Let me start with the question I get more than any other on the procurement side: "Why would anyone pay $300 for a motor when an MG90S servo is $3?"

It's a fair question. And the short answer is: sometimes you shouldn't. I manage a six-figure annual budget for precision motion components, and I've documented every order in our cost tracking system for six years. The one thing I keep telling engineers is that the price on a purchase order is not the cost of a motor.

The comparison I want to walk through here isn't "which motor is best." It's which one is cheapest to own in your actual application. I'm comparing four common paths:

  • a maxon motor dc drive system, specified from the maxon motor catalog
  • a hobby-class MG90S servo motor
  • a NEMA 17 stepper with a typical stepper driver
  • an AC induction motor driven by a VFD

I'll judge them the same way I judge vendors: total cost, integration effort, what the documentation really tells you, and what it costs when the thing fails.

First, What Are We Actually Comparing?

I'll keep the definitions short, because this is a cost conversation, not a physics lecture.

maxon-motor is a Swiss manufacturer of precision DC motors, brushless DC motors, servo motors, and gearboxes. Their catalog is the gold standard for anyone who needs a motor with full specifications—torque-speed curves, mechanical drawings, CAD files, the works.

MG90S is a small, cheap analog servo with plastic gears. It costs a few dollars, runs on 5V, and is everywhere in hobby robotics.

Stepper motors move in fixed increments, which makes positioning relatively straightforward—assuming the wiring and driver are set up correctly.

And what's a VFD? A variable frequency drive is an electronic controller that varies the frequency and voltage of the power going to an AC induction motor. That's how you get adjustable speed out of a motor that otherwise just spins at one speed when you plug it into the mains.

OK. Now the money part.

Round 1: Sticker Price vs. Total Cost

Let's put ballpark numbers on the table. Don't hold me to these exactly—they vary with supplier and quantity—but this is the range I've seen in real quotes:

  • MG90S servo motor: $3-8
  • NEMA 17 stepper + driver: $15-50
  • VFD + AC induction motor: $200-600
  • maxon DC motor with encoder, from the catalog: $200-1,200, depending on the combination

The MG90S looks like the obvious winner until you do what I do with every line item: ask what happens when it fails.

Here's the thing: the cheap items on a BOM are rarely the expensive part of a failure. The expensive part is the labor to swap it, the machine downtime, and the re-testing. When I audited our 2023 spending, I found one product line where $4 servos were causing $1,200 rework events when the plastic gears stripped in the field.

That's a 30,000% markup hiding in a line item.

Now, to be fair: not every application stresses an MG90S to that point. If you're building a hobby robot arm on your desk, a $4 servo is the right call. But if that motor is inside a product you sell, the total cost picture changes completely.

The same logic applies to the maxon option. On paper, a $600 maxon motor is 200x the price of the MG90S. But the maxon motor catalog gives you the engineering data to know—before you buy—whether the motor will handle the load. That's not a luxury. That's what turns a motor purchase from a gamble into a decision.

And what about the VFD path? The hardware can look reasonable, but I've never fully understood why VFD commissioning quotes vary so wildly between vendors. My best guess is that "installation" means different things to different people: some include parameter setup and testing, some just hang the box on the wall. That's a hidden cost you need to chase down in writing.

Round 1 conclusion: lowest purchase price wins only if the cost of failure is near zero. In most of our projects, failure is the line item that actually breaks the budget.

Round 2: What the Datasheet Actually Tells You

The maxon motor catalog is, honestly, the reason maxon can charge what they charge. You get torque-speed curves, efficiency data, mechanical drawings, electrical connection diagrams, and CAD files. For an engineer, this is everything. You can simulate the motor in your application before spending the money.

According to the maxon motor catalog, each motor line includes full torque-speed curves, mechanical drawings, and electrical connection diagrams. That documentation is the difference between spec'ing a motor and guessing at one.

Compare that to the MG90S. The typical spec sheet lists "stall torque: 1.8 kg·cm" and "operating voltage: 4.8-6V" and that's about it. No curve. No lifespan figure. No indication that the torque number is measured at the ideal voltage with a new battery. What most people don't realize is that these specs are best-case, and the plastic gearbox wears quickly. But you won't see that in the datasheet.

Stepper motors are a middle ground: the datasheets are decent, but there's a catch that trips people up. A stepper's torque drops as speed increases. The motor that works at 300 RPM might lose half its torque at 1,500 RPM. If you spec based on the low-speed torque figure, you end up with a system that stalls under load in production.

And the VFD route? The datasheet is only half the battle. A VFD needs to be programmed with motor parameters, acceleration curves, and protection settings. In my experience, the hardware is rarely the problem—the configuration is where budget disappears. That leads to a practical question: why does this matter so much? Because documentation is the part of a motor system that saves you money every single time you build a new machine.

