When you receive a mill certificate or a weld inspection report for duplex stainless steel - for example SAF 2205 or SAF 2507 - you will see two numbers that look almost interchangeable: a Ferrite Number (FN) and a ferrite percentage (%). One line may read "52 FN" while another reads "48% ferrite." Are they the same quantity? Can you compare them directly? If the specification calls for "40–60% ferrite," does a result of "52 FN" automatically pass?

The short answer is no. Ferrite Number and ferrite percentage are related but not identical scales, and confusing them is one of the most common causes of material rejection, unnecessary rewelding, and disputed inspection results in duplex stainless steel fabrication. This guide explains what each number means, how each is measured, why the two scales diverge, and exactly how to read them on a real inspection report.
What Are Ferrite Number (FN) and Ferrite Percentage?
Ferrite Number (FN) is a standardized, instrument-based index of how magnetic a stainless steel weld is; ferrite percentage (%) is the actual volume fraction of the ferrite phase in the microstructure. FN is reproducible and non-destructive, but indirect; ferrite percentage is a direct microstructural measurement that requires laboratory metallography.
Duplex stainless steels get their name from a two-phase microstructure: roughly half austenite (non-magnetic, face-centered cubic) and half ferrite (magnetic, body-centered cubic). Engineers care about this ratio because it controls strength, corrosion resistance, and toughness. Two measurement languages have grown up around it:
Ferrite Number (FN) is defined by AWS A4.2M (and ISO 8249). It is not a percentage. It is an arbitrary scale calibrated against a set of standardized reference blocks, so that any ferritescope anywhere in the world reads the same FN on the same block. The practical measurement range runs from about 0.1 to 90 FN.
Ferrite percentage (% ferrite) is the true volume fraction of the ferrite phase, measured by examining a polished and etched cross-section under a microscope. It is what metallurgists actually mean by "how much ferrite is in this steel."
The crucial point: a ferritescope does not count ferrite grains. It measures magnetic permeability. Magnetic permeability depends on ferrite content, but also on alloy composition, residual stress, temperature, and phase morphology. AWS A4.2 solved the reproducibility problem by defining FN against calibration standards - but in doing so, it deliberately stopped claiming that FN equals volume percent.
Why Do Duplex Inspection Reports Show Two Different Numbers?
FN is the field-friendly, non-destructive, contractually standardized number used for weld acceptance; ferrite percentage is the lab-based metallurgical value used for material conformance and critical service qualification. Reports include both because welding codes specify limits in FN, while product specifications and sour-service requirements are often written in volume percent.
During fabrication, inspectors need a fast, non-destructive test they can run on a completed weld. A handheld ferritescope gives an FN reading in seconds, on site, with little or no sample preparation. This is why weld procedure qualification (WPQR) and production weld acceptance are almost always written in FN.
At the mill, and in critical service such as sour (H2S) environment per NACE MR0175 / ISO 15156, buyers require proof of the actual phase balance. That requires polishing a coupon, etching it, and counting ferrite points per ASTM E562, or by automated image analysis per ISO 17781. The result is reported as volume percent. Both numbers appear on certificates because they support different decisions: FN for weld acceptance, and % ferrite for material fitness-for-service.
How Does a Ferritescope Measure Ferrite Number?
A ferritescope measures the magnetic induction of the near-surface layer and converts it to FN using a calibration traceable to AWS A4.2 reference standards. It is fast, non-destructive, and reproducible, but it reads only near-surface ferrite and its uncertainty widens at very high ferrite contents or on curved, rough, or thin sections.

The instrument works on the magnetic induction principle. The probe contains a coil that creates a low-frequency magnetic field. Because ferrite is ferromagnetic while austenite is essentially non-magnetic, the field response changes with the amount of ferrite present. The instrument is calibrated against AWS A4.2 calibration blocks traceable to national standards, spanning roughly 0.1–90 FN. Common instruments include the Fischer FeritScope series and equivalent handheld gauges.
