What Chips Are Inside an Electric Car? EV Chips Explained

Where you see it: a car that pulls away quietly

An EV has no engine noise, so it pulls away quietly. When you lift your foot off the accelerator, it slows down smoothly, even without the brake pedal. You can also charge it at home. These are the ways it differs from a combustion-engine car.

Semiconductor chips handle all of this. A combustion-engine car also holds many microcontrollers for control ("What Is a Microcontroller?"). An EV (here, a car that runs on its battery alone) makes its driving force from electricity. So it adds chips that handle large amounts of power.

A "map" of the inside of an EV

The figure below is a simple sketch of an EV's main electrical parts and how they connect. How the parts are split up and where they sit differ by maker and model.

The table lists the main chips in each part. We explain the words in the table one by one below. How an IC (integrated circuit chip) works is covered in "What Is an Integrated Circuit?"

Battery pack: many cells, and chips that watch them

An EV battery pack holds many small lithium-ion battery units in rows. Each unit is called a cell. One cell gives about 3.7 volts ("How Lithium-Ion Batteries Work").

To reach hundreds of volts, many cells are joined in a single line. This is called connecting them in series, and their voltages add up. For example, 96 cells in series give about 355 volts (our calculation).

Each cell has a safe voltage range. Charging it past the upper limit, or draining it too far, can damage it or make it unsafe. But the charge left in cells in series slowly drifts apart. As soon as even one cell reaches its upper limit, charging must stop for the whole pack.

So the pack uses chips that closely measure each cell's voltage and temperature. These are battery monitor ICs. For example, Texas Instruments' automotive part BQ79616-Q1 can measure up to 16 cells with one chip. For 96 cells, that works out to six chips (our calculation).

These chips also even out the drift in charge. They let a little energy leak away from cells that hold more charge, to match the others. This is called cell balancing.

The readings pass from chip to chip, like beads on a string (a daisy chain). But the cells at the two ends differ by hundreds of volts. So the chips pass only the signal, without a direct electrical connection. This is called isolation. In this part, a capacitor or a transformer sits in between.

A microcontroller uses the collected readings to work out the remaining charge and how much power the battery may deliver. This whole management system is called the BMS (battery management system).

Inverter and motor: making the driving force

The inverter turns the battery's DC into the AC that spins the motor. We explained how it works in "What Is an Inverter?", so here we look at the chips inside.

At its heart are switches that turn large currents on and off fast. EVs mainly use silicon IGBTs or MOSFETs made of SiC (silicon carbide). Both are kinds of power semiconductor. "IGBT vs Power MOSFET" sums up how they differ.

A motor fed by three wires basically needs six switches. One chip usually cannot carry such a large current alone, so several chips with the same job are used side by side. A 2018 teardown found 24 SiC parts in the inverter of Tesla's Model 3 ("What Is Silicon Carbide?").

A microcontroller's signal cannot drive a power semiconductor directly. The electrode that turns the switch on and off (the gate) has to be driven quickly with a fairly large current. The chip that does this is called a gate driver. In EVs, it isolates the microcontroller side from the switch's high-voltage side (TI's explainer).

A microcontroller works out when to turn on each switch. It bases this on the current flowing into the motor and on the rotation angle, and sensors measure each one (Allegro's explainer). Current sensors often use a Hall-effect sensor, which measures the magnetic field that the current creates.

Regenerative braking: making power while slowing down

When you lift off the accelerator in an EV, the car slows down and power flows back into the battery. This is called regenerative braking.

A motor turns when you send electricity through it. The reverse also works: if something outside turns its shaft, it makes electricity. That is what a generator does. A moving car has momentum. So even after you lift off, the wheels keep turning the motor's shaft.

At this point, the inverter changes how it switches, and makes the motor act as a generator. Generating power makes the shaft harder to turn. This resistance is the force that slows the car. The AC that comes out is turned back into DC by the inverter and stored in the battery. The same power semiconductors work in both directions, out and back.

A car cannot always regenerate the same amount. The Nissan LEAF owner's manual says regeneration is reduced automatically when the battery is fully charged. The same applies when the battery is too hot or too cold. Both limits protect the battery, and the ordinary (friction) brakes take over that share.

The charging path: onboard charger and fast charging

Power from a home outlet or a standard charging station reaches the car as AC. A battery can store only DC, so the car turns the AC into DC inside. This unit is called the onboard charger (Bosch's glossary).

Power semiconductors work inside the onboard charger too. They make the AC flow in one direction, then set the voltage and current to suit the battery. The flow is similar to a phone charger's, but the unit handles far more power ("How AC Adapters and Chargers Work"). According to ROHM's material, SiC diodes and MOSFETs are used here too.

In fast charging, a large unit inside the charging station does the AC-to-DC conversion. DC reaches the car and goes into the battery without passing through the onboard charger.

The 12-volt world and the DC-DC converter

An EV also has many parts that run on 12 volts, just like in a combustion-engine car. These include the lights, the cameras and many control units (Panasonic's explainer).

