Quick answer: Ultrasonic measurement can be suitable for cast-iron wall thickness, but a steel preset and a gauge range label are insufficient. Prove the setup on a representative known-thickness sample, confirm an interpretable backwall echo and document the zones that cannot be measured reliably. A missing reading does not mean zero wall thickness.
Define the inspection task first
State whether the goal is manufacturing wall dimensions or remaining thickness in service. Provide the casting grade, expected thickness span, surface condition, access and drawing of the proposed locations. Mark bends, ribs and features whose far surface may not face the probe. A supplier can assess a sample more usefully with this information than with “cast iron” alone.
Why a steel velocity setting can mislead
A time-of-flight gauge converts travel time to thickness using the configured sound velocity. Graphite structure and other metallurgical differences can change velocity in cast iron. Determine the setting using a mechanically measured, representative sample and the instrument procedure. A sample from another grade or casting process may not represent the job. If different zones disagree with independent checks, investigate material variation before accepting one setting for the whole casting.
Choose the probe by demonstration
Ask for a test that covers the thin and thick ends of your intended measurement span. The gauge, probe, frequency and setup must work as a system. Coarse structures can weaken or scatter the signal; suitable probe selection and setup may improve the result, but there is no universal frequency that guarantees performance on all cast iron. A catalog maximum on favorable material is not a demonstrated limit for your casting.
Distinguish coupling problems from unusable geometry
Prepare a consistent contact area and use a couplant suitable for the surface and temperature. Avoid rocking the probe over a rough peak. Reposition at a nearby marked location if the echo is poor. Highly inclined far walls can direct the echo away from the probe. Record an unmeasurable zone explicitly rather than continuing to increase gain until a plausible number appears. Surface preparation should follow the agreed inspection procedure.
Check whether the accepted echo is the right one
A stable numeric display is useful but does not by itself identify the far wall. Where signals are ambiguous, require an instrument or validated setup that allows the echo to be assessed by a trained operator, and compare selected locations with independent thickness checks where practical. Unexpected jumps, weak echoes and values inconsistent with the geometry need investigation. A routine thickness number is not a complete crack, porosity or nodularity examination.
Use a practical acceptance record
Before inspection, agree sampling locations, repeated-measurement practice, required confidence and the engineering limit for the job. Record part and zone IDs, gauge/probe identification, velocity setting, reference sample, surface preparation, accepted readings and failed locations. Repeat the reference check after an interruption or setup change. Request application testing before buying; the result should show where the chosen setup works and where another inspection method is needed.