Company Blog
High quality high performance excellent service
Search productSearch post
High quality high performance excellent service
Most battery packs fail for a boring reason: nobody wrote the numbers down before anyone ordered a cell. The conversation jumps straight from “we need 12 volts” to a quotation, and the pack arrives with more energy than the device needs, or less energy than the device quietly assumes. Both failures cost the same amount of money and take about the same time to fix.
What follows is a sizing method you can run yourself, in five steps, before you talk to any supplier. It is deliberately a method and not a formula, because the arithmetic is the easy part — the judgement calls are where packs go wrong. By the end you will have a worksheet with a voltage, a capacity, a BMS configuration, a connector, and a defensible quantity. If your project is a custom lithium battery pack rather than a catalogue item, these five steps are what a good supplier runs anyway; doing them yourself means the quotation and the sample approval conversation start from facts rather than adjectives. The same worksheet also works when you are buying a standard model with an unusual lead, or when a custom lithium battery pack is the outcome of a design that has outgrown its original cell.
Two quick notes on the numbers below. Everything is worked in nominal voltage because that is what cell datasheets use. And every calculation carries a derating factor, because a battery that meets a runtime target on a bench at 25 °C will not meet it in a hot enclosure in August.
You cannot size a pack without knowing what it has to do, and “it draws about 500 mA” is not a load profile. You need four numbers, and the fourth one is where most estimates quietly fail.
If you do not have current measurements yet, get them. A USB power meter and a bench supply will tell you what a datasheet does not, and the difference between a device that draws 180 mA and one that draws 620 mA is the difference between a 2 Ah pack and a 7 Ah pack. Vendor documentation often lists the maximum, not the average, and sizing to the maximum is how people end up carrying dead weight.
Capacity comes from energy, not current alone. The chain is short:
1. Required energy (Wh) = average current (A) × voltage (V) × hours of use per cycle
2. Raw capacity (Ah) = required energy (Wh) ÷ nominal voltage (V)
3. Nameplate capacity = raw capacity ÷ usable fraction ÷ temperature factor
Target = nameplate capacity — and then round up to the next cell format that actually exists.
The third line is where sizing goes wrong. A reasonable set of derating assumptions:
| Factor | Typical value | Why it is there |
|---|---|---|
| Depth of discharge | 0.80–0.90 | Cycle life is quoted at a specific DoD; running to empty shortens life and leaves no reserve. |
| End-of-life capacity | 0.80 | The datasheet number is new-cell capacity. The pack should still meet its runtime target at end of life. |
| Temperature | 0.85–0.95 | Cold reduces usable capacity; heat reduces it too and accelerates wear. |
| Wiring and BMS losses | 0.95–0.98 | Thin leads, a connector with a few tenths of a volt of drop, a protection board dissipating something. |
Multiply them together and you are typically looking at a 1.3–1.5× multiplier on the naive calculation. That sounds aggressive until you run the arithmetic on a real case.
Consider a battery for CCTV cameras — a single camera on a pole, running 14 hours a night on a 12 V nominal system, drawing 1.1 A average with a 3.5 A peak when the IR illuminator switches on. Take 3.7 V cells in a 3S configuration: 11.1 V nominal.
Energy per night = 1.1 A × 11.1 V × 14 h = 171 Wh
Raw capacity = 171 Wh ÷ 11.1 V = 15.4 Ah
Derated (0.85 DoD × 0.80 EOL × 0.90 temperature × 0.97 losses = 0.594) = 15.4 ÷ 0.594 = 25.9 Ah
Target: a 3S pack in the 26 Ah class — roughly 5 Ah cells in 2P, or 2.6 Ah cells in 5P.
Now the honest part: a 2.6 Ah pouch cell in a 5P configuration makes for a large, soft pack with more cell-to-cell variation to manage. If the enclosure has room, 5 Ah cells in 2P are usually the better engineering answer — fewer parallel strings to balance, better current capability per cell, and a pack that is easier to keep consistent from batch to batch. KBT’s 3.7 V polymer lithium-ion packs from 30 mAh to 10,000 mAh cover the cell formats this step actually lands on, and the same calculation applies if you move to cylindrical cells — the 18650 cells at 2,200–3,350 mAh are the workhorse option for 3S and 4S builds. That is a decision the worksheet cannot make for you, but at least you now know the question is about 26 Ah in 3S, which is a much better place to start a conversation than “we need a 12 V battery.”
One thing the calculation never tells you: the surveillance use case has a hidden constraint. Many cameras and recorders draw a small but continuous current while “off” to keep the RTC and detect PIR motion. If your device sleeps rather than powers down, add that standby draw to the average, and specify a BMS with a low quiescent current or your runtime will be shorter than the arithmetic promised. A second, less obvious requirement for battery for CCTV cameras is outdoor temperature — a pole-mounted enclosure sees both cold nights and hot afternoons, and charging happens at the base station, not at the camera. KBT’s 11.1 V 5,200 mAh DC5521 pack for CCTV, LED lighting and OEM/ODM projects is the shape of pack this calculation lands on for lower-power installations, with a stated 5 A maximum continuous discharge and a matched 12 V charger.
The worksheet gives you a nameplate number. Runtime is what the user experiences, and the gap between them is where complaints come from. Three adjustments catch most of it:
One more runtime factor worth stating plainly: standby draw. A pack specified for 14 hours of operation and 10 hours of standby is not a 14-hour pack. Add standby into the hours line of Step 2 before you size, not after.