- Practical limitations that affect the number on your report:
- Measurement depth is only a fraction of a millimeter, so surface condition (paint, scale, heat tint, machining marks) matters.
- Curvature, wall thickness below roughly 5 mm, and rough weld caps require compensation or special probes.
- Residual stresses and cold work can introduce strain-induced martensite, which is also magnetic and inflates the reading.
- At high ferrite levels (above roughly 60 FN in duplex), the magnetic response becomes nonlinear and the stated measurement uncertainty grows.
How Is Ferrite Percentage Measured in the Laboratory?
Ferrite percentage is measured destructively by metallographic examination: a polished cross-section is etched, and the ferrite area fraction is counted by systematic point counting (ASTM E562) or by automated image analysis, then reported as volume percent. This is the referee method, but it requires a coupon and a laboratory.
The specimen is cut, mounted, ground, polished, and etched - commonly with Beraha-type reagents that color ferrite and austenite differently. Under a light microscope, the metallographer places a grid over the microstructure and counts how many grid points fall on ferrite grains, following the systematic manual point-count procedure in ASTM E562. Modern laboratories use automated image analysis, which is faster and more repeatable when etch contrast is good.
ISO 17781:2017 specifically governs quality control of the microstructure of duplex stainless steels for the petroleum and petrochemical industries, and it relies on this metallographic approach. Because this method actually looks at the grains, it is the referee measurement when a dispute arises between ferritescope readings and the specification.
Can You Convert FN Directly to Ferrite Percentage?
There is no exact, universal conversion. Below about FN 10 - the range typical of austenitic 304/308 welds - the two scales nearly coincide; above FN 10, which covers the entire duplex range, they diverge, and the relationship depends on alloy grade. Any conversion factor is an approximation, and acceptance criteria must always state which unit they use.

This is the single most important point in this article. The WRC-1992 constitution diagram uses FN iso-lines precisely because the older Schaeffler and DeLong diagrams - which read out in volume percent - were found to be inaccurate for modern nitrogen-alloyed and duplex weld metals. The welding community moved to FN to make measurements reproducible between instruments and laboratories, not because FN is easy to convert into percent.
Approximate relationships reported in industry literature and instrument practice:
For 22Cr duplex (e.g., 2205 base plate, ER2209 filler): ferrite percentage is roughly 0.65–0.75 times the FN value. A reading of 50 FN corresponds to about 35–40% ferrite.
For 25Cr super duplex (e.g., 2507, ER2594): published round-robin work on single-pass welds has reported FN ≈ 1.1 × ferrite fraction, i.e., ferrite percent ≈ 0.9 × FN. A reading of 55 FN corresponds to roughly 50% ferrite.
These factors vary with weld pass count, dilution, nitrogen loss from shielding gas, and heat input. They are not contractually interchangeable. The practical rule: if your specification says "40–60% ferrite," do not pass a ferritescope report that says "40–60 FN" without a documented, grade-specific conversion agreed with the engineering contractor, or a metallographic confirmation. Always check which unit the acceptance criterion uses, and make sure the instrument was calibrated and the method documented in the same unit.
What Ferrite Range Is Acceptable for 2205 and 2507?
For solution-annealed 2205 and 2507 base metal, the target ferrite window is 40–60%. For weld metal, NACE MR0175 / ISO 15156 and API Technical Report 938-C broadly require 35–65% ferrite for wrought material and roughly 30–60% for weld metal, with project specifications often tightening this to 40–60%.
