Most charging guides stop at the chemistry. This one starts where procurement actually gets stuck: how much capacity a shift really needs, whether a charger will talk to the battery, what cold storage does to your charging window, and what the payback looks like once real conditions are included.
The charging rules themselves are short, so they come first. Everything after that covers what those rules mean for a fleet: capacity sizing, charger matching, cold-store warm-up, and ROI.
The charging rules, in full, before we move on
- Charge with CC-CV. Constant current up to the pack termination voltage, then constant voltage until the current tapers to the cut-off the BMS specifies.
- Match the charger to the cell count, not the nominal voltage. A "48V" pack can be 13S NCM, 15S LFP or 16S LFP, and those termination voltages are not interchangeable.
- Charge and discharge windows are different. A pack rated to operate from -20°C to 60°C is not a pack you may charge at -20°C. Standard LFP packs must not be charged below 0°C unless an integrated heater brings the cells above the threshold first.
- Depth of discharge, charge rate and cell temperature are the three variables that drive ageing. Everything else is secondary.
Capacity Selection: More Complicated Than You Think
The most common question buyers ask is "Is 400Ah enough?" That cannot be answered directly, because "enough" depends on too many variables that sit on your side of the quotation.
Calculate daily energy consumption first. Don't skip this.
Daily consumption (kWh) = Average power (kW) × Operating hours (h) × Load factor
Load factor depends on actual working conditions: light duty 0.3~0.4, medium 0.5~0.6, heavy 0.7~0.8. The most common sizing error is using rated plate power times hours, which overstates consumption by roughly 30% against metered draw.
Example: 2-ton electric forklift, rated power 8kW, operates 10 hours daily, medium load.
Daily consumption = 8 × 10 × 0.55 = 44kWh
In a 48V system, 44kWh corresponds to roughly 920Ah. Since you shouldn't discharge below 20%, usable capacity is about 80%, meaning you need around 1150Ah to get through a day on one charge.
But that's just theory.
Calculated and actual consumption rarely match. Operator habits, floor gradients, cargo weight variation and cab heating or cooling all sit outside the formula. The practical response: add roughly 20% buffer to the theoretical value, or meter a rented unit over a month before committing to a number.
Is bigger always better? Not necessarily.
Buying well above the calculated requirement "in case we grow" is a common decision, and it has three predictable consequences:
Problem 1
The larger pack adds mass, and on a counterbalance truck that mass can cost you rated fork capacity
Problem 2
Charger power has to be upgraded to match, and that can pull electrical infrastructure work in behind it
Problem 3
If the growth does not arrive, the pack spends most of its life at a mid-range SOC, which does nothing to help calendar ageing
The practical view: If current throughput is stable, size to what you actually need (theoretical value + 20% buffer). If it stops being enough in three to five years, replace it then. Cell pricing and energy density both move, and the "high capacity" you pay a premium for today may be ordinary in five years. If the budget, the compartment and a firm growth plan are all there, sizing up is not wrong either - it is just a different bet.
What to check before you commit to a number
Six inputs decide whether a capacity figure is right. If you have all six, any competent supplier can size against them; if you have none of them, you are buying a catalogue number.
| Input | What you should have measured | Why it changes the answer |
|---|---|---|
| Average power draw | Metered, not rated plate power | Rated × hours typically overstates by around 30% |
| Operating hours per shift | Including idle and staging time | Idle load is routinely left out of the model |
| Load factor | Observed, by duty class | The single largest source of sizing error |
| Charging windows | Length and frequency of real breaks | Decides charge rate, not just capacity |
| Ambient temperature range | Both charging and discharging | Cold-store duty changes usable capacity and charge availability |
| Compartment size and weight limit | From the truck OEM specification | A bigger pack can cost you fork capacity |
Polinovel builds packs from 24V to 144V and from 80Ah to 800Ah, so the binding constraint is usually the battery compartment and the charging window rather than the cell count. Send the six inputs above and we will size against them. See the full custom pack voltage and capacity range, or start from the standard electric forklift battery range. For calculated requirements above roughly 500Ah at 48V, start with the heavy duty forklift battery range instead.
