Why I Think Most Engineers Underestimate Bearings (Until It Costs Them)

Most Engineers Think Bearings Are Simple. I Did Too. Then I Lost $3,200.

When I first started handling transmission component orders back in 2017, I assumed bearings were mostly interchangeable. Need a needle roller bearing? Grab one that fits. Need a thrust bearing? Same thing, right? I was wrong. And that wrongness cost roughly $3,200 in scrapped parts, plus a week of production delay. The worst part? It wasn't the bearing's fault. It was mine.

I've now documented 47 significant mistakes over the past eight years—errors I made personally while specifying or ordering components like INA needle roller bearings, ball bearings, and linear guides. My view today is this: most engineers don't respect the precision required in these parts. They treat them as commodity items. And that's a costly mindset.

My Argument: Precision Components Deserve More Respect Than They Get

Let me be clear. I'm not saying bearings are the most complex part of a drivetrain. They aren't. But they're often the most misunderstood. And when you get them wrong, the consequences ripple through the entire system. Here are three reasons why I've changed my thinking.

Reason 1: The 'What's a Thrust Bearing?' Problem

I used to think a thrust bearing was just a bearing that handles axial load. Technically true. But the nuance matters. Thrust bearings aren't all designed the same way. Some handle high speeds but low loads. Others handle massive axial force but at slower speeds. I once ordered 200 INA thrust bearings for what I thought was a standard application. They arrived, we installed them, and within 48 hours, three failed. Why? I'd selected a bearing optimized for continuous rotation, but our application involved frequent start-stop cycles with high shock loads. The bearing wasn't wrong—my selection was.

That mistake cost $890 in redo work and a 3-day production delay. The lesson? If you can't explain the difference between a thrust bearing for axial loads vs. one for combined loads, don't click 'buy.'

Reason 2: Linear Guide Rails Are Not All the Same

I remember a project in September 2022 where we needed linear guide rails for a precision assembly line. The spec called for smooth motion with minimal play. I selected what I thought was a decent INA linear guide rail. It fit. It moved. But it didn't move smoothly enough. The issue wasn't the rail—it was the preload class. I'd chosen a light preload, thinking 'lighter is better for smooth motion.' Actually, for that application, a medium preload would have eliminated the tiny wobble causing the problem.

We caught the error during testing, but not before we'd assembled 12 units. $450 wasted in labor, plus the embarrassment of explaining to the client why we needed another week. Since then, I maintain a checklist comparing load type, speed, accuracy requirements, and preload classes before any linear guide purchase.

Reason 3: People Forget the Linear Actuator Motor Pairing

This one surprises people. A linear actuator motor is often spec'd based on power and stroke length. That's fine—but it's not enough. The motor's acceleration curve, duty cycle, and thermal characteristics all interact with the bearing choice. In Q1 2024, I approved a motor-actuator combo that looked perfect on paper. The linear actuator motor seemed to have plenty of torque. But we'd paired it with a bearing that had higher friction than the motor was designed to overcome continuously. Result: the motor overheated after 45 minutes of use.

We caught that one before it went to the customer, but it still cost us a day of rework and a $300 expedited shipping fee for the correct bearing. What I learned: never spec a motor and bearing in isolation. They're a system, not two separate parts.

Anticipating the Pushback: 'But Bearings Are Commodities'

I hear this a lot. 'Bearings are standard parts. Just pick the right size.' And yes, if you're talking about a generic ball bearing in a low-stress application, that's mostly true. But here's the thing: the application defines the bearing, not the other way around. An INA needle roller bearing that works perfectly in a high-speed conveyor might fail in a heavy-load press. And a thrust bearing that's fine for a slow rotary table might overheat in a high-speed spindle.

The industry has evolved. What was considered 'good enough' in 2018 may not be adequate today. Machines run faster, tolerances are tighter, and uptime expectations are higher. The old mindset of 'just pick the cheapest that fits' is a liability.

Don't get me wrong—I'm not saying you need to over-engineer everything. But I am saying that treating all bearings as interchangeable is a shortcut to expensive failures. The fundamentals of load, speed, and environment haven't changed. But the execution—the precision, the material quality, the design refinements—has transformed.

Final Thought: The Best Engineers I Know Are Slightly Paranoid About Bearings

I've worked with engineers who seem overly cautious about bearing selection. They double-check specs, ask about lubrication, and question dynamic load ratings. I used to think they were wasting time. Now I realize they've probably made the same mistakes I have—and they're avoiding a repeat.

If you take one thing from this article, let it be this: respect the bearing. Whether you're choosing an INA ball bearing for a fan assembly or an INA thrust bearing for a heavy press, ask the questions you'd ask of any critical component. What's the load profile? What's the duty cycle? What's the environment? The answers will save you money, time, and a lot of embarrassment.

Oh, and keep a checklist. Trust me on that.