547Mo Alloy Flange Sealing Surface Surfacing: A Practical Review of Electrode Selection, Cracking Control and Heat Treatment
Literature Overview and Core Technical Challenge
The paper by Fu Zhijian, Bi Xiaomin and Fan Fei addresses a persistent shop-floor headache: surfacing of 547Mo alloy onto flange sealing faces. 547Mo is a high-alloy, martensitic-type steel with substantial molybdenum content, used where corrosion resistance and hardness must coexist. The technical difficulty is not the chemistry of the deposit but the metallurgical behaviour during and after welding. The authors report three failure modes: microcracking in the as-welded layer, cold cracking, and detachment of the surfacing layer. All three lead to rework or scrapping of expensive flanges, and any of us who has run a heavy-wall vessel shop knows the cost of a scrapped forged flange late in the fabrication schedule.
The core of the paper is a comparison between two consumables. The conventional D547Mo electrode, a manual metal-arc electrode, produces an as-welded deposit with high hardness. That sounds good for wear resistance, but high hardness in a martensitic matrix means low ductility and poor resistance to hydrogen-assisted cold cracking. When multiple layers are deposited, the restraint from the previous layer and the flange base metal creates a triaxial stress state that the hard, brittle deposit cannot accommodate. The authors therefore propose QD547Mo, a flux-cored wire for CO2 shielded welding, with a modified alloy system. The modification lowers as-welded hardness, improves crack resistance, and allows the hardness to be raised later by a controlled tempering heat treatment.
Technical Points: Consumable Design, Process Window and Heat Treatment
The paper does not give the full chemical composition of QD547Mo, but the logic is clear from the described behaviour. The alloy system is adjusted so that the as-deposited structure is not fully martensitic. It may contain retained austenite or a softer transformation product, giving the deposit enough ductility to withstand the restraint stresses of multi-layer surfacing. After deposition, a tempering treatment transforms the structure to a harder, more stable condition. This is a classic two-step approach: weldability first, properties second.
The authors also emphasise that welding parameters must be controlled. Heat input affects cooling rate, which affects the as-welded microstructure. Too low a heat input gives fast cooling and a hard, crack-sensitive deposit. Too high a heat input gives excessive dilution from the base metal and a coarser structure. The paper does not tabulate the exact parameters, but from my experience with similar molybdenum-bearing martensitic steels, the practical window is roughly:
| Parameter | Typical Range | Shop-Floor Rationale |
|---|---|---|
| Welding process | FCAW with CO2 shielding | Higher deposition rate than SMAW, better operator control |
| Current | 180–260 A | Depends on wire diameter; avoid low current for stability |
| Voltage | 24–30 V | Ensure stable arc and adequate penetration |
| Travel speed | 25–40 cm/min | Balance heat input against dilution |
| Preheat | 150–250 °C | Reduce cooling rate and hydrogen risk |
| Interpass temperature | 200–300 °C | Avoid excessive heat build-up in multi-layer deposits |
| Post-weld tempering | 600–700 °C, 2–4 h | Raise hardness and relieve residual stress |
The heat treatment is the critical step. The authors state that tempering temperature and holding time must be adjusted to achieve the required hardness. Too low a tempering temperature leaves the deposit hard but brittle. Too high a temperature softens it below the service requirement. The holding time must be long enough to homogenise the structure but not so long that grain growth or over-tempering occurs. In practice, I would specify a tempering temperature based on the actual as-welded hardness and the target hardness, and I would verify with hardness traverses across the deposit and into the heat-affected zone.
Engineering Practice: What the Shop Floor Actually Cares About
From a production perspective, the choice between D547Mo electrode and QD547Mo flux-cored wire is not just a metallurgical decision. It is a decision about productivity, operator skill, and rework risk. SMAW with D547Mo is slow, requires highly skilled welders to avoid arc strikes and undercut on the sealing face, and produces a deposit that is difficult to machine because of its hardness. FCAW with QD547Mo is faster, more forgiving, and produces a smoother surface. The wire feed system is more complex, but on a flange sealing face, the benefits outweigh the equipment cost.
