An MG995 Servo Motor Replaced a Brushless DC Motor—and Stopped a Production Line

A quality inspector explains what happened when an MG995 servo motor was wired into an industrial packaging line instead of the specified brushless DC motor from Teco Electric and Machinery—plus a stepper motor primer.

It was 2:38 on a Tuesday afternoon in March when the phone rang. The voice on the other end belonged to a maintenance supervisor at a packaging plant, and he didn't waste time with pleasantries.

“The motor on the flap tucker finally let go. We need a replacement by Friday.”

The machine he was talking about is a case packer we built a couple of years ago. It runs 16 hours a day, cycling roughly 20 times a minute. The motion package was specified from the teco electric and machinery catalog: VFDs for the conveyors, steppers for the indexing dial, and a small brushless DC motor for the flap-tucker axis.

For context, I’m a quality and compliance manager at an automation builder. I review machines before they leave our floor—I’ve signed off on maybe 180 machines since 2019. Actually, closer to 160, I’d need to check the log. My job is to catch what the drawings don’t say. And this call turned into one of those cases where the drawing was the least of our problems.

First, the motor that wasn’t the problem

We’d seen this machine a few weeks earlier for the same fault. The drive had been logging over-current alerts for days. The line team kept clearing the alarm because the machine was needed for production, and they assumed the motor was just getting old.

It wasn’t old. The motor was being pushed sideways.

When we got to the flap-tucker mechanism, we found the real culprit: a cam-follower bracket had worked loose. The bracket shifted just a couple of millimeters under load, creating friction that the motor had to fight on every cycle. That constant mechanical drag made the motor current climb until the winding insulation gave up. Classic overload failure.

We replaced the motor with the correct spare, torqued the bracket to spec, and thought we were done. That was my mistake—assuming the lesson had landed.

Then came the MG995 servo motor surprise

Three weeks later, same machine, same call. But this time, the maintenance tech had already “fixed” it before I arrived.

I opened the electrical cabinet and smelled it immediately—that sharp, burnt-electronics smell you never forget. And there it was, zip-tied to the motion bracket: an MG995 servo motor, the kind you’d buy for an RC car or a hobby robot arm. Wired through a little 5-volt buck converter. The smoke was still fresh.

The tech explained what had happened. The original brushless DC motor had failed again—this time because the bracket had come loose a second time. No spare was on the shelf, and he didn’t want to lose a full shift waiting for the correct motor to arrive, so he went looking for a quick replacement. He found the MG995 servo motor listed as a “servo motor,” saw a torque number that looked reasonable, and decided it was good enough.

In fairness, it did run for a few minutes in test mode with no product in the machine. But the first real box hit the flap tucker, the mechanism stalled under load, and the MG995’s output stage gave up. That was maybe the fifth cycle. The MG995 is a small DC motor with an analog potentiometer and a gear train inside—not a continuous-duty industrial motor. It wasn’t designed for 20 cycles per minute, 16 hours a day.

The frustrating part is that the original motor’s part number was still readable on the nameplate. The tech had even looked it up in the teco electric app while trying to find a substitute. But he searched by the word “servo” rather than by the machine’s bill of materials, and the search results sent him somewhere completely different.

Servo motor, brushless DC motor, stepper motor: the differences matter

Part of the problem is that “servo motor” means two different things in two different worlds.

  • MG995 servo motor (hobby/RC class): a self-contained geared motor with a potentiometer for position feedback. It usually rotates only about 180 degrees, expects a 5–6 V RC-style PWM signal, and is designed for low-cost, low-duty applications.
  • Brushless DC motor (industrial class): a three-phase permanent-magnet motor with electronic commutation. It rotates continuously, needs a matching controller or drive, and is built for longer duty cycles. With the right feedback and drive, it becomes what people usually mean by an industrial servo motor.
  • Stepper motor: a motor that moves in fixed angular increments—typically 1.8 degrees per step—by pulsing current through its coils. It can run open-loop, meaning it doesn’t need an encoder to know where it is, but it can also stall silently if the load exceeds its torque.

When someone asks me, “what stepper motor should I use?” my first answer is usually: it depends on your load, speed, and whether you need closed-loop feedback. It is never “whatever has the same shaft size.” That sort of thinking is exactly how a $7 MG995 ended up in an industrial cabinet.

One distinction I try to explain to customers: a motor is only one part of a system. The drive, feedback device, wiring, and mechanical load all have to match. You can’t swap a brushless DC motor for a hobby servo just because both are called “motors.” The voltage is different, the control signal is different, the duty cycle is different, and the mechanical feedback is different.

What the fix actually required

After we removed the melted MG995 servo motor, we put the correct brushless DC motor back in place. But we also did the thing that should have been done the first time: we checked the entire mechanical path.

The loose bracket was re-torqued. The coupling was aligned. We measured the motor current at full production speed—it dropped from about 2.9 amps under load to 1.1 amps after the mechanical fix. That difference is the whole story: the motor wasn’t failing on its own. It was being asked to push against a machine that was slowly tearing itself apart.

Since then, we’ve added a simple rule to our quality documentation: after any motor replacement, the technician has to record the running current and verify that the mechanical load is within spec before restarting production. It takes ten extra minutes, and it would have prevented both failures.

The honest bottom line

I don’t want to make the MG995 servo motor sound like junk. It isn’t. For a hobby robot arm, a camera gimbal, or a low-cost bench prototype, it’s a perfectly reasonable component. I’ve used them myself in test rigs. But that doesn’t make it a replacement for an industrial brushless DC motor, and pretending otherwise is how lines go down.

If your machine specifies a motor from a manufacturer like teco-electric, there is usually a reason related to duty cycle, voltage, feedback, or interface. The cheapest substitute is rarely the actual fix. At least, that’s been my experience with light-to-medium packaging machinery over the last several years. And it’s a lesson that cost this customer a full day of downtime, a burnt MG995, and a second failed industrial motor before the real problem—a loose bracket—was ever addressed.

(Note to self: add “verify mechanical load after motor replacement” to the next quarterly audit checklist. And remind me to keep a spare in stock.)