Casting defects · recording

The casting defect register

A blowhole to one inspector is a pinhole to the next. Those two names carry different repairs, so a register that names defects loosely can send the plant after the wrong cause.

By Versatile Equipments Pvt Ltd · equipment manufacturer & calibration provider · updated 2026-09-07

Why the name decides the rest

Everything a foundry does about rejections rests on the name written at inspection. The morning meeting works from it, the corrective action is chosen because of it, and the monthly rejection report is only those names added up. Where the name is unreliable, so is everything built on it.

Several common defects are easy to confuse, because they can look alike while having quite different causes. A blowhole and a pinhole are both gas cavities, but the first points at mould and core gas and the second at the melt. Shrinkage porosity and gas porosity are both internal voids whose fixes pull in opposite directions. Sand inclusion and slag inclusion both appear as dark, dirty pockets, one sending you to the mould and the other to skimming and gating. Each pair carries one appearance and two different repairs.

There is a second cost that rarely gets counted. Where two people disagree, the disagreement tends to be settled by argument rather than evidence, and the register slowly becomes a document the plant does not fully trust.

What a paper sheet loses

• It is often written from memory. A defect seen mid-morning is commonly written up near the end of the shift, so counts drift and the heat or pattern can be attached to the wrong casting.
• The sheet has to survive a foundry. Oil, sand and damp make entries hard to read, and sheets go missing between the floor and the office.
• Some rejects go unwritten. Where the register is used to apportion blame, the softer number is the one that tends to get recorded.
• Every retyping adds error. Floor sheet to spreadsheet to monthly report is two more chances to lose or mistype a figure.

What reaches management is then a total with the reasoning stripped out. It can say that rejections rose last month, but not which pattern, which impression, which heat or which shift, and it cannot show anyone the casting.

Recording it consistently

• Capture at the bench, not at shift end. A photograph taken before the casting moves keeps the evidence, so a disagreement six weeks later can be settled by looking rather than remembering.
• Pick from a shared list. When every inspector chooses from the same short list of defects, matched to what the plant actually casts, names drift far less between shifts.
• Let a senior call settle it. A supervisor or metallurgist, who may sit in another plant, can review the same photograph and correct the classification. Keeping both the original and the correction gives you training material drawn from your own castings.

Down to the cavity and the heat

One casting comes from one impression. A pattern with eight cavities produces eight castings from the same ladle, in the same sand, seconds apart, so when one of them rejects far more often than the other seven, the difference is more likely to be that impression than the metal - a worn, poorly vented or differently fed cavity. That question can only be asked if the cavity number is recorded with the defect, and cavity numbers are cast into most patterns already.

With the heat recorded as well, the register can separate a bad heat from a bad impression from a bad shift. Those three have different remedies, and until the data can tell them apart the plant is choosing which to investigate by instinct.

Turning it into action

• Rank before you act. A small number of defect types usually account for most of what is scrapped, which says where effort belongs and which problems can wait.
• Read it against the sand lab. Many green sand defects follow moisture, strength and grain fineness, so a register read alongside moisture and compactability and permeability records is more likely to suggest a cause. Those numbers need calibrated instruments to be worth reading.
• Check the fix held. The step most often skipped is looking at the weeks either side of a change to see whether the defect actually fell.
• Let the audit take care of itself. Photographs, timestamps, who classified and who changed it cover much of the nonconformance evidence expected under IATF 16949, as a by-product of ordinary work.

For the wider picture of defect types and their causes, start from the casting defects guide.

Frequently asked questions