Budget structure, not a price list
Cell prices move quarterly and land differently in every market, so any figure published here would be wrong by the time you read it - and wrong in your currency besides. What does not move much is the structure of the budget, and which lines get forgotten.
| Line item | Usually the largest share | Commonly under-budgeted |
|---|---|---|
| Battery packs | Yes | - |
| Chargers | No | Quantity. Opportunity charging needs more units, not bigger ones |
| Charging station install, cabling, panel work | No | Yes |
| Electrical capacity upgrade | Varies enormously | Yes - can be zero, or the largest single line |
| Spare or swap packs | Only in swap configurations | Swap keeps the handling labour that lithium was supposed to remove |
| Commissioning and protocol integration | No | Yes, whenever battery and charger come from different vendors |
- A swap configuration is rarely the cheapest answer unless the battery compartment is fixed and cannot take a larger pack
- Electrical capacity expansion is the line that most often breaks a budget, and it is entirely site-specific
- Ask for a line-item quotation, not a pack price - the pack is the part that is easy to compare
Charger Selection: Where Most Problems Occur
Battery selected, just grab any charger? Charger mismatch is one of the most common causes of commissioning delays and early field failures - and almost all of it is avoidable at the purchase order stage.
Voltage matching isn't that simple
Batteries all labelled "48V" can have very different charge termination voltages, because the nominal voltage does not tell you the cell count:
| Battery Type | Cells | Cell Termination | Pack Termination |
|---|---|---|---|
| NCM Ternary | 13S | 4.2V | 54.6V |
| LFP (Iron Phosphate) | 15S | 3.65V | 54.75V |
| LFP (Iron Phosphate) | 16S | 3.65V | 58.4V |
15S and 16S LFP chargers are NOT interchangeable. A 15S charger on a 16S pack terminates at roughly 3.4V per cell, and the pack never reaches a full state of charge. The reverse is more dangerous: a 16S charger on a 15S pack drives cells past their termination voltage, and you are then relying on the BMS to catch an overcharge condition that should never have been created.
Always verify cell count during procurement. Nominal voltage alone isn't enough.
Communication protocols are honestly a mess
In theory, chargers with CAN communication interact with the BMS in real time and adjust charging parameters to battery status. In practice:
Different manufacturers use different application layer protocols. CAN 2.0 only specifies the physical layer. What happens above that is vendor-specific.
The failure modes that follow from that are predictable:
- Battery from one brand, charger from another: the CAN cable is connected, the handshake never completes, and the installation ends up running as a "dumb charger"
- A supplier claims compatibility with a published standard such as GB/T 27930, but only the basic commands are implemented and the extended set is not
- The battery manufacturer declines to share protocol documentation, citing trade secrets, and integration stalls
What to ask for, in writing, before the charger is ordered:
- Cell count and chemistry
- Pack charge termination voltage and maximum charge current
- The application-layer protocol, and the specific commands actually implemented - "CAN compatible" describes the physical layer and tells you nothing about whether the handshake will complete
- Whether a commissioning report is provided for mixed-vendor installations
The simplest way to avoid the problem is to take battery and charger from one source, or to get a written compatibility guarantee with a commissioning report. Separate purchasing can save real money if you have electrical engineers who can handle protocol integration in-house - it is a false economy if you do not. Polinovel packs can be specified with CANbus, RS485, Bluetooth or GPRS communication; the available communication options on custom packs are set at the design stage, so raise the protocol requirement with the drawing, not after delivery.
