Put a Number on the Gas Before Blaming the Die
Two shop-floor methods
The reduced pressure test takes a small sample, solidifies it under partial vacuum, and compares its density with a sample solidified at atmospheric pressure. Gas that would stay dissolved in the melt comes out of solution under vacuum and forms visible porosity, so the difference in density gives a readable index of gas content.
Density measurement alone — weighing a sample in air and in water — gives the same information in a simpler form and is often used as the routine check, with the vacuum sample reserved for when a reading looks wrong. Both are comparative tools: the absolute number matters less than the trend against a baseline established for that alloy and that furnace.
Reading the result without over-reading it
A rising gas reading points at the melt: wet charge stock, damp tools, a degraded flux practice, or a degassing step that has been shortened. A stable gas reading alongside unexplained porosity moves the investigation to the die and the process — air entrapment from the shot profile, a blocked vent, or a vacuum system that is not holding.
Sample discipline is what makes the number trustworthy. Take the sample at the same point in the cycle, with the same scoop and the same cooling, and record it against the furnace and the shift. A reading taken from a different location in the bath at a different time is not comparable, and comparing incomparable numbers is how a real trend gets missed.
FAQ
What is the reduced pressure test?
A sample is solidified under partial vacuum so dissolved gas forms visible porosity. Comparing its density with an atmospheric sample gives an index of melt gas content.
Is density measurement enough on its own?
It gives the same information more simply and suits routine checks. Use the vacuum sample when a reading is questionable or when the trend shifts.
What raises gas content in the melt?
Wet or contaminated charge stock, damp tools and ladles, degraded flux practice, and a degassing step that has been shortened or skipped.
How do I make the readings comparable?
Fix the sample point in the cycle, the scoop, and the cooling, and record furnace and shift. Only then does the trend mean anything.
All figures above refer to standard product data sheets; final confirmation is subject to written quotation.