Servomoteur et servocontrôleur

In motion control, “servo drive” and “servo controller” get used interchangeably all the time. That’s understandable — but it’s wrong, and the confusion costs engineers hours of troubleshooting they shouldn’t need to do.

Here’s the clearest way to put it. The servo controller is the brain: it plans the motion, generates commands, and coordinates axes. The servo drive is the muscle: it takes those commands and converts them into the precise current and voltage that makes the motor move, while monitoring feedback to keep everything on track.

A drive without a controller waits. A controller without a drive just thinks. And when both are chosen and sized correctly, you get a machine that runs reliably instead of one that constantly surprises you.

This article breaks down the practical differences, explains how the two collaborate inside a real servo architecture, and gives you a structured checklist for choosing and sizing the right hardware.

What is the difference between a servo drive and a servo controller?

Think of it like a GPS and a car engine.

The GPS (servo controller) decides the route, issues turn-by-turn commands, and recalculates when something changes. The engine and drivetrain (servo drive) convert those commands into actual forward motion — managing the torque, controlling the speed, and delivering the power the journey requires. Both matter. But they’re doing fundamentally different jobs.

In the control hierarchy, the servo controller (also called a motion controller) generates motion profiles, handles multi-axis coordination, and typically manages the position loop — or delegates it to the drive depending on architecture. The servo drive (also called an amplifier) closes the current loop, often the velocity loop, and sometimes the position loop, while supplying the motor with the high-power output it needs.

One important detail: the position loop doesn’t have a fixed home. It may run in the controller. It may run in the drive. The architecture determines it — there’s no universal rule.

In short: the controller decides what to do; the drive makes it happen. (And yes, some vendors sell integrated units that combine both into one box — we’ll get there.)

In even simpler terms…

Still not clicking? This analogy usually does it.

Picture an orchestra. The conductor (servo controller) reads the score, sets the tempo, and cues each section at exactly the right moment. The musicians’ hands and instruments (servo drive) convert those cues into actual sound — with the right force, the right timing, and no missed entries.

Neither can replace the other. A conductor waving at an empty stage produces no music. Musicians playing without a conductor might still perform, but not together, and not in time. You need both — working in sync — to get a performance worth listening to.

The servo world is identical. The most capable motion controller won’t save you if the drive can’t execute cleanly. Get both working in sync and you unlock the precision both were designed to deliver.

What roles do the drive and the controller each perform in the servo system?

You’ve got the concept. Let’s get into the timing.

Think of it like a kitchen during service. The controller is the head chef calling out orders — what to make, in what sequence, at what speed. The drive is the line cook’s hands: executing each step precisely, adjusting in real time when something doesn’t behave as expected.

The controller handles trajectory planning, interpolation, and multi-axis coordination. In high-performance EtherCAT systems, it typically runs at cycle times of 0.25–1 ms — fast enough for accurate path generation and coordinated axes. EtherCAT has grown to 105.2 million installed nodes globally as of 2025, with 16.9 million new nodes added that year.

The drive runs faster still:

• Current loop: 25–50 µs — this is what keeps motor torque accurate
• Velocity loop: 0.1–0.5 ms
• Position loop (drive-based): 1–4 ms when run in the drive

That’s 10–20× faster than the controller’s cycle time. Jitter at any level — especially the current loop — is what produces the torque ripple and positioning errors that take days to diagnose. Low jitter isn’t optional. It’s the foundation.

Safety also lives here. Safe Torque Off (STO) is implemented in the drive, as defined by IEC 61800-5-2. For SIL 2 or PLd compliance under IEC 62061:2021 — the functional safety standard for machinery control systems — drive-level STO is the established approach.

Why is a servo drive also called an amplifier?

Because that’s literally its job. No hidden meaning here.

Think of a home thermostat and a gas furnace. The thermostat sends a small, low-power signal — “heat to 20°C.” The furnace receives it and delivers the actual thermal energy needed to warm the building. The thermostat decides; the furnace acts. It doesn’t produce heat — it determines when heat should be called for.