How to choose charge rate
Charge rate follows the shift pattern, not the other way round:
| Scenario | Recommended Rate | Notes |
|---|---|---|
| Single shift, 8+ hours overnight charging window | 0.3C~0.5C | Slow charging is gentlest on batteries |
| Double shift, charging at lunch and overnight | 0.5C~0.8C | Balance between speed and longevity |
| Triple shift continuous, only brief gaps | 1C | Opportunity charging scenarios |
| Emergency | 1.5C | Occasional use only, not standard practice |
Sustained charging above 1C accelerates degradation, but by how much is not settled across the industry. Published laboratory data varies widely with cell format, cooling and state of charge window, and field data sets are still small. Treat any single number you are quoted - from any supplier - as a figure that needs its test conditions attached before it means anything.
ROI: Don't Get Fooled by Ideal Numbers
Lithium ROI analyses tend to look beautiful: a payback in the mid-twenties of months, a large five-year saving. They are built on best-case assumptions - full utilisation, stable energy prices, no downtime, no operator error. Projects rarely hit all four.
Where the projection and the outcome part company
Utilisation outliers.
A subset of the fleet almost always runs harder than the model assumed - often because of a driver incentive scheme that never came up during scoping. Those packs age ahead of the fleet average, and they, not the average, set the replacement schedule.
Procedural failures in the first months.
Cold packs plugged in before they have warmed through, night-shift staff who were never trained on the protocol, chargers swapped between trucks with different cell counts. These are training problems, but they show up on the balance sheet as BMS lockouts and downtime.
Spare parts lead time.
A charger mainboard failure can idle a truck for weeks if the part ships internationally. Ask what the supplier holds in stock in your region before you sign, not after the first failure.
Energy price movement.
Payback models almost always assume a flat electricity price across the period. Over a two-year payback that assumption has been wrong in most markets recently, and it moves the result in only one direction.
The rule that survives contact with reality:
Supplier ROI calculations are best-case scenarios. Build your own budget at 70% of the projected saving. If the project still clears your hurdle rate at 70%, it is probably solid. And ask any supplier - including us - for the assumptions behind the payback figure rather than the figure alone: utilisation rate, energy price, depth of discharge, and assumed pack life at your actual duty cycle.
Cold Charging: Special Considerations for Cold Storage
That standard LFP packs should not be charged below 0°C is well established. The practical question in a cold chain operation is a different one: how long after a battery comes out of the freezer before it can charge?
There is no universal answer, because warm-up speed depends on:
- Battery mass (100kg vs 300kg makes a huge difference)
- Case material and thermal path (aluminium conducts heat faster than plastic)
- Ambient temperature and airflow
- Whether an active heating system is fitted
What matters more than the exact minutes is the shape of the curve. The surface of the pack reaches room temperature long before the cells do, and the BMS reads the cells:
| Time | Core Temp | Surface Temp | Chargeable? |
|---|---|---|---|
| 0min | -18°C | -18°C | ✗ |
| 60min | -12°C | -4°C | ✗ |
| 120min | -6°C | +8°C | ✗ |
| 180min | +1°C | +16°C | ✗ (approaching threshold) |
| 210min | +5°C | +19°C | ✓ (slow charge OK) |
Indicative figures for a pack of roughly 200kg in a metal enclosure, with core temperature taken from an internal sensor and surface temperature by IR thermometer. Warm-up time scales with pack mass and enclosure, so ask for the profile of the pack you are actually specifying rather than planning shifts around this table.
Notice the gap between core and surface temperature. Operators touch the case, decide it is "not cold anymore", and cannot understand why the charger refuses. The BMS is reading the cells, which are still below the threshold. That refusal is correct protection. Do not bypass it, and do not let a night shift work around it with a different charger.
Is a heating system worth installing?
An integrated heater removes the wait rather than shortening it - warm-up drops from a matter of hours to a matter of tens of minutes. Whether that is worth the option cost is a frequency question, not a technology question:
The rule of thumb:
If the equipment enters and leaves cold storage more than twice a day, specify the heater. If it goes in occasionally, do not - a room-temperature staging area costs nothing. Park the truck there when the BMS raises its low-temperature flag, and plug in once the core reading, not the case, has cleared the threshold.
