CATL’s automotive-grade NMC811 prismatic cell ships with a headline safety feature: a “Dual Vent Safety Design.”
The specifications:
- NMC811 cathode
- 142.9Ah | 3.65V | 521.6Wh
- 32.6 × 221.3 × 119.8mm, 2kg
- 262Wh/kg
- ~0.36mΩ DC internal resistance
- 510A peak discharge
- 2C fast-charge
We believe this cell is for ACC.
Jerry Wan shared details of this CATL cell on LinkedIn and it got us thinking:
Why vent design matters more on NMC811
A cell’s vent only has one job: get gas out faster than pressure can build up. How much margin that job has depends on the chemistry. NMC811’s high nickel content is what gives it the energy density in the spec sheet above, but nickel-rich cathodes are also less thermally stable than lower-nickel or LFP chemistries. They start decomposing exothermically at lower temperatures, and once thermal runaway starts, the gas volumes involved are large. Published characterisation work on large-format prismatic cells has measured gas generation rates in the region of several litres per amp-hour of capacity during runaway, internal pressures reaching roughly 200kPa, and vented gas temperatures well above 600°C [1]. On our own reference page on cell venting, we note that prismatic NCM cells typically actuate around 0.9MPa, against roughly 0.6–0.7MPa for prismatic LFP. NCM already needs a higher-pressure, higher-flow vent than the chemistry most of the industry benchmarks against [2].
A single vent sized for an older, lower-energy-density cell can become the bottleneck in that scenario: if gas can’t leave fast enough, pressure keeps climbing, and the failure mode gets worse: case rupture, jetting flame, faster heat transfer into neighbouring cells. That’s the design pressure (literally) that a claim like “dual vent” is responding to.
What the physical cells actually show
Photos of the cells settle part of the question before any patent search does. The top cover of each cell carries two separate round vent structures on the same face: not one central vent, and not a cell-to-pack arrangement, but two vents built into a single cell’s cover plate. Each vent sits outboard of its nearest terminal: terminal, then vent, then the edge of the case on one side; the mirror image on the other; the polarity marks and a QR-coded traceability label occupy the space between the two terminals in the middle.
That layout tracks with the cell’s proportions, and it’s the 221.3mm length rather than the width that most plausibly drives it. This is a long, thin can, and a gas-generating event doesn’t necessarily start in the middle: an internal short or localised failure can just as easily originate near one end. If the only vent sat at the centre, gas and heat from an event at one end would have to travel up to roughly 110mm internally before reaching it. Over that distance, there’s a real risk that local pressure builds faster than it can propagate to a vent that far away, and the case wall itself gives way first. Once the case has failed, the vent is irrelevant: gas and particles are escaping through an uncontrolled rupture rather than the engineered path, which defeats the entire purpose of designing a vent in the first place. Putting a vent near each end instead means that wherever along the 221.3mm length an event starts, there’s a relief path close by rather than one a full half-length away.
The dual-vent structure is not simply about increasing pressure-relief capacity. It also helps define and control the gas/flame release path, keeping the high temperature venting zone separated from the BMS/sense-wire routing area.
This is very much aligned with the evolution of automotive battery safety concepts according to CATL’s NP2.0 approach and thermal-electrical separation: when thermal runaway occurs, the objective is not only to prevent cell rupture, but also to control the DIRECTION of heat, gas and flame propagation while protecting electrical monitoring and isolation functions.
In a traditional design, hot gas or flame can potentially damage nearby sensing wires, causing insulation failure, short circuits, loss of cell monitoring, or even secondary ignition.
So I would look at the two vents as part of a broader “thermal runaway pathway management” architecture — separating the flame/gas zone from the wiring and electrical control zone. Jerry Wan
The cell genuinely has two vents. What’s still open is which of CATL’s patents, if any, covers this specific two-vents-per-cell layout.
This post has been built based on the support of and sponsorship from: illumion, Custom Power, Quarto Technical Services, Eatron Technologies and About:Energy
What CATL’s patents actually describe
CATL files a large number of incremental patents on cell venting: dozens are listed under Contemporary Amperex Technology Co., Ltd. and its Hong Kong-registered sister entity. We pulled the four that came up as the closest match to a “dual vent” style claim and checked each one directly on Google Patents rather than relying on secondary summaries.
A vent that resists nuisance-opening at low pressure
US20230093863A1: Pressure Relief Mechanism, Battery Cell, Battery, Electric Apparatus, and Manufacturing Method Thereof (CATL, priority date 2021-09-29)
This one isn’t about having two vents: it’s about making one vent behave better across the cell’s life. The vent is built directly into the housing plate: an outer ring anchors it, an inner panel angles into the cell and meets a deliberately thinned weak zone, which connects to a second panel angling back outward. At normal operating pressure, that geometry is self-compressing: the two panels press the weak zone together rather than pulling it apart, which is the opposite of what happens in a flat stamped vent disc, where the weak point is under tension from the moment it’s formed. CATL’s stated motivation is exactly the failure mode you’d expect from a flat design: a vent that’s rated correctly on paper but creeps and fails early in normal service because the weak point is permanently under some tensile load. Only once internal pressure crosses the design threshold does the geometry invert and put the weak zone into tension, so it tears on cue rather than early.
Nothing in this patent limits a cell to using one of these. It describes a single vent’s internal geometry, not how many of them get fitted to a cover plate or where, so this is a plausible candidate for the actual vent unit built into each of the two locations on the cell photographed above, used twice rather than once.
