It Started With a Pallet in Receiving

Last Tuesday—around 10:30 AM—I walked past the receiving dock and saw a pallet shrink-wrapped in SKF-branded boxes. Standard stuff. But something about the way the stack was tilted caught my eye. I pulled the packing slip. 4,000 units of radial ball bearings (6000 series). Ordered by a packaging machinery company I'd worked with before.

The spec sheet said they were for a linear actuator assembly—a linear actuator kit they'd been developing. And that's when my gut said: this is wrong.

Honestly, I'm not sure why nobody caught it before. My best guess is the buyer saw 'SKF bearings' and checked the box. But radial ball bearings in a linear application with both radial and axial loads? That's a textbook mismatch.

The Moment of Doubt

I called the project engineer, a guy named Dave. 'Dave, you ordered 4,000 radial ball bearings for your linear actuator? What about thrust loads?'

'We just needed something to move the carriage,' he said. 'Standard ball bearings work fine.'

And that's the assumption—that radial ball bearings handle everything. Actually, the reality is: radial ball bearings take mostly radial load. If you've got axial (thrust) forces—which every linear actuator has when starting and stopping—you need thrust bearings SKF or a bearing that can manage both.

I told Dave, 'These bearings will likely fail within 6 months. Let me show you some numbers.'

Ball Bearing vs Roller Bearing: The Real Trade-Off

This is where the 'which is better ball bearing or roller bearing' debate gets practical. Ball bearings (like deep groove or angular contact) are great for high-speed, low-load applications. Roller bearings handle heavier loads and shock, but often at lower speeds.

For Dave's linear actuator, the main loads were moderate radial + moderate axial. A tapered roller bearing design (like SKF tapered roller bearings) can take both radial and thrust loads in one unit. It's a classic TCO trade-off: ball bearings are cheaper per unit, but if they fail and cause downtime, that single event wipes out the savings.

I pulled up SKF's technical bulletin on bearing selection (SKF.com, March 2025). The recommended bearing for his load profile was a single-row tapered roller bearing—specifically, SKF 30205 series—supported by a small thrust bearing for pure axial loads.

I ran a quick cost comparison:

  • Current radial ball bearings: $2.10 each → $8,400 total
  • Replacement with SKF 30205 tapered + thrust bearing: $5.80 each → $23,200 total
  • But the ball bearings would need replacement every 8 months (based on SKF L10 life calculation under actual loads). Over 3 years: 4.5 replacements → $37,800 in bearings + labor + downtime.
  • Tapered roller option: 1 set lasts 3+ years → $23,200 total.

'The $500 quote turned into $800 after revision fees'—or rather, the $2.10 bearing turned into $14,600 extra cost over 3 years. My God, Dave was shocked.

The Redo

We rejected the entire batch. I'll admit, I felt a little sorry for the sales team who had to scramble, but not sorry enough. The vendor (a local distributor) re-sourced the correct SKF tapered roller bearings and thrust bearings SKF at their own cost—lesson learned for them too.

Dave's team also updated their linear actuator kit design to accommodate tapered rollers. I checked in last week: the first 200 units are in field trial, zero failures after 4 months. Dodged a bullet.

What I Learned

People think expensive bearings cost more. Actually, correct bearings cost less when you account for total cost of ownership. I now include a TCO calculator in every quote review.

One more thing: if you ever ask 'which is better ball bearing or roller bearing'—the answer is always 'it depends on your load profile.' But if you're guessing, you're probably wrong. Consult an engineer, or at least check SKF's bearing selection tool online.

Prices as of April 2025. Verify current pricing at skf.com as rates may have changed.