Imagine a service bench receiving a mechanical seal after a pump begins to leak intermittently. Under angled light the carbon face shows several small raised islands. Someone reaches for abrasive paper to see whether they will polish off. Stop before altering the part. Determine whether the spots stand above the original face, whether the mating ring has matching transfer marks, and what the pump was doing around the leak. Blistering describes a surface feature; seeing that feature does not establish a unique cause.
Make the first inspection without changing the evidence
Isolate the pump electrically and relieve any stored pressure through the approved service procedure. Record which face came from which side of the seal and preserve the faces as a matched pair. With the assembly in a safe inspection area, photograph the carbon face in low-angle light from several directions and add a scale. A raised feature casts a different shadow from a depression, although a photo cannot reliably establish the height. Note its position relative to the running band. Do not scrape the face to test whether the material is soft or apply solvent before the evidence is recorded.
A deposit can sit above the surface. Abrasive particles often leave circumferential tracks. An impact chip is missing material, frequently near an edge. Blisters can appear as localized raised areas in a carbon face and may later leave damage as they break or wear. Material transfer, contamination and earlier cleaning can confuse the picture. A shop with appropriate inspection equipment can check the profile and examine the mating ring; it may send a preserved sample to the seal supplier when the feature cannot be distinguished non-destructively.
Reconstruct the duty around the leak
Carbon grades differ in pore structure, impregnation and suitability for a particular fluid and operating cycle. In some hard-face and carbon pairs, viscous lubricants and repeated cold starts increase shear stress at the interface and can contribute to blistering. The mechanism depends on the material and conditions; a raised mark alone is not evidence that a particular pressure or temperature event occurred. Ask for the actual fluid, time stopped, duration and frequency of restarts, temperature history and changes in the seal chamber or cooling. Separate observations from inferences. If the pump ran in water but the affected face is lubricated by chamber fluid, identify which liquid it actually saw.
As an illustrative comparison, one returned seal has a uniform circumferential scratch on both faces and visible grit in the drain sample. Another shows discrete raised areas with a broken island on the carbon face, and a matching local polished patch on the mating ring. Both can leak. The first points the inspection toward particles and the source of contamination; the second merits a profile check and discussion of the carbon grade and service cycle. Neither example warrants changing to a harder pair without checking lubrication and compatibility.
Ask the supplier for a replacement decision you can test
Send the face pair identity, the photos, operating sequence and relevant fluid information. Ask whether the proposed material is suitable for that specific seal design, fluid and duty. If a material change is recommended, compare repeatability under a controlled test representing the problematic starts, rests and operating temperatures, with leakage and face condition recorded. A single room-temperature pressure hold may not exercise the observed event. Do not reuse damaged faces or return a seal to service on the strength of a polished appearance.
For the report, state the failure feature and the confidence in the cause separately. “Raised features confirmed; cause under evaluation” is more useful than calling every raised mark dry running or every broken island a casting defect. Preserving the original surface lets the next investigator answer the question that matters.
Frequently asked questions
Does a blister prove the pump ran dry?
No. Dry running can damage a face, but a raised feature alone does not establish the operating conditions that produced it.
Can a picture confirm a blister?
It can show shape and distribution. Profile inspection and comparison with the mating ring are more reliable for distinguishing raised carbon from a deposit.
Should both faces be kept if only the carbon looks damaged?
Yes. The other face can retain transfer or contact marks that help reconstruct the event. Preserve their association.
