You have run the test. The result is above the limit. The customer is asking questions. The production manager is asking for answers. And everyone is looking at the furnace.
The furnace is probably not the problem.
I have spent 25 years troubleshooting aluminium melt cleanliness failures for aerospace applications. I have seen hydrogen failures, inclusion failures, and failures that turned out to be measurement errors. The pattern is always the same: the furnace is blamed first, but the root cause is usually in the melt treatment system or the sampling procedure.
This article walks through the most common causes of aerospace cleanliness test failures, in the order I check them.
First, Confirm the Test Is Valid
Before you change anything in the furnace, confirm the test result is real. I have seen more than one “failure” that turned out to be a sampling error.
| Test | Common Sampling Error | How to Verify |
|---|---|---|
| Hydrogen (RPT) | Sample taken from surface, not bulk | Sample from mid-bath, below surface |
| Hydrogen (Telegas) | Probe not purged, moisture in carrier gas | Purge probe, check gas dew point |
| PoDFA | Sample mould contaminated, wrong filtration | Clean mould, verify filter integrity |
| K-Mold | Sample not representative, fracture surface dirty | Sample from mid-bath, clean fracture |
A foundry in Europe failed a hydrogen test three times in one week. They were ready to replace the degasser. I asked to see the sampling procedure. The operator was taking the sample from the surface of the holding furnace, where hydrogen had already escaped. We retrained the operator. The next sample passed.
What to do: Verify your sampling procedure before you change the furnace. Check the probe, the mould, and the sample location. A clean sample is the foundation of a valid test.
Cause 1: Hydrogen Content Too High
Hydrogen is the most common cleanliness failure in aerospace aluminium. The target is below 0.06 ml/100g Al. Most commercial foundries run at 0.15–0.25 ml/100g.
Hydrogen comes from moisture. Every time aluminium meets water vapour, it absorbs hydrogen. The hydrogen stays dissolved in the liquid metal. During solidification, the solubility drops sharply. The hydrogen forms bubbles. Those bubbles become porosity.
Sources of hydrogen:
| Source | How It Enters | How to Control |
|---|---|---|
| Wet scrap | Moisture on charge | Pre-dry scrap, covered storage |
| Damp refractories | Moisture in lining | Proper dry-out, sealed furnace |
| Humid air | Atmosphere | Sealed furnace, inert gas blanket |
| Wet degassing lance | Moisture on lance | Pre-heat lance before insertion |
| Wet flux | Moisture in flux | Store flux in sealed containers |
| Leaking water cooling | Water in furnace | Check cooling system for leaks |
I visited a foundry in Southeast Asia that was failing hydrogen tests. The degasser was working. The argon was pure. The problem was the scrap. It was stored outdoors. After rain, it was charged wet. We built a covered drying rack. Hydrogen dropped from 0.18 to 0.09 ml/100g in two weeks.
What to do: Check your scrap storage. Check your refractories. Check your degassing practice. Hydrogen is almost always a moisture problem, not a degasser problem.
Cause 2: Inclusion Count Too High
Inclusions are non-metallic particles that remain in the melt. The aerospace target is below 1,000 particles/kg (≥40 μm). Most commercial foundries do not measure inclusions at all.
Inclusions come from three sources: oxidation, refractory reaction, and filtration failure.
Oxidation inclusions form when molten aluminium is exposed to air. They are alumina films and spinel clusters. They form during transfer, pouring, and holding.
Refractory inclusions come from the furnace lining. A cracked or spalling refractory releases particles into the melt.
Filtration failures occur when the filter is bypassed, damaged, or not properly sized.
A foundry in Europe was failing inclusion tests. The PoDFA analysis showed a mix of alumina films and refractory particles. The root cause was the transfer launder. It was unheated. The metal froze on the walls. When the flow resumed, the frozen metal broke off and carried oxide into the mould.
We insulated and heated the launder. The inclusion count dropped by 60%.
What to do: Check your transfer system. Check your refractory. Check your filtration. Inclusions are a system problem, not a furnace problem.
I have written about how to achieve aerospace-grade aluminium melt cleanliness. The principles apply here.
Cause 3: Degassing Practice Is Inconsistent
Hydrogen removal depends on the degassing system. A manual lance degasser is not consistent enough for aerospace work. A rotary degasser is.
| Degassing Method | Hydrogen Removal | Consistency |
|---|---|---|
| Manual lance | 30–50% | Poor |
| Rotary degasser | 80–95% | Good |
| Vacuum degassing | 90–98% | Excellent |
If your degasser is not achieving the target, check three things: gas flow rate, rotor speed, and degassing time.
| Parameter | Target Range | Common Problem |
|---|---|---|
| Argon flow | 15–25 L/min | Too low = poor dispersion |
| Rotor speed | 500–700 rpm | Too low = large bubbles |
| Degassing time | 8–12 minutes | Too short = incomplete removal |
| Gas purity | 99.99% | Industrial grade = contamination |
A foundry in China was using a rotary degasser but still failing hydrogen tests. The argon flow was 8 L/min. The recommended range was 15–25 L/min. We increased the flow. Hydrogen dropped from 0.14 to 0.07 ml/100g.
What to do: Measure your degassing parameters. Check gas flow, rotor speed, and time. Verify gas purity. Measure hydrogen before every pour.
Cause 4: Furnace Atmosphere Is Not Controlled
Aerospace melting requires a controlled atmosphere. If the furnace is open to air, oxygen and moisture enter continuously. Hydrogen absorption increases. Oxidation increases.