Same principle. The motion controller produces small-signal command references. The drive amplifies those into the high-current, high-voltage motor phase outputs that create actual torque. That’s amplification — taking a low-power decision and turning it into high-power action.

More specifically: the drive’s PWM power stage chops a DC bus voltage into precisely timed AC waveforms, phase by phase. Simultaneously, it reads encoder or resolver feedback to keep the current aligned with the rotor’s angular position — the process called commutation. I’ve seen systems quietly lose torque efficiency from a single misconfigured commutation offset — the kind of error that doesn’t announce itself, just makes everything slightly worse until someone actually measures it.

The “amplifier” label is accurate and useful. It reminds you that the drive is fundamentally a power conversion device, not a decision-making one.

Is a “servo controller” and “motion controller” the same thing?

You’d think two terms used this often would have one clear, consistent meaning. They almost do.

Think of it like “sofa” and “couch” — the same piece of furniture, two names depending on where you grew up. In everyday engineering use, “servo controller” and “motion controller” refer to the same device: the higher-level unit that generates motion commands, plans trajectories, and either closes the position loop or delegates it to the drive. “Motion controller” is technically more precise — it covers servo and stepper systems alike — but you’ll hear the two swapped freely in conversation and datasheets.

The nuance: some vendors sell integrated units that combine the motion controller and servo drive into one enclosure. When they say “servo controller,” they mean brain-plus-muscle in a single box. We make integrated units too, and they’re a solid choice where simplicity matters. But they’re a different product category from a standalone controller.

One useful data point: digital servo drives — the type with meaningful on-board processing — now hold approximately 55% of the servo drive market (SNS Insider, 2023). That share reflects how much on-board intelligence has improved, which is why the boundary between “drive” and “controller” keeps getting blurrier.

Can a servo drive function as a motor controller?

For the right application? Absolutely. And it often makes economic sense.

Many modern servo drives include built-in motion functions: point-to-point indexing, homing sequences, basic I/O logic. Think of a smart TV with built-in streaming versus a basic screen that needs a separate streaming box — one device, fewer cables, less to manage. (No shame in that.) For single-axis applications like a rotary indexer or a simple positioning stage, the drive may handle everything without a separate controller.

Here’s where it breaks down.

The moment your application requires multi-axis interpolation, electronic camming, robotic kinematics, or synchronized coordinated movement — the drive’s built-in functions run out of runway. These tasks need trajectory generation, coordination logic, and deterministic cycle times that a drive’s internal processor was never designed to handle.

The decision rule isn’t axis count. It’s motion type.

Two independent axes doing point-to-point moves? A capable drive with built-in indexing can manage it. Two axes running a coordinated cam profile? You need a dedicated motion controller. The moment “interpolated” or “synchronized” enters the application description, the decision is essentially made.

How do you choose and size a servo drive and controller for your application?

This is where most selection mistakes happen — not because the process is difficult, but because it’s easy to skip steps that seem obvious in the moment.

Think of a pilot’s pre-flight checklist — worked through every single time regardless of experience, because the cost of a skipped step is too high to leave to memory. Servo selection works exactly the same way.

Work through these eight points before you specify anything:

1. Motion profile — Travel distances, velocities, accelerations, and duty cycles. No numbers, no valid sizing.
2. Load inertia ratio — Reflected load inertia vs. motor rotor inertia. Target 5:1 or lower for stable, well-behaved control. Go significantly higher and tuning becomes increasingly painful.
3. Torque and speed envelopes — Plot peak and continuous requirements against the motor’s published curves. The drive must cover peak current for acceleration et continuous current for steady-state.
4. Exigences de sécurité — STO, SS1, SS2? IEC 61800-5-2 and IEC 62061 define the requirements. SIL 2 / PLd is the standard target for most machinery.
5. Coordination requirements — Independent moves or coordinated/interpolated motion? This one question determines whether you need a standalone motion controller. Answer it early.
6. Fieldbus compatibility — EtherCAT, CANopen, PROFINET, EtherNet/IP? Industrial Ethernet now accounts for 71% of new network node installations (HMS Networks, 2024). Fieldbus mismatch means replacing hardware, not reconfiguring it.
7. Environmental constraints — Enclosure rating, ambient temperature range, vibration, available panel space.
8. Vendor ecosystem — Technical support, tuning software, diagnostic tools. You’ll rely on these across the full machine service life, not just at commissioning.