Once the numbers are fixed, the BMS turns into a specification with decisions in it. Four parameters carry most of the meaning:
| Parameter | How to choose it | What goes wrong if you get it wrong |
|---|---|---|
| Continuous current | At least the measured average, with headroom; the charge current rating often needs to be higher still. | Thermal throttling or shutdown in an enclosed, hot installation. |
| Peak current / response time | Must cover the shortest measured peak and survive the required trip delay. | Protection trips during a legitimate inrush, or fails to trip on a real fault. |
| Cell chemistry | Li-ion, LiFePO4 and Li-polymer use different charge and cut-off thresholds. KBT’s large-capacity LiFePO4 energy storage batteries are the reference point if your pack sits in a storage application rather than a handheld one. | Over-charge, or a battery that is cut off while it still holds usable energy. |
| Balance method | Passive balancing suits slow rates; active balancing suits high-capacity parallel groups. | Cell divergence in a multi-parallel pack, which shows up as reduced runtime years later. |
Three configuration questions to settle before the sample, because they affect tooling and cannot be changed cheaply afterwards:
The connector is the smallest decision on the list and the one that most often causes a return. It is a mechanical part that has to carry current, survive vibration, and be matched to whatever the customer actually owns.
Four rules cover most of it:
KBT builds several DC5521-output packs, including an 11.1 V 2,400 mAh pack with DC5521 male output rated to 3 A continuous — a useful contrast with the 5 A 5,200 mAh unit, since the current rating is a property of the pack and its protection, not of the connector family. If you are talking to a DC5521 battery pack manufacturer, the useful questions are about the crimp and the current rating, not the moulding: what gauge of conductor is crimped to the pin, whether the contact is plated or bare, and what continuous current the assembly is rated for. Where the enclosure is exposed, the outdoor waterproof pack with DC5521 connector shows the same output format combined with a sealed housing.
Not every project needs a production run. Prototypes, pilot builds, a single product launch, or a customer who needs to hold inventory: there are legitimate reasons to buy small batch lithium battery packs instead of committing to thousands of units. Two practical realities govern how this works.
First, standard catalogue models are usually orderable by the carton, which is often the cheapest path to a handful of units. Custom configurations — a different cell, a different enclosure, a different connector, custom labelling, or a certification requirement — carry a minimum order quantity that depends on which of those you change. The cost drivers are predictable: cells and tooling, a custom enclosure, and certification. A pack that differs only in label or lead length is a much lighter commitment than one with a new shell. If your pilot is small enough that a standard pack plus a custom lead will do, that is a legitimate way to start — and because the worksheet above is the same either way, a small batch lithium battery packs order and a production order can share one specification document.
Second, small batches and custom work are compatible, but the process is the same one used for large runs — only the quantities differ. On KBT’s support page the project path runs requirement review, engineering evaluation and feasibility, sample customisation, pilot run, sample validation, then batch production with IQC, PQC, TQC and final QA. A 50-unit pilot can sit on that path without a shortcut, and running it properly is what protects you when you scale to volume later: the sample you approve is the same build that production will follow.
The trap in small-batch buying is skipping the validation step to save time. The point of a pilot run is not the units themselves — it is discovering whether the battery fits the enclosure, whether the connector is right, whether the BMS thresholds behave as intended, and whether the runtime calculation matched reality. Those are cheap to find at 50 units and expensive to find at 5,000. A run of small batch lithium battery packs is therefore a specification exercise as much as a purchase, and the worksheet at the end of this article is the document that makes it one. KBT’s support and OEM project page sets out the same sequence and is worth reading before you send the first brief.
Here is the whole method compressed into the artefact you would actually keep. Fill in the left column, and you have a brief that any competent supplier can quote against without a single clarification call.
| Line | Your answer |
|---|---|
| Average current (A) / peak current (A) and duration | |
| Hours per cycle / days per week | |
| Device voltage range; cell configuration (nS) | |
| Required energy (Wh) and raw capacity (Ah) | |
| Derating factors and final target capacity (Ah) | |
| Cell format and parallel count (target and acceptable alternate) | |
| BMS: continuous / peak current, chemistry, balance method | |
| BMS features: fuel gauge, temperature protection, low quiescent current | |
| Connector: family, gender, polarity, current rating, mating lead included | |
| Enclosure: dimensions, material, sealing, cooling | |
| Quantity: prototype, pilot, annual forecast | |
| Required documents for the destination market |
Send that with an inquiry and you will get a quotation you can compare against another quotation, because both will be priced against the same numbers. KBT confirms the applicable documents for the specific model against the destination market before quotation, and lists what to provide — application, voltage, capacity, maximum dimensions, continuous and peak discharge current, connector and wire requirements, quantity, and target country — on the support page, with a contact route on the contact page.
The method is not exotic, and it is not proprietary. Its value is that it moves the conversation away from voltages and towards requirements, which is the only place where a sizing decision can actually be defended — to your engineering team, to your customer, and later to whoever has to explain why the pack was that size.
Tags: OEM battery sourcing, custom lithium battery packs, battery procurement, cylindrical cells Total body word count: 1,537 Reading time: 8 minutes Choosing an OEM battery manufacturer is a product-risk decision, not simply a search for ...
View detailsLearn how to specify a custom lithium battery pack for OEM projects, including voltage, capacity, BMS, connectors, cell type, UN38.3, MSDS, UL and CE.
View detailsLearn how to specify a custom lithium battery pack with cable and connector. Use this buyer checklist and contact KBT for an OEM quote.
View detailsThe short answer is sometimes, but matching the "12V" label is not enough. A pack can fit the compartment and still be wrong for the motor surge, charger, connector or wet environment. So, can a lithium battery replace lead-acid in your product wi...
View details