What to look for:
| Feature | Requirement | Why It Matters |
|---|---|---|
| Sealed chamber | Gas-tight lid and seals | Prevents air ingress |
| Inert gas blanket | Argon or nitrogen | Displaces oxygen |
| Oxygen monitoring | Continuous O₂ measurement | Proves atmosphere control |
| Moisture monitoring | Continuous dew point | Proves dryness |
A foundry in Europe was failing hydrogen tests despite good degassing. The holding furnace was open to air. The metal was absorbing hydrogen during holding. We sealed the furnace and added a nitrogen blanket. Oxygen dropped from 8% to 0.5%. Hydrogen dropped from 0.12 to 0.06 ml/100g.
What to do: Seal the furnace. Add inert gas blanketing. Monitor oxygen and moisture. A controlled atmosphere prevents hydrogen pickup and oxidation.
Cause 5: Temperature Is Too High
Higher temperature increases hydrogen solubility and oxidation rate. If you are running hotter than necessary, you are making the cleanliness problem worse.
| Alloy | Recommended Melt Temperature | Maximum |
|---|---|---|
| 6061 | 700–730°C | 750°C |
| 7075 | 680–710°C | 730°C |
| A356 | 700–730°C | 750°C |
A foundry in Southeast Asia was running their furnace at 780°C. Hydrogen was 0.20 ml/100g. We lowered the setpoint to 720°C. Hydrogen dropped to 0.12 ml/100g. The lower temperature reduced hydrogen absorption and oxidation.
What to do: Set the furnace to the lowest temperature that achieves the required melt rate. Measure the actual metal temperature, not just the controller display.
Cause 6: Refractory Is Reacting with the Melt
Some refractories react with aluminium. They introduce silicon, iron, or other elements into the melt. They also release particles that become inclusions.
What to look for:
| Refractory Type | Suitability for Aerospace | Notes |
|---|---|---|
| High-alumina castable | Good | Low reactivity |
| Silica-based | Poor | Silicon pickup |
| Phosphate-bonded | Poor | Phosphorus pickup |
| Non-wetting castable | Excellent | Designed for aluminium |
A foundry in China was failing inclusion tests. The PoDFA analysis showed refractory particles. The lining was a silica-based castable. It was reacting with the aluminium. We replaced the lining with a high-alumina, non-wetting castable. The inclusion count dropped by 70%.
What to do: Check your refractory specification. Use non-wetting, high-alumina castables for aerospace work. Verify with your supplier.
A Real Case from Aerospace
In 2024, an aerospace foundry in Europe was failing cleanliness tests on 7075 alloy. The hydrogen was 0.09 ml/100g (target: 0.06). The inclusion count was 2,200 particles/kg (target: 1,000).
They had tried: longer degassing, more argon, new filters, new refractory. Nothing worked.
I spent three days on site. The findings:
| Area | Finding | Action |
|---|---|---|
| Sampling | Hydrogen sample from surface | Retrain operator |
| Degassing | Argon flow at 10 L/min | Increase to 20 L/min |
| Transfer | Open launder, 9 m long | Cover and shorten |
| Filtration | Single stage, 30 ppi | Add second stage, 60 ppi |
| Atmosphere | Open furnace, no blanket | Seal and add nitrogen |
| Temperature | 760°C | Reduce to 710°C |
Results after one month:
| Parameter | Before | After | Target |
|---|---|---|---|
| Hydrogen | 0.09 ml/100g | 0.05 ml/100g | ≤0.06 |
| Inclusion count | 2,200 particles/kg | 620 particles/kg | ≤1,000 |
| Cleanliness test | Fail | Pass | — |
The foundry passed the next customer audit. The cost of the changes was €28,000. The value of the retained contract was €4.2 million per year.
A Simple Troubleshooting Checklist
-
Verify the test. Check sampling procedure before changing anything.
-
Check hydrogen first. Target below 0.06 ml/100g. Measure before every pour.
-
Check degassing. Argon flow 15–25 L/min, rotor 500–700 rpm, 8–12 minutes.
-
Check atmosphere. Sealed furnace, inert gas blanket, O₂ below 0.5%.
-
Check temperature. Lower is better. 700–730°C for most alloys.
-
Check transfer. Covered, heated, short. Minimise turbulence.
-
Check filtration. Two-stage ceramic foam. 30–50 ppi + 50–80 ppi.
-
Check refractory. Non-wetting, high-alumina. No silica.
-
Check scrap. Dry, clean, sorted. Covered storage.
-
Train operators. Sampling, degassing, and recording.
When to Call an Expert
Aerospace cleanliness failures are system problems. The furnace, the degasser, the transfer, the filtration, and the sampling all matter. Finding the root cause requires measurement and experience.
If you are failing cleanliness tests and you have tried the obvious fixes, bring in someone who has solved this problem before. The cost of a consultant is small compared to the cost of a lost contract.
Final Thoughts
Why is your aerospace aluminium melt failing cleanliness tests? It is probably not the furnace. It is the system: hydrogen, inclusions, atmosphere, temperature, transfer, and sampling.
In my 25 years, I have seen foundries pass aerospace audits and foundries fail. The difference is always the same: understanding the root cause before changing the equipment.
If you are failing cleanliness tests, start with the sampling. Then check hydrogen. Then check inclusions. You will find the cause.
About the Author: Hu Shenyue is a veteran aluminium melting furnace expert with 25 years of hands-on experience in China, Southeast Asia, India, the UK, Europe, and North America. He writes practical, experience-based content on furnace selection, maintenance, and optimisation at SmeltPro.
Internal links: How to Achieve Aerospace-Grade Aluminium Melt Cleanliness, How to Reduce Hydrogen Content in Aerospace Aluminium Melts, Case Study – How a Chinese Expert Achieved Aerospace Melt Purity for a US Supplier, Case Study – Solving Aerospace Aluminium Inclusion Problems for a European Foundry, What to Consider When Buying a Vacuum Furnace for Aerospace Aluminium.
