Shimano STEPS Power Transmission OEM logo
Technical article

What Happens When a Linear Actuator Fails—and How to Tell Which Scenario You're In

2026-11-20 / Engineering Desk

There's no single answer, because there are at least three different problems

I work as a quality and brand compliance manager at an industrial OEM. Four years on the receiving inspection side, roughly 200+ delivery batches a year—gearboxes, spur and worm gears, brushless servo motors, linear actuators. Our customer base is mostly mid-range industrial operations in North America.

Every time someone calls about a failed linear actuator, the conversation starts the same way: "Is this a quality problem?"

Sometimes it is. But most returned actuators I've handled have no single manufacturing defect. They failed for one of three completely non-overlapping reasons—and the usual "just size up" advice only solves one of them.

So instead of a universal recommendation, here's what those three scenarios actually look like. Find the one that matches your setup.

Scenario A: The actuator is fine—but the spec was wrong from the start

From the outside, a dead actuator looks like a manufacturing failure. You open it up and the motor windings are burnt, the bearing is dry, the screw looks intact—because the load never damaged the screw. It stressed the motor until the insulation cooked.

This is the most common pattern I see in failure review logs. More than half of the early-life failures I've tracked over the past three years traced back to a sizing calculation made at the quotation stage: someone picked the actuator against nominal load instead of measured peak dynamic load. Normal engineering practice calls for a linear actuator's maximum load rating to be at least 2× your application's maximum dynamic load. If that actuator is sized to run at its maximum working load, your motor is running near the edge every time the mechanism accelerates or friction changes.

We had one batch of 300 actuators for a packaging line pusher—eleven motor burnouts in under six months. The supplier insisted the units were within spec. They were. But the customer's duty cycle wasn't smooth. A sudden acceleration phase in each shift pushed the load to roughly 1.8× steady state. The actuator had been selected for the steady-state figure. Certified fine. Product fine. Sizing wrong.

Here's something vendors won't tell you: catalog load ratings are typically established under one specific condition set—ball screw, horizontal mounting, 25°C ambient, perfect alignment, intermittent duty. Deviate on any of those—inclined mounting, continuous operation, no forced cooling—and your usable load drops below the nameplate number.

Sizing software since 2023 calculates peak dynamic load before making a recommendation. The catch is that someone has to actually use it.

Scenario B: The environment is harsher than the label suggests

IP ratings are a quietly misleading label. Under IEC 60529, IP65 means dust-tight and protected against water jets—from a nozzle. It does not cover continuous washdown, prolonged submersion, or the kind of sustained abrasive dust that grinds against seal faces during normal operation.

A moisture-ingressed actuator often looks fine on the receiving dock. The seal appears intact. There's no way to tell visually how the protection scheme will hold up over eighteen months.

One OEM customer put an IP65 actuator on a food-processing machine rinsed twice per shift. The seals degraded at about fourteen months—just past warranty. Supplier clean. Customer furious. Nobody understood why until we dug into it: the IP65 water-jet test has no relationship to continuous washdown.

Practical check: if the application involves frequent washdown, condensation, or direct contact with cutting fluid, don't just look at the IP column. Ask the supplier specifically about seal material and design assumptions. If the answer is vague, file that as one of the answers.

Scenario C: The motion profile is slowly killing it

This is the hardest one to diagnose because everything looks inside spec. Load is fine. Environment is fine. And it still fails.

The usual pattern: high acceleration, frequent cycling, no dwell. Every start and stop sends an elastic shock through the screw. Rolling elements micro-pit at the contact surface before the rest of the assembly catches up.

The counterintuitive part—you may not need a larger actuator. You need a slower motion profile. Extending acceleration ramp time from 0.25 seconds to 0.5 seconds, or dropping acceleration by 30%, extends fatigue life substantially, and throughput on most mid-speed automation lines barely moves. The bottleneck is typically loading and unloading, not the actuator stroke.

Motion-profile planning software can quantify the tradeoff between throughput and fatigue life before you commit. Manual tuning is basically guesswork—"it feels smooth" → "hmm, maybe slightly jittery" → "let's try a bigger size." With quantified validation, you know which parameter is actually costing life.

How to tell which scenario you're actually in

Start with three questions:

  • Have you measured actual peak dynamic load—not just static load—including the acceleration phase?
  • Does the installation match the catalog condition assumptions, or is it inclined, continuous-duty, or exposed to washdown?
  • What are the acceleration and cycle frequency? If you're running tens of thousands of cycles per shift, fatigue life won't show up on a static load rating.

Teardown gives you the fastest answer:

  • Burnt motor, intact screw → Scenario A (overload/sizing)
  • Pitted screw or leaking seal → Scenario B or C (environment or motion profile)
  • Intact seal, decent screw, failed bearing → probably C

When in doubt, don't reach for "bad quality" as the default explanation—it hides the real root cause.

My experience is based on roughly 200 mid-range industrial delivery batches, mostly North American customers. Extreme-temperature environments, high-purity semiconductor applications, or severe corrosion sites need a different framework entirely.

Last thing

If you made it this far waiting for an answer about which model number to pick, I don't have one. The three scenarios point to three different systemic problems—load, environment, motion profile—and they require different fixes. Sometimes more than one at once.

Most of the time, the number that needs changing isn't on the actuator datasheet. It's in your sizing calculation, your installation assumptions, or your motion profile settings.

The most reliable starting point is a documented teardown of the failed unit. Your own data is the only thing that will actually tell you which scenario you're in.

If you need components matched to your scenario, Boston Gear's catalog covers speed reducers, gearboxes, spur and worm gears, brushless servo motors, and linear actuators—available through distributor networks including the Charlotte, NC region. But I'd still work through the checklist above before opening a catalog.

(If you landed here searching for "2026 Boston Marathon gear" or "natural blood pressure reducers"—those are unrelated to anything in this article. Boston Gear is an industrial power transmission brand, not a running apparel label. And a "pressure reducer" in transmission context means a mechanical device, not a medical one. Search terms overlap sometimes. Topics don't.)

Shimano STEPS Engineering Desk

Application notes from drive unit, brake and service documentation teams.