Phase analysis is one of those topics that tends to generate more questions the deeper you go, which is exactly what happened after our recent webinar. What follows are the questions we didn’t get to fully address live, plus a few we anticipated ahead of time because they come up consistently in the field.
New to phase analysis? Check out Augury’s introduction to Phase Analysis first. It will give you the foundation to get the most out of what’s below.
Q: How precise is the time synchronization between wireless sensors? Are we talking milliseconds or microseconds?
This question gets at the hardest engineering problem behind making phase analysis work wirelessly.
For phase to mean anything, sensors need to be capturing data at the same moment in time, or close enough that the difference doesn’t affect the phase angle being measured. Augury’s wireless sensors achieve synchronization in microseconds (one microsecond = 1/1,000,000 of a second), which sounds impressive, but to put it in perspective: at 1,800 RPM, one full shaft rotation takes about 33 milliseconds (one millisecond = 1/1,000 second). There’s plenty of margin to get a valid phase reading.
The harder challenge was achieving that precision without compromising battery life. Tight synchronization is power-hungry. If continuous sync were required, customers would be swapping batteries every six months instead of every two to three years. Not a viable solution when you have thousands of sensors deployed. Augury’s approach is to take synchronized phase snapshots selectively, so you get the data quality you need without the power cost. That balance is what made this practical rather than just theoretically possible.
Q: Are there scenarios where phase data becomes unreliable or difficult to interpret?
Yes, and it’s worth knowing these limitations upfront. Phase is a powerful tool, but it’s only as good as the signal it’s measuring.
There are two situations where caution is warranted:
- Low signal-to-noise ratio: If the 1X vibration amplitude is very low, the phase angle will wander without settling on a meaningful value. A dominant, clean 1X signal is required for phase to be useful. If a machine is running smoothly with minimal vibration, phase data at that frequency will be unreliable. This is generally fine, because there’s likely nothing to diagnose.
- Competing faults at 1X: If two problems are occurring simultaneously at run speed — minor unbalance and misalignment, for example — the sensors will capture a vector sum of both fault signatures. The resulting angle won’t match the classic pattern for either fault cleanly, leading to what I’d call a “mushy” reading. In these cases, phase indicates that the motion is complex, but it won’t isolate a single root cause. Think of it as phase telling you that multiple things are off, without pointing to a single source.
When you encounter either of these scenarios, phase becomes one input among several rather than a standalone diagnostic tool. Cross-reference with amplitude trends, historical data, and recent maintenance events to build the full picture.
Q: Can phase and amplitude data assist in troubleshooting faults that vary with operational conditions?
In some cases, yes. For example, if vibration amplitude increases with a speed increase while the phase remains relatively stable, this points to Mass Unbalance. Conversely, with a similar vibration signature, if the amplitude is increasing while approaching a specific speed —accompanied by a rapid phase shift—this is evidence the machine has a Structural Resonance.
Q: At what point in our diagnostic workflow should we bring phase data in? Is it a primary check or a secondary one?
For now, treat it as a secondary tool, something you reach for once the vibration data has already indicated something is worth investigating.
Augury’s analysts follow this general sequence: vibration, temperature, and magnetic data surface the initial concern. The analyst reviews the data and determines whether there’s a meaningful change that warrants a health status update or further investigation. Once that threshold is crossed — when the question shifts from “is something elevated?” to “what is this and where is it coming from?” — that’s when phase enters the picture.
Phase is most valuable in three situations:
- when differentiating between unbalance, misalignment, and looseness
- when validating whether a repair was effective
- when a vibration pattern doesn’t fit a clean fault signature and a tiebreaker is needed
The longer-term goal is to integrate phase signals directly into Augury’s detection algorithms, so phase can surface as part of the initial alert rather than always as a follow-on step. Getting the data quality right first is the necessary foundation for that.
Q: When will Phase Analysis be available across all sensors, and does it need to be activated?
Augury began the rollout at the start of 2026, initially across approximately 100 machines. The expansion is deliberate, adding machines as battery performance and data quality are confirmed to be holding up as expected. Full rollout is planned once behavior has been validated across roughly 1,000 machines.
One important clarification: Phase Analysis is supported on Augury’s current-generation EP sensors. Hazardous-area sensors built on the older hardware platform will not support phase analysis in this first generation. Future hardware iterations will likely address this, but it’s worth confirming which sensors you have deployed before assuming the capability is available.
If you’re unsure whether your sensors are eligible, your Augury team can tell you exactly where you stand.
Still Have Questions?
The on-demand recording from the webinar is available now and includes a full product walkthrough and applied use cases: among them, a real-world example of an 80° phase shift that identified rotor unbalance on a fan well ahead of a failure event.
For those newer to the topic, Augury’s Phase Analysis overview covers the fundamentals of how phase relationships reveal fault sources that amplitude analysis alone can’t identify. You can also download a copy of our one-pager
Want to share what you’ve learned? Download the Phase Analysis one-pager for a quick-reference overview for your team, or request a demo to see how it works in your environment.
