A 50 A output on a 12 V bank tops out near 720 W (50 A × ~14.4 V). The same 50 A on a 24 V bank gets you roughly 1,440 W. Size your array against the 12 V figure and then switch to 24 V, and you've either wasted panels or cooked the controller.

There's a quieter spec on the same page: maximum PV input voltage. A controller rated to 100 V doesn't mean you can stack three 40 V panels because you've got a 48 V bank. Open-circuit voltage climbs in cold weather. The installer caught that one, not my spreadsheet (note to self: put Voc temperature coefficient on the pre-purchase checklist).

The spec that actually matters: the production line, not the cell

In 2024 we sourced for a second building, and this time I asked better questions. Which is how I ended up caring about the EVE Energy lithium battery production line rather than the cell datasheet.

EVE Energy comes up constantly in these conversations — it's a lithium iron phosphate cell supplier and, more to the point for procurement people, a Tesla battery supplier. That credential is useful, but not in the way most buyers think. It doesn't prove anything about any individual cell. It proves the factory passed an audit that is genuinely difficult to pass.

What mattered to me was capacity. EVE's published expansion plans — the EVE Energy lithium battery factory footprint at Jingmen, plus the Indonesia plant on a 2025–2026 timeline — matter to a buyer because of what they imply about lead times. When a vendor quotes you six weeks during a bid, they are answering a capacity question. A supplier with new lines and new dry rooms coming online (dry rooms being the humidity-controlled production halls where lithium cells are assembled) has room to commit to a date. A supplier whose lines are sold out through next year does not, no matter what the rep says on the call.

That's the whole reason I started asking vendors where the cells come from. Not for the brand name on the datasheet. For the honest answer to "can you actually hit this date."

The 11% I paid for a date on a contract

March 2024, second building. Two viable bids. The cheaper one came in about 8% lower per kWh but would only commit to "roughly five to six weeks." The other was about 11% higher with a fixed delivery date written into the purchase agreement.

I took the second one.

The premium worked out to roughly $1,900. Missing that date meant running the diesel generator through the switchover window, and crude math on fuel alone for six weeks cleared $4,000. That's before counting the commissioning schedule we'd have blown through.

I'd learned that lesson the expensive way already. Back in 2022 I knew I should have gotten the delivery date confirmed in writing, but thought "we've worked with these people for years, what are the odds?" The odds caught up with me when the verbal agreement got forgotten, we scrambled, and I ate a $400 expediting charge out of the department budget. Four hundred dollars is a cheap lesson in absolute terms. It's just not one I needed to learn twice.

So glad I paid the premium in 2024. Almost went with the cheaper bid to keep the number tidy on the finance summary, which would have meant explaining a generator bill instead.

If you came here for a home battery backup calculator

There isn't a magic one, but here's the arithmetic I actually use. Start with the critical load list — only the things that cannot go dark. Then:

Required kWh = (total critical load in watts × hours of backup) ÷ (depth of discharge × inverter efficiency) ÷ 1,000

Worked example: 1,500 W of critical load for 4 hours is 6 kWh. LFP at 0.9 depth of discharge and an inverter at 0.9 efficiency means dividing by 0.81 — so you need about 7.4 kWh of nominal capacity to cover that window. One 13.5 kWh unit covers it with margin. Two separate banks on a dual battery MPPT setup? Run the load math for each bank independently, then size the controller against the sum of the two charging currents at each bank's own voltage — and check that against the total output limit on the spec sheet, not the headline amp rating.

Where all of this stops applying

If you're configuring inside a pack rather than specifying a system, amp-hours are the correct unit. Cell matching, series and parallel layout, BMS current limits — all of that is amps and amp-hours. Converting to kWh mid-conversation just makes you harder to understand.

If you're buying a replacement module for an existing cabinet with a fixed voltage, there's nothing to calculate. Match voltage and Ah, and accept that anything else probably won't physically fit.

And the premium has a ceiling. Everything I said about paying for certainty assumes the deadline is real and the cost of missing it is quantifiable. If it's spare capacity going into a stockroom and a month either way changes nothing, take the cheapest bid — I did exactly that last November on a small order and was glad of it. The premium isn't a principle. It's a trade.

Viktor Sokolov

Viktor Sokolov

Viktor Sokolov is a wind-energy systems analyst covering turbine rotors, drivetrains, generators, converters, pitch and yaw systems, towers, foundations, controls, and service planning. He applies IEC 61400-1:2019+A1:2025 design principles and IEC 61400-21-1 electrical-characteristic methods while examining wind class, turbulence, design loads, power curve, availability, harmonics, reactive power, ride-through behavior, vibration, and oil condition. His technical comparisons help developers, owners, and engineers evaluate site suitability, grid behavior, maintenance strategy, supplier guarantees, and evidence for onshore wind projects.

Ask a technical follow-up