From HEV to EV: How Liquid Cooling Became the Foundation

A decade and change ago, packs were small and mild. Today they carry the car. That shift turned thermal into a first-principle problem. If you run procurement, design modules, or build packs on a real line, your checklist is simple: hold ΔT, run ΔP, hit SOP. Everything else—materials, flow paths, sealing, tooling, PPAP—exists to protect those three.

We start with how real vehicles evolved from HEV to PHEV to EV, then break down the hardware paths—cold plates vs cooling tubes—and close with how a full-service supplier like XD THERMAL bridges the gap between concept and scalable production.

The Evolution (HEV → PHEV → EV)

HEV — Batteries as Assist (late 1990s → 2010s)

Early hybrid systems were built around downsized gasoline engines with small battery buffers. The pack mostly handled engine start, torque smoothing, and light electric creep. Heat generation was modest; air cooling and passive strategies typically sufficed.

● 1997 Toyota Prius (XW10/XW11). Nickel-metal hydride (Ni-MH) pack around ~1.7–1.8 kWh (273.6 V × 6.5 Ah in NHW11-era documentation). Cooling relied on ducted air; no liquid plate required for normal operation.

● 1999 Honda Insight (ZE1). Ni-MH pack ~0.94 kWh (144 V × 6.5 Ah across 120 “D-cell” elements). Thermal studies and teardown notes describe fan-assisted air as the primary strategy.

In HEV, the battery was the sidekick. Liquid cold plates were not an engineering necessity.

PHEV — Batteries as Co-Lead (early 2010s )

Plug-in hybrids elevated both energy and power. The pack needed to deliver tens of kilometers of all-electric range, while still supporting hybrid peaks. Heat flux and cycle depth rose, and manufacturers increasingly adopted active liquid loops to keep cells in their comfort zone and to protect calendar/cycle life.

● 2012 Prius Plug-in (PHV, first gen). ~4.4 kWh lithium-ion pack—an order of magnitude more energy than early HEV buffers—pushing systems toward active thermal solutions under sustained load.

● 2013→2022 Mitsubishi Outlander PHEV. Pack capacity moved from ~12 kWh to ~13.8 kWh and then ~20 kWh in the latest generation. The rising energy and higher continuous power pressed OEMs to tighten temperature uniformity and durability targets, where liquid solutions help a lot.

As soon as range and continuous power matter, liquid starts to look like the baseline rather than an upgrade.

EV — Batteries as the Powertrain (2012 → now)

Full battery-electric platforms are the stress test for thermal: long range, DC fast charge, high continuous power, tight packaging, and safety margins that must survive millions of cell-hours. They rely on an integrated EV battery liquid cooling system to keep temperatures balanced during both charge and discharge. Without liquid cold plates or well-designed cooling tubes, uniformity and aging go sideways.

● 2012 Tesla Model S. Large (70–90+ kWh) packs built from 18650 cylindrical cells with liquid coolant loops running through modules—essential for both range and repeatable performance.

● 2017 Tesla Model 3. Moved to 2170 cells; pack capacities commonly ~50–82 kWh depending on variant. The platform balanced cost and fast-charge speed with careful coolant routing and pressure-drop budgeting.

● 2022 Tesla Model Y (Texas-built, 4680 structural pack). Larger-diameter cylindrical cells and structural pack with foam potting changed both mechanical load paths and cooling geometry. Industry teardowns frequently highlight same-side inlet/outlet with U-turn (“U-flow”) and side-cooling ribbons/plates between cell columns. This layout became a benchmark for Tesla 4680 cell cooling design and side-cooling integration.

● 2021 Hyundai IONIQ 5 (E-GMP). 800 V architecture enabling very high DC fast-charge rates, which drives stringent ΔT and ΔP targets across plates and manifolds.

● 2020 BYD Han EV (“Blade” LFP). Long prismatic “blade” cells improve volumetric utilization and safety margins. Pack-level liquid cooling still matters to hold uniformity during charge and sustained loads.

For EV, liquid thermal is foundational. The only questions are how you move coolant, where you touch cells, and how you keep it manufacturable and leak-tight for years.

With each vehicle model iteration, battery cell form factors haven’t evolved in a single direction, but rather along two parallel lines:

One is the continued growth of the square/prismatic/LFP type (e.g., BYD, VW, etc.);

The other is the migration toward larger cylindrical (46 series) cells (e.g., BMW, which shifted from its existing design to a 46mm cylindrical cell; Tesla, which has always used cylindrical cells and has upgraded to the 4680).

Because interface shapes drive heat paths, the choice of cell format becomes a choice of hardware. Prismatic surfaces pair well with plate cooling for uniformity and stiffness; large cylindrical arrays package better with cooling tubes/ribbons along the sidewalls. Such prismatic cell thermal management remains the foundation for most mass-market EV and ESS designs. Many OEMs combine them—plates + tubes—to hit tighter ΔT without blowing the ΔP budget.

Liquid Cooling Plates: What They Offer

A single XD THERMAL liquid-cooling plate family covers a wide operating envelope and delivers a few clear promises:

four types of cooling plate

Cold Plates: Four Proven Routes

Each liquid cooling plate for electric vehicle applications must balance flow efficiency, manufacturability, and lifetime reliability. Here all four routes are proven in automotive and energy storage. Choose by envelope, stiffness, tooling, and schedule.

