# Casting Defect Root Cause: Porosity, Shrinkage and Scrap Analysis > Source: https://netray.co/resources/casting-defect-root-cause-porosity-and-scrap-analysis > Publisher: Netray (netray.co), AI and integration engineering for ERP and manufacturing systems. > Section: Knowledge Base ## Short answer Find root causes of porosity, shrinkage, cold shuts and inclusions by linking scrap to heat, die, pattern, shift and process data, and use AI to spot patterns. ## Category ai-automation ## Content **Intro:** Root cause analysis for casting defects works when scrap records connect to process data: which heat, die or pattern, which shift, which furnace temperature, which spray or sand lot. Porosity, shrinkage, cold shuts, inclusions and misruns each have distinct causes, but most foundries cannot see them because defect counts sit in one place and process data in another. Linking the two, then letting people ask questions across them, is where analysis finally stops being anecdotal. **Sections:** #### Defect Types and Where They Usually Start **Body:** Gas porosity often traces to melt practice, hydrogen pickup in aluminum, moisture in sand or coatings, or venting problems. Shrinkage points toward gating and riser design, pour temperature or localized hot spots. Cold shuts and misruns link to low metal temperature, slow fill or die temperature in die casting. Inclusions come from dross, slag, sand erosion or poor filtering. Soldering and heat checking in die casting relate to die temperature, lubrication and die wear. These are starting hypotheses, not diagnoses, and the right test is whether your data shows the defect clustering by a controllable variable. **Bullets:** - Gas porosity: melt degassing, hydrogen, moisture, venting - Shrinkage: gating, risers, chills, pour temperature - Cold shut or misrun: metal and die temperature, fill speed - Inclusions: dross, slag, sand erosion, filtering #### Capture the Variables That Explain the Defect **Body:** A scrap report that says 14 percent porosity on Part A is not actionable. You need the same entry to carry heat, furnace, ladle or holding furnace, die or pattern, machine, shift, date and time, and operator, and ideally process parameters such as pour temperature, shot profile or die temperature. For aluminum, include degassing logs and density index results if you run them. For sand, include sand tests such as moisture and compactability. Radiography or X-ray results, evaluated against a reference such as ASTM E505 for aluminum and magnesium die castings, give you severity levels rather than pass or fail alone. #### Analysis Methods That Work on a Plant Floor **Body:** Begin with a Pareto of scrap dollars by part and defect. Then stratify the top defect by shift, die, heat, machine and day of week to look for clusters. Control charts on key variables reveal drift before scrap spikes. Design of experiments helps when you have a candidate change but is a heavier tool. Most gains come from basic stratification done consistently, such as discovering that porosity doubles on third shift after a furnace charge change. Document each investigation, the action taken and the outcome, so the next engineer does not repeat the same discovery. **Bullets:** - Pareto by scrap dollars, not scrap count - Stratify by die, heat, shift, machine and furnace - Control charts on temperature and shot parameters - Keep a log of actions taken and results #### Using AI to Search Across Scrap, Process and Quality Data **Body:** A private AI assistant over your ERP scrap records, spreadsheets, SPC exports and inspection reports can answer questions like which dies had the highest porosity scrap in the last 90 days, or whether scrap on Part B correlates with a specific heat supplier. It can draft stratified tables and point to records for the engineer to examine. It is not a replacement for metallurgical judgment, and correlation in messy data can mislead, so keep an engineer in the loop. The practical value is speed: queries that took a day of spreadsheet work return in minutes. #### When This Is Not the Priority **Body:** If scrap is a few percent and stable, and customer returns are rare, deeper analytics may not be your best use of time. Focus first on consistent defect coding and basic data capture. Without clean codes, no analysis tool will help. Where scrap is concentrated in a few parts, a focused project on those parts will beat a plant-wide analytics initiative. **Cta:** If scrap data and process data live in separate places, we can help you connect them and query across them privately, with every answer pointing back to source records. **Faq:** #### What causes porosity in castings? **Answer:** Common causes include dissolved gas in the melt (especially hydrogen in aluminum), moisture in sand or coatings, trapped air from poor venting or turbulent fill, and shrinkage from inadequate feeding. The correct fix depends on which type you have, so confirm by sectioning or X-ray and by looking at where it appears. #### How do I reduce scrap in die casting? **Answer:** Stratify scrap by die, machine, shift and alloy to find clusters, then check die temperature control, spray practice, shot profile, venting and die wear. Fix the largest scrap-dollar part first and track results on a control chart. #### What standard is used to grade casting porosity by X-ray? **Answer:** ASTM E505 provides reference radiographs for aluminum and magnesium die castings, and ASTM E155 for aluminum and magnesium castings in other processes. Customers often specify acceptance levels from these references on the drawing or purchase order. #### Can AI find the root cause of casting defects? **Answer:** It can speed up finding patterns across scrap, process and heat data, but root cause still needs metallurgical judgment and often a trial. Treat its output as hypotheses with cited data, not conclusions. ## Related Slugs - spc-integration-erp-quality - casting-yield-and-scrap-cost-in-quotes - ai-predictive-maintenance-erp-integration - foundry-heat-lot-traceability-for-castings ## Search terms casting porosity causes, casting defect root cause, die casting scrap reduction, foundry defect analysis, shrinkage vs gas porosity, casting quality analytics, foundry scrap analysis --- Talk to an engineer: https://netray.co/contact?intent=erp_ai | Machine catalog: https://netray.co/agent/catalog.json