Eight years. That's how long I've been handling motion control orders as an application engineer for maxon-motor. In that time, I've personally made—and documented—9 significant selection mistakes, totaling roughly $38,000 in rework, returned product, and a few apologies I'd rather not repeat.

You'd probably assume the worst mistakes were calculation errors or misread datasheets. They weren't. The most expensive mistake I ever made was recommending a maxon-motor when the application clearly needed something else. I let the product I knew override the problem in front of me.

So here's my honest answer to the question engineers ask me all the time: do you actually need a maxon-motor? Sometimes, no. And being able to say that out loud is the most valuable skill I've developed in this job.

People outside engineering think this sounds weak. But in my experience, it's the opposite. The moment you admit a product isn't right for an application, everything else you say becomes more believable.

The expensive lesson no catalog PDF will teach you

One of the most common search terms that lands on our website is "maxon dc motor catalog pdf." I've used that catalog myself—it's an excellent document. But here's the hard truth: a catalog is a reference, not a selection guide.

My first real lesson came in March 2017. A packaging machine builder requested a quote for a maxon DC motor. They'd already picked a model from the catalog based on its stall torque spec. It looked fine—plenty of margin, or so I thought. I processed the order without asking about duty cycle or start frequency.

The motors passed bench testing flawlessly. On the customer's packaging line, with 60 start-stop cycles per minute and a high inertial load, they thermally shut down in under 40 minutes. All twelve motors on the line. That was a $4,800 order that became a full return, plus two engineering visits to fix the application properly.

What the catalog couldn't show me was the duty profile. Under IEC 60034-1, a motor rated for short-time duty (S2) or intermittent periodic duty (S3) behaves completely differently when it's treated as if it were rated for continuous duty (S1). The static numbers all looked right. The dynamic application proved they weren't.

After the re-design, we specified a larger motor with a proper thermal margin and re-matched the gearbox. The line ran 18 months without a motor failure. But the customer never quite trusted us the same way again. That's the part I don't write in the official report: $4,800 for the wrong motors, roughly $1,500 in free engineering time, and a relationship that took two years to rebuild.

Why the right answer was a $5 SG90 servo motor

Everything I'd read about technical sales said the same thing: be the supplier who says yes, who always finds a way to make it work. In practice, I found that the opposite builds more trust. Customers remember the vendor who tells them the truth.

The moment that changed my thinking happened in 2020. A robotics startup asked for "a small servo." My default reflex—or rather, my lazy reflex—was to spec one of our premium brushless servos. Then I asked what they were building: a camera gimbal for a research drone, with a payload under 200 grams.

I went back and forth for an afternoon. The maxon option would have been absurd overkill—a 400-gram motor carrying a 200-gram payload, at fifty times the cost. The customer had already asked, "wouldn't an SG90 servo motor work?" I didn't want to admit it, but they were right. The TowerPro SG90 gives you roughly 1.8 kg-cm of stall torque at 4.8V in a 9-gram package. For that gimbal, it wasn't a compromise. It was the correct engineering answer.

It sounds kinda absurd, I know—a specialist recommending a $5 servo. But I hit send on my reply and immediately felt the doubt. What if they thought we were weak? What if they never came back? The next 72 hours were uncomfortable. Then the reply arrived: "Thanks for being straight with us." They didn't order anything that day. But fourteen months later, when their next product needed a real motion control system, they came back. We qualified early in the design phase. Not because of a discount—because of the truth.

What a failed liftgate taught me about linear actuators

The most expensive mistake in my documented list happened in September 2022. A fleet maintenance manager asked for a maxon liftgate motor replacement. He gave me the part number, I confirmed compatibility, and we shipped a standard replacement unit.

What I didn't do was ask a single question about the application. That truck did 200+ stops per day in city delivery. The liftgate cycled fifty times per shift—not the occasional-use profile that our standard replacement motor is built for. Six weeks later, it failed.

That failure cost $3,700 in parts, labor, and a delivery truck sitting in the shop for two days. It also changed how I think about replacement orders. A "standard replacement" doesn't exist. Every application is a new application, and it needs verification.

I now ask a question that sounds weird in a sales conversation: what happens when a linear actuator fails? Because the answer tells me everything about whether we're the right supplier.

A linear actuator failure is rarely sudden. Typically it's an overheating story: the actuator reaches end of stroke, the controller keeps commanding motion, and the motor sits at locked-rotor current—several times its continuous rating. If the motor wasn't selected for locked-rotor conditions, the winding insulation degrades over weeks until it simply fails. That's not a mechanical failure. That's a specification failure, and in my experience, it's on the engineer who signed the selection.

At least, that's been my experience with industrial applications over the last eight years. Some actuators fail from contamination or mechanical wear. But when I trace a failure back to its root cause, the majority trace back to motors sized for the wrong duty. The mechanical parts are usually fine. The motor paid for the mistake.

The objection: "you just don't make everything"

I know the skepticism. "This is a trust-building trick so he can upsell us later." I get it—I'd think the same. So let me be direct about our boundaries.

We don't build single-phase AC motors. If you need a straightforward induction motor for a pump or a fan, you shouldn't buy a maxon-motor, and I'll tell you that. There are excellent manufacturers for commodity AC motors, and a Swiss precision drive would be a waste of your budget. That's not false modesty. That's the entire basis of our credibility. When I say "this is precision motion, this is what we do best," it means something precisely because we don't claim to do everything.

To be fair, our own datasheets are extremely detailed—almost dangerously so. They make it easy to focus on the numbers in front of you and forget the machine behind them. A datasheet tells you what a motor can do in a test condition. It doesn't tell you what it will do in your machine. That requires questions, patience, and sometimes a vendor who says "no."

What I'd tell a new engineer on day one

Eight years in, I maintain our team's motor selection checklist. The first line is one I would have laughed at in 2017: "Confirm the application before confirming the part number." That checklist has caught 71 potential mis-specifications in the past 18 months.

So here's where I land. Strong professional boundaries don't limit a specialist. They define it. When a customer needs a motor for a surgical robot, a space mechanism, or a precision automation cell—that's when I don't hesitate. That's when a maxon-motor isn't just an OK answer; it's the obvious one. And the customer trusts me when I say it because they've watched me say no first.

I honestly do not know how to make this sound less like a sales pitch. So I won't try. It's a boundary, it's expensive, and it's the most valuable lesson eight years of mistakes ever taught me.