1) Hydroformed / Stamped (plate‑to‑plate joining)

2) Extruded Cooling Plates

3) FSW (Friction‑Stir‑Welded) Plates

4) Machined Plates (CNC‑milled channels)

What Matters in Practice

Sealing approach: joint geometry, surface prep, and inspection define leak‑tightness; helium testing enables full end‑of‑line checks.
Coolant compatibility: choose alloys, surface treatments, and elastomers for the specified coolant (e.g., water‑glycol) and temperature band.
Interface thermal resistance: control TIM type, thickness, and compression; plate flatness and roughness at the interface dominate real‑world heat paths.
Pressure‑drop design: channel height/width, bend count, and branch splits are tuned to hit ΔP while equalizing flow.

Cooling Tubes for Cylindrical Modules

three types of cooling tubes

Side‑contact often makes more sense with cylindrical cells, especially larger diameters. especially larger diameters, where a cylindrical cell side cooling tube provides an efficient contact path. The lineup includes three patterns: the options and benefits below mirror the product panels.

C2P Ribbon‑Style Tubes

C2M Bended Tube Sets

Tri‑Way (Multi‑Ribbon Parallel) Tubes

Shared Tube Guidance

Interface drives results: pressed contact + TIM often dominates the thermal path more than the tube wall.
Bend economics: each bend costs ΔP; balance branch lengths or add controlled restrictions so branches share flow.
Leak exposure: fewer connections usually mean fewer potential leaks—one reason multi‑ribbon concepts score well in design‑for‑assembly reviews.

4680 and the Story Behind “Side Cooling”

cylindrical battery cell-- liquid cooling tube

When Tesla scaled up the 4680 cell, the diameter grew and the tabless design improved how current and heat move inside each cell. The structural pack and foam potting made the whole assembly much stiffer, but they also tied the cooling parts, structural frame, potting compound, and build order together.

In most teardowns and engineering chats, three words keep cropping up — side-cooling, single-ended, and U-shaped loop. None of this is mystical. A single-ended layout makes the pack easier to assemble and connect, a U-loop lets coolant flow out and back to smooth out temperature differences, and side-cooling takes advantage of the larger surface area and shorter heat path of the cylindrical wall, forming what’s often referred to as a Tesla 4680 cooling plate assembly in teardown discussions.

That’s why so many RFQs now mention things like “side cooling plate” or “cooling plate assembly.” What customers really want isn’t just “a pipe for water to run through,” but a buildable, manufacturable design that holds ΔT and ΔP within spec inside a very tight space. In the Tesla side cooling plate context, this concept ensures stable ΔT under high charging loads.

FAQs

Often it’s a mix of insufficient end-of-loop flow and high interface resistance. Start by redistributing flow through the circuit topology, then align your TIM thickness, compression, and surface finish within a controllable window.

Break the “ideal ΔT” into a realistic, reachable ΔT. Parallel branches, equivalent cross-sections, larger bend radii, and flow splitters can all help squeeze margin. If that’s not enough, balance the pump power and circuit design together rather than in isolation.

No. It depends. For cylindrical-cell side cooling or tight packaging, tubes are often the more agile answer. Just keep a close eye on interface resistance and assembly tolerances—they decide how well the idea works in practice.

There’s no single rule. Real-world constraints—structure, potting, assembly order, and service strategy—often push the design toward side- or bottom-cooling. Engineering chooses whichever path is more stable and manufacturable for that platform.

For high-energy or fast-charging packs, yes. Liquid systems carry heat away more evenly and keep cells within a narrow temperature band—something air cooling can’t manage once power density climbs.

Because electric motors convert huge electrical power into torque, and the copper windings and rotor get hot fast. Liquid cooling channels near the stator pull heat out efficiently and keep motor efficiency and magnet life stable.

It showed up early—around the first modern plug-ins in the late 2000s. The 2010 Nissan Leaf used air-cooling, but by 2012 Tesla’s Model S had gone fully liquid-cooled, setting the template most high-energy EVs still follow.

  • Air cooling: used in early HEVs and small packs.

  • Liquid cooling: the norm for today’s EVs and PHEVs.

  • Refrigerant direct cooling (RDC): newer systems circulate refrigerant directly around modules for faster thermal response.
    Most modern vehicles blend these approaches—liquid for the battery and motor, refrigerant for cabin and pack pre-conditioning.

The path from HEV to EV hasn’t just been about bigger batteries—it’s been about learning to manage heat as carefully as power.Whether a pack relies on prismatic cells with plate cooling, or cylindrical cells with side-contact tubes, the real craft lies in balancing ΔT, ΔP, and manufacturability.

XD Thermal as battery cooling plate manufacturer, the role in this story is simple but vital: turn thermal theory into buildable, testable, and repeatable cooling systems that survive production and real roads alike. As vehicles move faster and charge harder, good cooling won’t just protect the battery—it will define how far and how long it can run. That is precisely the focus of XD Thermal battery cooling solutions.

Get A Free Consultation
with Engineer

Ask For More Questions