Two-tier venting: cell-to-pack, not cell-to-cell
US12567647B2: Battery and electric apparatus (CATL, priority date 2023-06-07)
This is the one we expected to describe two vents on a single cell. It doesn’t: the “dual” structure here sits one level up. Each cell has its own first pressure relief mechanism, and the battery box (the pack enclosure, not the cell) has a second one, connected by a dedicated relief channel. Gas from a venting cell is funnelled through that channel to the pack-level vent rather than dumping straight into the pack interior. It’s a real two-stage venting architecture, and a sensible one for limiting propagation to neighbouring cells: it’s just describing pack design, not a second vent on the cell itself.
This appears to fit with the Zeekr 140kWh CATL Qilin pack benchmark where we see the cell vent at the bottom of the cell and within the pack there is a dedicated venting channel.
Keeping gas from pooling unevenly behind the cover
US12431582B2: Battery cell including pressure relief structure and cover assembly having first recessed portion… (CATL, priority date 2021-06-23)
This patent sits just behind the vent rather than describing the vent itself. It adds a recessed pocket, shaped around the electrode tab, on the inward-facing side of the cover plate that carries the pressure relief mechanism, plus at least one internal channel. The point is to stop gas pooling unevenly around the tab geometry before it reaches the vent, which would otherwise mean the vent sees pressure later, or less evenly, than the cell’s actual internal state.
Protecting the vent membrane from the electrolyte it sits next to
US12300846B2: Pressure relief mechanism, battery box, battery cell, battery, preparation method and apparatus (CATL, priority date 2020-07-10)
The last of the four is about longevity rather than gas flow. A separate protective sheet, held by a compression ring, sits between the actual pressure-relief sheet and the cell’s internal chemistry, so the relief sheet itself never sees direct electrolyte contact. The concern being designed against is corrosion of the vent membrane over years of service: a vent that’s structurally fine on day one but has been quietly weakened by electrolyte exposure by year five doesn’t help you.
Does anyone else use two vents on one cell?
Two vents at separate locations on a single cell isn’t something we’ve seen documented elsewhere, and two comparisons are worth making explicitly, because they pull in opposite directions.
BYD’s Blade cell is the obvious first check: an LFP prismatic cell around 960mm long (more than four times the length of the CATL cell here), which on the length argument alone should make an even stronger case for splitting vent locations. It doesn’t appear to work that way. Our own spec listing for the Blade cell records a single pressure relief valve, not two. That’s not a confirmed teardown count, so treat it as suggestive rather than settled, but it points the same direction: length by itself doesn’t automatically force a two-vent design. If a 960mm cell can get away with one vent, chemistry has to be doing real work in the CATL cell’s two-vent decision as well. LFP’s thermal runaway is slower to start and produces less gas than high-nickel NMC, so a single vent on the Blade cell may simply never be asked to move as much gas, as fast, regardless of how far internally it has to travel.
Tesla’s 4680 cylindrical cell is the second comparison, and it points the same way from the opposite end: it also uses an NMC811 cathode, but vents from a single point at the bottom, negative end of the can, based on published teardown analysis. The 4680 is a much smaller cell (46mm diameter, 80mm tall) than the one here, so the “distance from a localised event to the nearest vent” problem the CATL cell’s two vents are solving doesn’t really exist at that scale: there’s nowhere on a jelly roll that short for a local event to outrun a single vent.
Put the two together and the two-vent layout on this cell looks like a response to a fairly specific combination: a cell long enough for a localised event to outrun a single vent, built from a chemistry aggressive enough that the outcome matters if it does. Neither condition on its own (just length, or just NMC811 chemistry) seems to be enough to force the same design choice elsewhere.
What we couldn’t verify
None of the four CATL patents above confirms the specific two-vents-per-cell layout shown in the photo. US20230093863A1 describes a single vent’s internal geometry, which is a plausible match for the vent unit used at each of the two locations, but a mechanism patent like that doesn’t itself claim anything about count or placement, so it isn’t proof either way. The cell-to-pack scheme in US12567647B2, the gas-channelling patent, and the membrane-protection patent are relevant background on CATL’s broader venting work, but none of them describes two vents sharing one cover plate. It’s entirely possible the specific placement claim sits in a CN- or WO-family filing we haven’t checked, or in a design patent covering the cover plate layout rather than a utility patent covering the vent mechanism itself.
We also weren’t able to locate this specific 142.9Ah / 262Wh/kg SKU on CATL’s own site: their public passenger-EV solutions page confirms the same generation of chemistry (nickel-rich NCM, cell-to-pack integration, energy densities in the 265Wh/kg range) but doesn’t list individual cell datasheets publicly [3].
If you’re evaluating cells against this specific claim, the practical move is to ask the supplier directly for the patent number or internal test report behind “dual vent,” rather than taking the spec sheet’s description at face value, and verify actuation pressure, exhaust rate, and vent location yourself.
Further Reading on BatteryDesign.net
→ Cell Vent: actuation pressure ranges by chemistry and format, and the design trade-offs behind vent sizing → Thermal Runaway → CATL Cell Benchmarking → BYD Blade: the 960mm LFP cell referenced above → Tesla 4680 Cell: the NMC811 cylindrical cell referenced above
Footnotes
- Gas generation rate, peak pressure, and vent gas temperature figures are drawn from published thermal runaway characterisation studies on large-format prismatic Li-ion cells; see e.g. thermal runaway propagation studies on NMC-811 and LFP automotive cells in Journal of Power Sources Advances (ScienceDirect).
- See BatteryDesign.net’s own Cell Vent page for actuation pressure figures by chemistry.
- CATL, Passenger Vehicles Solutions, accessed August 2026.