Round 2 conclusion: if you need to know how the motor will behave before you integrate it, the maxon catalog wins on documentation alone. If you're experimenting and can afford trial-and-error, then a cheap servo is fine.

Round 3: Wiring, Integration, and the Budget Nobody Tracks

This is where I see engineers' time get burned. Let's talk about stepper motor wiring specifically, because it's a perfect example of a "cheap" motor that isn't free to integrate.

A typical NEMA stepper has 4, 6, or 8 wires. You need to figure out which wires belong to which coil, connect them to a stepper driver in the right order, make sure the current limit is set on the driver, and verify the direction logic. Get any of that wrong, and the motor either hums without moving, runs in the wrong direction, or runs hot and quickly degrades. I've seen an entire afternoon disappear into stepper wiring.

The MG90S is the opposite: three wires, 5V, ground, and a PWM signal. An Arduino library drives it in five minutes. But the low integration cost comes back later in performance—no feedback, no position verification, just a hobby servo guessing where it should be.

maxon motors are more work to wire than a hobby servo: power, encoder, hall sensors if it's brushless, plus a compatible motor controller. That's real effort. But here's what changes the equation: the wiring diagrams in the maxon documentation are precise, and the controllers are designed to match the motors. When I compared quotes for a new system in Q2 2024, the maxon vendor's documentation cut our integration time by roughly two-thirds compared to the previous supplier.

Then again, if your project is a one-off prototype, you might not care about two-thirds. It's a question of how many units you're building and what your labor is worth per hour.

The VFD side has its own integration story. Three-phase wiring, a HMI or remote control setup, motor parameter ID runs, and safety circuits. It's not the hardest thing in the world, but it's almost never a 15-minute job. And on a cost-per-engineer-hour basis, that setup time often exceeds the VFD itself.

Round 3 conclusion: the cheapest motor to buy is often the most expensive to integrate. Wiring and commissioning time don't show up on the purchase order, but they show up on the timesheet.

Round 4: What Happens When It Breaks

I'll be direct: precision matters most at the moment of failure, not at the moment of purchase.

A maxon motor is a Swiss-made precision instrument. Brushed DC motors eventually wear out their brushes, but the lifespan is documented, and the performance degrades predictably. When we need a motor to do the same job 10,000 times without drifting, that predictability is what we're paying for.

A stepper motor, with the right driver and wiring, is surprisingly reliable at low speeds. But it's inherently open-loop: if the load exceeds the torque, the motor loses steps, and the position silently drifts. Unless you add an encoder, you won't know until the product fails inspection.

An MG90S servo has plastic gears. Under continuous load, they strip. Not maybe. They strip. And when they strip, the whole unit needs replacing. On a bench, that's a minor annoyance. In a product in the field, it's a warranty claim, a service visit, or a lost customer.

As for the VFD + induction motor combination: this is actually the most robust option for high-power, continuous-speed applications. I'd pick it every time for conveyors, fans, pumps. What it doesn't do is precision positioning. A VFD controls speed, not position.

Round 4 conclusion: reliability is a cost, not a feature. When I analyze our total cost of ownership spreadsheet, the motors that just work for years are almost always the ones with the best documentation, not the lowest price.

So Which One Should You Buy?

Here's the part where I refuse to give you a single answer, because anyone who tells you one motor is always the right choice is trying to sell you something.

Buy the MG90S if you're prototyping, learning, or building something where a failure costs you nothing. For teaching and R&D, the $4 servo is a fine tool. Just don't put it in a product you're shipping.

Buy a stepper with a proper driver if you need repeatable positioning at low speeds, you're OK with open-loop limitations, and you want a good value. But budget for commissioning time, and learn the stepper motor wiring before you start.

Buy a VFD and AC induction motor if you need variable speed on a continuous process—pumps, fans, conveyors. It's a workhorse solution. Spend the time on commissioning, and keep spare parameters documented.

Buy the maxon motor if you need precision, reliability, and a real datasheet trail; if your product carries a warranty; if you're building multiple units and want the same result every time; or if a mid-project motor failure would cost more than the motor does. In those cases, a maxon motor dc system specified from the maxon motor catalog is honestly the cheapest option on this list.

If you're not sure which bucket you're in, my advice is to calculate what one field failure costs you—parts, labor, rework, customer trust—and compare that to the motor price. That number usually settles the debate faster than any spec sheet battle.

I've been tracking motor costs for years, and I still won't tell you there's a "best" motor. But I will tell you this: the cheapest motor on the quote is rarely the cheapest motor in the machine. That's true whether you're buying from maxon, from a hobby distributor, or from anyone else.