The authors do not discuss one issue that I have seen repeatedly: the effect of dilution on the sealing face composition. The first layer of surfacing always picks up base metal. On a flange made of carbon steel or low-alloy steel, the dilution can be 20–30%, which significantly reduces the chromium and molybdenum content of the first layer. The sealing face must have the required corrosion resistance throughout its thickness, so a minimum of two layers is usually necessary. The second layer has much less dilution and achieves the target composition. I would specify a minimum build-up thickness of at least 5 mm after machining, and I would verify the chemical composition of the final machined surface by spectrographic analysis.
Another practical concern is the preheat and interpass temperature control on a large flange. A flange of 500 mm diameter and 100 mm thickness has significant thermal mass. Maintaining a uniform preheat across the sealing face requires either a furnace or multiple gas torches. If the preheat is uneven, the deposit will cool at different rates around the circumference, leading to differential contraction and potential distortion of the sealing face. I have seen flanges that were perfectly round before surfacing and out-of-round after, simply because the preheat was applied on one side only.
The paper also mentions the risk of deposit detachment. This is usually caused by inadequate surface preparation. The sealing face must be machined clean, free of oil, grease, and rust, and preferably with a slight surface roughness to promote mechanical interlocking. If the surface is too smooth, the deposit can peel off under thermal cycling. I would specify a surface roughness of Ra 6.3 to 12.5 µm for the machined surface before surfacing, and I would inspect the surface with magnetic particle testing before deposition.
Key Questions and Reflections
The paper raises a fundamental question: why use a martensitic stainless steel for a flange sealing face at all? The answer is that 547Mo offers a combination of hardness and corrosion resistance that austenitic stainless steels cannot match. In a high-temperature, high-pressure hydrogen service, for example, the sealing face must resist both hydrogen attack and galling. Martensitic steels with molybdenum are a reasonable choice, but they demand careful welding practice.
The authors' solution — a modified flux-cored wire with lower as-welded hardness and a post-weld temper — is sound. I would add one caution: the tempering treatment must be applied before any machining of the sealing face. If the deposit is machined in the as-welded condition, the residual stresses will be released unevenly, causing distortion. The sequence should be: surface preparation, preheat, surfacing, stress relief or tempering, then final machining.
I would also question whether CO2 shielding is the best choice for this alloy. CO2 is an oxidising gas, and molybdenum has a high affinity for oxygen. Some molybdenum will be lost in the arc, and the deposit may have slightly lower molybdenum content than the wire. A mixed gas of 80% Ar + 20% CO2 would reduce oxidation and give a more stable arc. The authors chose CO2 probably for cost and availability, but I would specify argon-rich shielding for a high-alloy deposit.
Study Insights and Implications
This paper is a good example of the difference between a consumable specification and a welding procedure. The consumable specification tells you what the wire contains. The welding procedure tells you how to use it. The authors have done the right thing by focusing on the procedure — preheat, interpass temperature, heat input, and post-weld heat treatment — because that is where the cracking risk is actually controlled.
For the shop floor, the key takeaway is that a hard surfacing alloy is not inherently crack-resistant. The cracking resistance comes from the combination of a ductile as-welded structure and a carefully controlled heat treatment. If you try to achieve the final hardness in the as-welded condition, you will fight cracking every step of the way. If you accept a lower as-welded hardness and then temper to the required value, you will have a much smoother production run.
The paper also highlights the importance of qualification testing. Before any production surfacing, I would require a welding procedure qualification that includes: hardness traverses across the deposit, chemical analysis of the final layer, metallographic examination for microcracks, and a bend test to verify ductility. The qualification should be done on a test piece of the same material and thickness as the production flange, with the same preheat and interpass temperature.
Conclusion
The authors have made a practical contribution by demonstrating that a modified flux-cored wire, used with controlled parameters and a post-weld temper, can overcome the cracking problems associated with conventional D547Mo electrodes. The approach is transferable to other hard-surfacing alloys where as-welded hardness is the enemy of crack resistance. In my own experience, the most important lesson is that the welding procedure, not the consumable, is the primary control for cracking. A well-designed procedure with a ductile as-welded deposit and a proper temper will always outperform a hard consumable used with inadequate thermal control. I would recommend this paper to any engineer responsible for surfacing of high-alloy flanges, and I would add the practical caveat that preheat uniformity, surface preparation, and post-weld heat treatment sequence are as critical as the consumable choice itself.
Zhuojin Pipe Fitting Co., Ltd