Sizing example: peak current for acceleration

Suppose you need to accelerate a 0.01 kg·m² reflected load inertia to 3,000 rpm (314 rad/s) in 0.2 seconds.

Step 1 — Angular acceleration: α = 314 ÷ 0.2 = 1,570 rad/s²

Step 2 — Total inertia (load plus motor rotor): Don’t skip the motor. It’s the single most common sizing omission, and it leads directly to undersized drives.

J_total = J_load + J_motor = 0.01 + 0.002 = 0.012 kg·m²

Step 3 — Acceleration torque: T_total = J_total × α = 0.012 × 1,570 ≈ 18.8 N·m

Step 4 — Peak current: At a torque constant of 1.5 N·m/A: I_peak = 18.8 ÷ 1.5 ≈ 12.6 A

That’s your baseline. Add friction, gravity load, and safety margin, then compare against the drive’s peak and continuous current ratings — both columns matter.

The Importance of Consulting with Professionals

Servo systems are deceptively complex — the individual components often look straightforward on paper, right up until they don’t work together as expected.

Choosing the right drive and controller isn’t only about matching torque and speed numbers — you’re simultaneously aligning loop architectures, safety certifications, thermal management, fieldbus timing, and scalability. Getting any of these wrong can mean the difference between 20,000–30,000 operating hours of reliable service and a machine back on the bench well before its first scheduled maintenance.

The numbers make this concrete: roughly 80% of servo motor repairs are preventable with correct initial sizing and timely maintenance (Advanced Motion Controls, 2024). Most failures that reach a repair facility were selection or configuration decisions made at the design stage.

“The most common issues we see in the field trace back to inertia ratio or thermal derating — steps that were skipped or estimated too loosely during the design phase,” says Dan, one of our senior application engineers at AMC. Getting expert review in early is worth more than any amount of post-commissioning troubleshooting.

At ADVANCED Motion Controls, we’ve been supporting engineers across industrial automation, robotics, mobile platforms, and outdoor applications for 32+ years. Our FlexPro®, DigiFlex® Performance™, and AxCent™ servo drive families work with the fieldbuses and motion controllers you’re already running — and our application support team is engaged from pre-sale sizing through commissioning. Reach out to us directly if you’re early in the design process. We’d rather review your application now than troubleshoot it later.

Réflexions finales

Here’s what it comes down to: the controller plans, the drive executes. Think of it as the brain and the muscle — each doing its job, neither capable of doing the other’s.

The servo drive handles the fast, demanding low-level loops that keep the motor accurately tracking under real-world conditions — current at 25–50 µs, velocity below 0.5 ms, torque held steady despite load fluctuations. The motion controller sits above it, managing trajectory generation, axis coordination, and the logic that connects individual move commands into a coherent machine sequence.

Treating these two as interchangeable is how performance problems, failed safety audits, and scaling headaches find you. The global servo motor and drive market is projected to reach USD 22.5 billion by 2031 (Verified Market Research, 2024) — a figure that reflects how central precision motion has become to modern manufacturing. More engineers are making these selection decisions now, and many of them for the first time.

Take the selection seriously. Do the inertia math. Verify the fieldbus. Get an expert in the loop before the design is locked.

For a closer look at how the motor fits into this architecture, read our guide on What Is an AC Servo Motor?. And if you’re evaluating fieldbus options for a real-time multi-axis system, our EtherCAT implementation breakdown is the natural next read.