So the car carries a unit that steps the battery's hundreds of volts down to 12 volts. A circuit that changes a DC voltage into a DC voltage of another size is called a DC-DC converter. Inside it, power semiconductors switch on and off fast to lower the voltage.

Many EVs also carry a small 12-volt battery (such as a lead-acid battery), which the DC-DC converter charges. For safety, the high-voltage side and the 12-volt side are isolated (Bosch's product page).

Chips that help the driver: cameras, radar and SoCs

Some features warn you when you drift out of your lane, or keep a set distance from the car ahead. Together they are called ADAS (advanced driver-assistance systems). They are also common in combustion-engine cars.

The car sees its surroundings with camera image sensors ("How Do Image Sensors Work?") and with radar, which measures distance with radio waves. A large chip then picks out people and cars from what it sees. This chip combines a CPU and circuits for AI, and is called an SoC.

400 volts vs 800 volts: what changes for the chips?

Most EV batteries are in the 400-volt class (Bosch's glossary). The Porsche Taycan, launched in 2019, was the first production car to use 800 volts (Porsche's announcement). In December 2020, Hyundai also announced a dedicated EV platform that supports 800-volt fast charging.

Raising the voltage lets you lower the current. The energy that electricity carries each second is called power. It equals voltage times current. For the same 200 kilowatts, 400 volts needs 500 amperes, while 800 volts needs only half, 250 amperes (our calculation).

When the current halves, the heat lost in the wire falls to about a quarter, if the wire is the same. That is because the heat lost is proportional to the current multiplied by itself (our calculation, simplified). So the wires can be thinner, and the car can take in more power during fast charging (Bosch's glossary).

For the chips, the difference is the voltage they must withstand. The voltage a part can take without breaking down is called its voltage rating. A battery's voltage rises and falls with its state of charge. It also jumps for a moment each time a switch turns off. So in general, engineers choose parts rated well above the battery voltage. As a rough guide, 400-volt-class cars use 650–750-volt-class parts, and 800-volt-class cars use 1,200-volt-class parts (part makers' material).

At the 1,200-volt class, SiC MOSFETs gain a larger advantage over silicon IGBTs. For example, in 2021 Infineon released a 1,200-volt SiC part for 800-volt cars.

Why do EVs have so many chips?

In a combustion-engine car, the driving force comes from burning fuel, and electricity mainly runs the lights and control units. In an EV, the driving force itself is electricity. Enough power to move the car passes through the inverter's switches, which flip thousands to tens of thousands of times a second.

On top of that, power changes form in several places: the inverter, the onboard charger and the DC-DC converter. Each one needs power semiconductors, gate drivers and a microcontroller to control them.

An analysis by the consulting firm P3 (reported in 2022) counted about 1,300 chips in an EV and about 600 in a combustion-engine car. Most of the gap is in the parts that move the car (the drivetrain): about 600 in the EV and about 300 in the combustion car.

For value, there is an estimate from Infineon, a German chipmaker. Per car in 2025, it puts the chips at about $750 for a combustion-engine car and about $1,400 for an EV (the company's investor material). The count and the value come from different firms and different years. Even so, both put an EV at around twice a combustion-engine car.

Heat, cold and shaking: how tough car chips must be

Car chips work for years through winter cold, blazing summer sun, heat from the motor and the shaking of the road. How chips are cooled is covered in "How Are Chips Cooled?"

So car chips are used only after they have passed set tests. A typical test standard is AEC-Q100 from the Automotive Electronics Council (AEC) in the US. In short, it checks whether a chip can stand up to the heat, cold and humidity inside a car for a long time. It covers ICs such as microcontrollers. Single parts such as power semiconductors have a separate standard (AEC-Q101).

The tests heat and cool the chips again and again, or run them at high temperature for a long time, to look for weak spots. The range of surrounding temperatures a chip can work in has "grades." For example, Grade 1 is −40 to 125 °C.

The sheer number of chips also makes quality demands strict. Suppose just one chip in a million is faulty. Build a million EVs with 1,300 chips each, and about 1,300 cars would get a faulty chip (our hypothetical calculation).

So the car industry aims for zero faults. It combines design margins, tight control of manufacturing and tests that weed out doubtful chips. This approach is called zero defects.

Cars also need ways to stay safe even when something fails. One example is a system that moves the car into a safe state when it detects a fault. This idea is called functional safety, and the international standard for cars is ISO 26262.

Common myths

By the numbers

Read the value and count figures as estimates made at the time of each study.

Learn more

• See the differences in a table: Si vs SiC power semiconductors, converter vs inverter, CAN vs LIN vs automotive Ethernet (the wiring that links control units)
• Terms for the next step: ECU, ASIL, zonal architecture, temperature cycling
• Try the numbers: junction temperature calculator (power loss and chip temperature), DC-DC converter calculator
• Related articles: "What Is Silicon Carbide (SiC)?" and "What Caused the Global Chip Shortage?"