The Mysterious 0.6× Frequency — a Frequency That Shouldn't Exist (But Does)

Introduction

You know the rule: in a bearing with a stationary outer race, the cage rotates at roughly 0.4× of shaft speed. So when a peak appears at 0.4× in the spectrum, the first thought is a cage or rolling-element defect. But what if you see a clean peak at 0.6× instead? Not 0.4×, not 1×. Just a lone peak at 0.6×. Where does it come from? SDT340 ultrasound, which catches early-stage friction, and Artesis e-MCM electrical signatures can help confirm the hypothesis.

A thought experiment: changing the reference frame

Normally we stand on the ground — the outer race is stationary. The inner race rotates at 1×. The cage rotates at 0.4×.

Now imagine "sitting" on the inner race — as if the shaft weren't moving.

In this reference frame, the outer race rotates backward at −1×. The cage's speed relative to you: 0.4× − 1× = −0.6×. In other words, in this frame, the cage and rolling elements pass a fixed point on the inner race at a frequency of 0.6×.

Where does the 0.6× vibration come from?

Even with no damage on the inner race, any asymmetry between rolling elements creates a disturbance every time the "odd one out" rolling element passes a fixed point on the inner race.

Examples of asymmetry that generate excitation at 0.6×:

• One rolling element is slightly larger or smaller than the rest
• One rolling element has a small spall
• A cage pocket is worn or cracked, so one rolling element sits differently than the others

Every time this rolling element completes its cycle — at an interval of 0.6× — it slightly lifts or displaces the shaft. This micro-impact repeats at the 0.6× frequency.

Why isn't 0.6× always visible?

If the asymmetry is always present, why is 0.6× so rarely seen in spectra? It comes down to the signal transmission path.

Signal path	Description
Outer race defect — short path	Outer race → housing → sensor. Direct transmission, minimal attenuation.
Rolling-element asymmetry — long path	Rolling element → outer race → housing → sensor. Significantly more attenuation.

0.6× shows up in the spectrum only when the signal energy is sufficient to "survive" this longer path.

That's exactly why cylindrical roller bearings show 0.6× more often than ball bearings: the line contact of rollers transmits energy more efficiently than the point contact of balls.

What can cause 0.6×?

Cause	Description
Rolling-element size variation	Manufacturing tolerances or progressive wear that leaves one element larger or smaller than the rest.
A spall on one rolling element	A localized defect that periodically disrupts the load zone at an interval of 0.6×.
Cage pocket wear	Excessive clearance in one pocket causes that rolling element to move differently than its neighbors.
Cage crack	Structural damage that changes the position or motion of one or more rolling elements.

Diagnostic tip

If you see a clean peak at 0.6× with no other harmonics, don't rush to diagnose a "cage defect" (that's 0.4×). Think instead about an asymmetric rolling element viewed from a reference frame fixed to the inner race.

And check the bearing type: if it's a roller bearing, the probability of seeing 0.6× is significantly higher.

A question for discussion

If 0.6× appears in the spectrum, what other signatures might be present? Share what you've observed in practice — get in touch with our engineers.