A breakout in continuous casting isn't a quality problem, it's liquid steel escaping a torn shell onto the casting floor. Every caster operator knows the feeling of watching mold temperature readings and hoping nothing looks wrong. One plant replaced that hope with signal monitoring across the mold, and breakouts dropped sharply within a year of going live. Here's what got watched, and what changed. Sign up to see how the same signal monitoring could apply to your caster.
70% Fewer
Reduction in breakout incidents after signal monitoring went live
400+ Heats
Consecutive heats cast without a breakout on the monitored strand
Real-Time Signals
Mold data reviewed continuously instead of only when something looked off
The Problem
Breakouts were rare enough that they didn't happen every week, but frequent enough that they showed up on the safety and cost report every quarter. Each one meant an emergency stop, a cleanup, and a review that usually concluded the warning signs had been there in the data, just not looked at closely enough in time.
Signals That Predict A Breakout
Thermocouple Temperature Asymmetry
Mold thermocouples were compared side to side in real time, since an uneven pattern is an early breakout signal
Mold Level Fluctuation
Level stability was tracked continuously, since erratic level readings often precede shell tearing
Casting Speed Versus Temperature Correlation
Speed changes were checked against mold temperature response to flag a mismatch before shell thickness suffered
Copper Plate Wear Tracking
Plate thickness readings from every changeout were logged so a thinning plate never got reused past its limit
Watch The Signals That Actually Predict A Breakout
Oxmaint logs mold temperature, level, speed, and plate wear against each caster so a warning pattern is visible before it becomes an incident. Sign up for a free trial to see it against your own caster data, or book a demo to walk through your breakout risk points.
Breakout Causes Addressed
| Breakout Cause |
Before |
After |
| Sticker breakouts |
Occurred periodically, hard to predict |
Rare, caught through level and temperature signals |
| Thermal hot spot breakouts |
Discovered only after shell weakening |
Flagged early through thermocouple asymmetry |
| Speed and temperature mismatch |
Adjusted reactively after visible strain |
Caught through correlation monitoring before impact |
| Worn plate reuse |
Relied on visual inspection alone |
Prevented through logged thickness data per plate |
The Outcome
None of the four signals were new information the mold wasn't already producing. What changed was reviewing them together, continuously, against each mold's own history, instead of relying on an operator noticing something looked slightly off during a busy shift.
Frequently Asked Questions
Q
What causes a breakout in continuous casting?
A breakout usually starts with a weak spot in the solidifying shell, caused by uneven cooling, a thinned mold plate, or a sudden mismatch between casting speed and heat extraction.
Q
How does predictive maintenance detect a breakout risk in advance?
It watches patterns like thermocouple asymmetry and mold level fluctuation continuously, so a combination that historically preceded a breakout gets flagged as it starts forming, not after the shell has already weakened.
Q
Can a CMMS reduce breakouts without changing casting practices?
Yes, a CMMS focused on tracking mold condition and signal patterns catches maintenance-related breakout causes on its own, without requiring any change to casting speed or practice.
Turn Mold Data Into A Breakout Early Warning System
Oxmaint logs mold temperature, level, speed, and plate wear against each caster so warning patterns are visible before they become incidents. Sign up for a free trial to see it against your own caster data, or book a demo to walk through your breakout risk points.
