Every year I review a few hundred motor configurations before they ship—datasheets, first-article inspections, and QC reports. In 2024, I rejected about 6% of first-article inspections for tolerance drift or incomplete documentation. This FAQ answers the questions engineers and buyers actually send to our team. Some answers are straightforward. A couple might surprise you.

What makes maxon-motor different from a cheaper DC motor?

It's tempting to think a DC motor is a DC motor: coil, magnet, shaft, bearings, voltage in, rotation out. That's true until it isn't. With a commodity motor, the next unit off the line might have slightly different resistance, slightly different cogging, slightly different everything. For a one-off prototype, that's fine. For a machine you're building in quantities of 5,000, unit-to-unit variation becomes a real design risk.

What you're paying for with a maxon-motor isn't just the copper and magnets inside. It's the consistency between units. What I mean is that the price difference is largely the cost of making sure every motor that ships behaves like the one before it—which is the part you can't see on a datasheet until your second production batch doesn't work like the first.

In our Q1 2024 quality audit, we ran 400 identical brushless motors through a 24-hour performance test. The spread in electrical time constant between the fastest and slowest unit was under 2%. That repeatability is the real value behind the Swiss-made label: a measurement protocol, a tolerance budget, and a rejection process that refuses to ship a batch that doesn't match the published specs.

"A spec sheet is a promise. The quality system is what keeps it."

(Note to self: I should put that quote on the wall, not just in this article.)

Why choose a maxon brushless motor instead of a brushed DC motor?

Brushless motors are more expensive than brushed versions, and I'm not going to pretend otherwise. The reasons to go brushless: no mechanical commutation, no brushes to wear out, better efficiency, and more torque per gram. The reasons to hesitate: you need a controller, and the total upfront cost is higher.

This is where the "cheaper option" trap usually snaps shut. A customer once saved about $14 per motor by switching to a low-cost brushed unit on a 2,000-unit order. On paper, that's $28,000 in savings. Six months later, the first failures showed up—brushes worn out under continuous duty. Field service visits, replacement motors, and customer downtime cost them roughly $46,000. The budget choice looked smart until the callbacks started.

So when does a high-quality brushless motor actually pay off? Continuous duty. High cycle counts. Locations where you can't easily replace the motor (think robotic joints, medical pumps, actuators in sealed housings). If the motor will run thousands of hours and a mid-life failure would be expensive, brushless usually wins on total cost of ownership—sometimes within the first year.

What is the maxon AIR S motor?

The AIR S is part of maxon's lightweight drive platform—a motor family built around a simple question: how do you get precision motion without carrying excess mass? That question matters a lot in aerial vehicles, modern robotics, and mobile systems where every gram affects payload, battery life, or endurance.

Lightweight motors are harder to build consistently than bulky ones. Thin housings can deform when tolerances drift. Heat dissipation is tighter because there's less material to absorb it. So the AIR S design doesn't just chase a low number on the scale—it balances weight against stiffness, thermal behavior, and reliability. We vibration-test and thermal-cycle these units specifically because "light" has failed in the field before when it was engineered carelessly.

If you're evaluating the AIR S, ask for the torque at the speed you actually need, the duty cycle it can sustain, and the stiffness of the mounting interface. Those three numbers tell you more than the brochure hero shot.

Why add an encoder to a stepper motor?

Because stepper motors can lose steps—and a silent lost step is worse than a loud failure. Open-loop stepper systems assume every pulse results in one step. That assumption breaks down when you accelerate too hard, hit a load spike, push past the torque curve, or excite a resonance. At that moment, the rotor isn't where the controller thinks it is.

An encoder closes that loop. It measures actual rotor position and lets the drive correct errors in real time. In practice, a stepper motor with an encoder gives you much of the reliability of a servo at a fraction of the cost. That's why you find them in 3D printers, CNC machines, pick-and-place arms, and medical dispensing equipment—anywhere a missed step means a scrapped part or a mispositioned tool.

I know this from an uncomfortable experience. We delivered a pick-and-place axis in open-loop mode because "a properly sized stepper doesn't lose steps." What are the odds it would actually miss one? Well, the odds caught up with us during a customer demo when a stalled motor let the gripper drop a part. The part cost $40. The downtime cost three hours. The customer's engineer asked—very politely—why we hadn't included an encoder. That question was the expensive part.

NEMA 23 stepper motor — does the frame size tell you the torque?

NEMA 23 describes the faceplate: 2.3 inches square, with fixed mounting hole positions. That's all it guarantees. It says nothing about torque, current, inductance, stack length, or whether the motor is a good match for your load.

The "just use a NEMA 23 and it'll be fine" advice ignores a messy reality: two motors with the same NEMA frame can have a three-to-one difference in holding torque. A short-stack NEMA 23 might hold around 0.5 N·m; a long-stack version can push past 1.5 N·m. Different winding configurations shift the torque-speed curve as well. Same bolt pattern, very different personality—kinda like assuming two runners with the same shoe size have the same marathon time.

This matters because I've seen engineers design a mount around the frame size and then discover the selected motor can't deliver the required torque at the required speed. Fixing that always costs more than the original selection would have. Frame size gets your foot in the door; the torque-speed curve, the inertia match, and the supply voltage decide whether the machine actually works.

How fast can a stepper motor turn?

No-load speed is a party trick. The real answer lives in the torque-speed curve.

A typical small stepper might reach 1,000–3,000 RPM with zero load. But torque falls off quickly as speed rises, mainly because motor inductance and back-EMF limit how fast current can build in the windings. With a real load, most steppers deliver useful torque only in the 300–800 RPM range. Push past that and the motor can stall entirely—which brings you right back to the encoder conversation.

The practical levers:

  • Higher supply voltage (36–48V) flattens the curve and extends usable speed
  • Lower-inductance windings trade low-speed torque for high-speed range
  • Microstepping improves smoothness but doesn't create torque that isn't there

I've been burned by trusting a rated speed before. Now I demand the measured torque-speed curve from every vendor—including my own team. So how fast can a stepper motor turn? Fast enough if you respect the curve, and not nearly fast enough if you ignore it. And if your application needs high speed and high torque at the same time, it's worth asking whether a brushless motor is the better starting point. That's a conversation we have a lot—and the honest answer is sometimes yes.