To size the decision, work out what an idle truck actually costs you per hour, multiply by the hours currently lost to warm-up, and multiply by the days per year your operation is actually affected. In a year-round cold chain that arithmetic usually settles the question quickly; in a facility that sees freezer work for a few weeks a year it usually does not.
Polinovel offers self-heating and IP-rated pack options for cold weather duty, with IP54, IP65 and IP67 enclosure ratings for condensation and washdown exposure. Cold storage work most often sits on narrow-aisle equipment, so the reach truck battery range is usually the right starting point. Send the store temperature, the number of daily entries and the truck model, and we will come back with the warm-up behaviour for that specific pack.
Questions the Industry Hasn't Settled
It is worth being direct about the things that are still genuinely open, because suppliers who present them as settled are telling you something about themselves:
1. What's the real cycle life of LFP batteries?
Manufacturer specifications commonly show 3000~6000 cycles, sometimes more. Those are laboratory conditions: constant 25°C, 0.5C charge and discharge, a defined depth of discharge. Real industrial environments have temperature swings, variable charge rates and deeper discharges. What proportion of lab life translates to field life is not something anyone can answer with confidence yet, because the industry does not have enough multi-year field data at scale. Treat cycle life as a comparison figure between products tested the same way, not as a promise about your site.
2. Does opportunity charging actually shorten lifespan?
In principle lithium cells count cycles proportionally, so partial charging should not cause extra degradation. Some research suggests frequent shallow cycles accelerate SEI layer growth. The question is still open in the literature, and anyone claiming a definitive answer in either direction is ahead of the evidence. What is not in dispute: charging a cold pack, or holding one at a high state of charge at high temperature, does measurable damage.
3. How will the used lithium battery market develop?
There is no mature secondary market for industrial lithium packs today. Retired units go to second-life applications or recycling. As the first large wave of industrial lithium retires, that may change - and if used packs become genuinely tradeable, the strategy of paying a premium for maximum life needs revisiting. It is a reason to keep residual value out of your business case rather than to build it in.
These are flagged so you can factor the uncertainty into long-term planning, not to avoid a position.
Working With Polinovel
Polinovel has built lithium packs for motive power applications since 2006, from a 50,000 m² facility in Shenzhen with an in-house engineering team, supplying OEMs and distributors internationally.
| Item | Detail |
|---|---|
| Voltage and capacity range | 24V to 144V, 80Ah to 800Ah |
| Chemistry | LiFePO4 with integrated smart BMS |
| Communication options | CANbus, RS485, Bluetooth, GPRS |
| Environmental options | Self-heating for cold weather; IP54, IP65, IP67 enclosures |
| Warranty | 5 years on the material handling range - ask for the conditions that apply to your model |
| Documentation | CE, IEC, UL, UN38.3 and MSDS documentation; certificate numbers and the models each covers are available on request |
| Quotation response | Within 12 hours of receiving your requirement |
| What we need to quote | Voltage, capacity or duty-cycle inputs, compartment dimensions and weight limit, charging windows, ambient temperature range, connector and communication requirements |
More on the company, the facility and the engineering team: about Polinovel.
Final Thoughts
If you are doing preliminary research for a lithium battery purchase:
- Understand your actual needs first: daily consumption, charging windows, operating environment. Measure or calculate these yourself; don't take a sizing figure on trust
- Get quotes from 2~3 suppliers: compare configuration proposals, warranty terms and after-sales response, not just the pack price
- Ask for same-industry reference sites: better still if you can visit one and talk to the people who run the fleet
- Discount ROI projections by 30%: supplier numbers are best-case. Give yourself the margin
- Get the cell count, termination voltage and protocol in writing before any charger is ordered
If you have the six sizing inputs, send them over and we will size against them and come back within 12 hours - or tell you if the requirement is better served by a configuration we do not make. You can start from the full material handling battery range if you would rather look at standard configurations first.
Guidance in this article is general engineering and procurement practice and does not replace the datasheet, BMS specification or charging instructions supplied with a specific battery.
