Most B2B battery inquiries start with a headline: 48V, 100Ah, 5 kWh. Then the buyer asks for the spec sheet, and that is where the comparison actually happens — usually in a spreadsheet, against two or three other suppliers. The sheet is not a formality. It is the document that decides whether your pack fits their charger, their rack, their truck, and their warranty terms.
Having answered these sheets on both sides of the table for years, we see the same six figures decide most deals. Here is what each one means, and the question a buyer should ask before signing.
1. Nominal voltage is a label, not an operating range
A LiFePO4 cell sits at about 3.2V nominal. Sixteen in series gives the familiar 51.2V pack; fifteen gives 48V. Both get called “48V systems” in conversation, and neither behaves the same at the terminals. What matters on the sheet is the pair of cut-offs: the charge voltage the BMS will accept and the discharge voltage where it stops. A 16S pack typically charges near 58.4V and cuts off around 40V; a 15S pack tops out near 54.75V.
If those two numbers do not match the charger and inverter already on site, the pack is either derated or refused by the system. Ask for both cut-offs, plus the per-cell limits, before you compare anything else.
2. Amp-hours mean nothing without the voltage
100Ah at 12.8V is 1.28 kWh. 100Ah at 51.2V is 5.12 kWh. The amp-hour figure is a current rating over time, not a quantity of energy, and sales pages mix the two constantly. Convert everything to kWh before you build a comparison table, and then convert again to usable kWh using the depth of discharge the supplier permits. A pack rated 5.12 kWh at 90% DoD gives about 4.6 kWh the inverter can actually draw.
3. Continuous current and peak current are two different promises
The continuous rating is what the pack can deliver for hours without the BMS intervening. The peak rating is a short pulse — usually ten seconds, sometimes three — for motor start-up or a surge load. Written as a C-rate, a 100Ah pack rated 0.5C continuous delivers 50A; a 2C peak rating means 200A for the stated pulse length.
Undersizing this is the most common cause of “the battery shuts off on hills” complaints. A three-wheeler climbing a loaded incline can pull two to three times its cruising current for a minute, not ten seconds. Ask what the BMS does when the peak window expires, and whether the rating was measured at the pack terminals or at the cells.
4. Cycle life is a test condition disguised as a number
“6,000 cycles” on its own is marketing. The number only means something with four conditions attached: depth of discharge, charge and discharge C-rate, ambient temperature, and the end-of-life criterion — usually 80% of initial capacity. A pack cycled at 0.5C to 80% DoD at 25°C will comfortably outlast the same cells cycled at 1C to 100% DoD in a 45°C cabinet.
Ask whether the figure is cell-level or pack-level, and whether it comes from a third-party lab or an internal report. For a fleet or a storage project, the difference between those two answers is measured in years of service.
5. Dimensions and weight decide the freight, not the datasheet headline
Buyers routinely compare product dimensions and then get surprised by the packed volume. A wall-mounted pack with a shipping carton, foam and pallet can cube out 30–40% larger than the product itself, and lithium cells ship under UN3480 or UN3481 rules with their own paperwork. Ask for packed dimensions, gross weight, and the number of units per pallet. That single line often changes the landed cost more than a 5% price difference.
6. The BMS protocol is where integration projects stall
“CAN communication” is not a specification. There are several common protocol dialects in this industry, and an inverter that speaks one will not talk to a pack that speaks another. Before ordering, confirm the protocol variant, baud rate, connector pinout, and whether the supplier has already commissioned a closed-loop link with the inverter model you plan to use. A ten-minute question here prevents a month of field debugging.
What no spec sheet will tell you
Four things never appear on the page, and all four show up in year two: where the cells came from and whether the batch is traceable, how the pack was assembled (laser welding, torque control, busbar sizing), how heat is managed under sustained load, and what the factory’s outgoing QC actually samples. None of these can be verified from a PDF.
The practical workaround is to ask for the test documentation alongside the sheet: a UN38.3 test summary and MSDS for transport, an IEC 62619 report if the supplier has one, and a short video of the assembly line and the aging rack. A supplier who can produce those quickly is usually a supplier who knows their own process.
How we handle it
We publish nominal voltage, capacity, energy and configuration on every product page, and we list the standards we design to — UN38.3, IEC 62619, CE (LVD/EMC) and RoHS — with the note that the actual certificate status is confirmed in writing with the quotation. Full specification sheets, including cut-offs, current ratings, dimensions and test conditions, are issued with the quote rather than posted publicly, because they change with cell batch and configuration.
If you send the load profile — voltage, continuous and peak current, daily energy, ambient temperature — we will reply with the configuration that fits and the sheet to go with it, rather than a catalogue page.
If you are sizing a residential system, the 5.12 kWh wall-mounted pack page shows the configuration we build most often. Transport and conformity documents are in the download centre.