How to Verify a Thermowell for Vibration Risk

Back to Blog

What You Need

Gather the well's datasheet and the original wake-frequency or vibration calculation if the site has one, the current operating conditions for the line, and the well's dimensions: unsupported length, root and tip diameters, and bore. The background physics is the same vortex-shedding mechanism described in vortex shedding and the von Karman street, except here the shedding body is the instrument itself.

Confirm the Operating Case Against the Design Basis

A thermowell calculation is only valid for the conditions it assumed. Compare today's normal and maximum flow against the design case: debottlenecking projects, larger pumps, lighter fluids, and gas instead of liquid all raise velocity, and shedding frequency scales with velocity across the well. Design practice keeps the shedding frequency comfortably below the well's natural frequency; a velocity increase erodes that margin from below, and passing through the coincidence point is what drives resonant failure.

If the velocity has grown materially since the last calculation, or nobody can produce a calculation at all for a well in high-velocity service, treat the well as unverified and get the wake-frequency analysis done or redone by a qualified engineer per the site's engineering process. That is the escalation point; everything else on this page is screening around it.

Screen the Geometry for Slenderness

Long, thin wells are the vulnerable ones. The natural frequency of a cantilevered well falls rapidly as unsupported length grows and as the stem thins, so a long straight well in a large line or a deep insertion into a duct deserves more suspicion than a short tapered well in a small pipe. Tapered and stepped profiles exist precisely to keep the tip light and the root stiff; a straight well of the same length is the weaker geometry, as covered in what a thermowell is.

Check the installation details that change effective length: a well mounted on a long nozzle or standoff carries extra unsupported length that the catalog dimensions do not show, and a well repurposed from a thicker-walled line may sit deeper in the flow than its design intended.

Inspect for the Signatures of Fatigue

At the next opportunity the well is accessible, inspect the root area - the highest-stress point - for circumferential cracking, and check the well for tip wear or polishing that indicates the well has been contacting the sensor or shaking hard. From the outside, an audible hum or ringing from the well area at certain flow rates, or a temperature reading whose noise grows with flow, are both worth taking seriously as resonance clues; a sensor being shaken in its well can also fail repeatedly, so chronic element failures at one location are a thermowell question, not a sensor question, a connection discussed in commissioning an RTD in a thermowell.

Verify the sensor is spring-loaded and in tip contact, since a loose element rattling in a vibrating well amplifies damage. Document what you find; a well that screens clean today but sits near its limits should be on a re-check list whenever operating conditions change.

Verifying the Result and Common Mistakes

The verification is complete when three things are true: the operating case is inside the calculated design basis, the geometry and mounting match what the calculation assumed, and the physical inspection is clean. Record all three, because the next process change invalidates the first one silently.

Common mistakes: assuming a well is fine because it has survived years at the old flow rate, ignoring the nozzle length when judging insertion, treating repeated sensor failures as a sensor-quality problem, and letting anyone but a qualified engineer sign off the wake-frequency margin for high-velocity or high-pressure service. When in doubt, defer to the site's mechanical-integrity process.

Frequently Asked Questions