Table 1. Typical ferrite targets for common duplex stainless grades
|
Grade |
Base metal ferrite (solution annealed) |
Typical weld target |
Primary reference |
|
Lean duplex 2101 (1.4162) |
60–80% |
Project-specific |
ASTM A240 / A790 |
|
2205 duplex (S31803 / S32205, 1.4462) |
40–60% |
30–70 FN |
ASTM A790; NACE MR0175 / ISO 15156 |
|
2507 super duplex (S32750, 1.4410) |
40–55% |
30–70 FN |
ASTM A790; API TR 938-C |
|
255 super duplex (S32760) |
40–55% |
30–70 FN |
NACE MR0175 / ISO 15156 |
Note: a WRC-1992 prediction from weld chemistry is only a screening tool. The measured FN per AWS A4.2M is the contractual acceptance value when a specification sets one.
What Happens If Ferrite Is Too Low or Too High?
Too little ferrite (below roughly 30%) erodes the duplex grade's strength and chloride stress-corrosion-cracking advantage; too much ferrite (above roughly 70%) reduces toughness and ductility and accelerates formation of brittle intermetallic phases such as sigma in the heat-affected zone.
Low ferrite (over-austenitic microstructure). The steel behaves more like a standard austenitic grade. Its yield strength - the main mechanical advantage of duplex, roughly double that of 304/316 - drops. Chloride stress corrosion cracking resistance falls, and the benefit for sour-service and pressure-vessel design is lost. Welds with very low ferrite also risk solidification cracking if solidification switches to primary austenite mode.
High ferrite (over-ferritic microstructure). Excess ferrite reduces room-temperature impact toughness and ductility. More importantly, the ferrite phase is where chromium nitride (Cr2N) precipitation and sigma-phase formation occur during slow cooling or multi-pass welding. Sigma phase makes the steel brittle and sharply reduces pitting corrosion resistance. This is why duplex welding procedures strictly control heat input and interpass temperature, and why post-weld solution annealing is often required for thick sections.
Which Standards Govern Ferrite Inspection Reports?
The measurement method is governed by AWS A4.2M / ISO 8249 for magnetic FN and by ASTM E562 / ISO 17781 for metallographic volume percent; acceptance limits come from the product specification (ASTM A790), the welding code, and sour-service requirements (NACE MR0175 / ISO 15156).
Table 2. Key standards referenced on duplex ferrite inspection reports
|
Standard |
Title (abbreviated) |
What it controls |
|
AWS A4.2M:2020 |
Calibrating magnetic instruments to measure delta ferrite in austenitic and duplex weld metal |
FN measurement method and instrument calibration |
|
ISO 8249:2018 |
Determination of Ferrite Number (FN) in austenitic and duplex Cr-Ni stainless weld metals |
International equivalent of AWS A4.2 |
|
ASTM E562 |
Determining volume fraction by systematic manual point count |
Metallographic % ferrite measurement |
|
ISO 17781:2017 |
Quality control of microstructure of duplex stainless steels for petroleum and petrochemical industries |
Phase-balance QC in oil and gas service |
|
ASTM A790 |
Seamless and welded ferritic/austenitic stainless steel pipe |
Product specification (typically 35–65% ferrite) |
|
NACE MR0175 / ISO 15156 |
Materials for use in H2S-containing environments in oil and gas production |
Sour-service ferrite limits (typically 35–65% wrought, ~30–60% weld) |
|
API TR 938-C |
Use of duplex stainless steels in oil and gas production |
Industry guidance on ferrite windows by zone |
|
WRC-1992 Diagram |
Constitution diagram for stainless steel weld metals |
Predictive FN from weld chemistry (Creq / Nieq) |
How Should You Actually Read a Duplex Inspection Report?
Read the report in this order: (1) confirm the unit; (2) confirm the method and standard; (3) confirm the location and number of readings; (4) compare the reading to a criterion stated in the SAME unit; and (5) flag any result near the limit for metallographic confirmation.
A practical checklist for engineers, inspectors, and procurement staff:
- Find the unit first. Is the result reported as "FN" or as "% ferrite" / "vol% ferrite"? If the report only says "52" with no unit, call the laboratory. Never infer the unit.
- Check the method standard. A ferritescope result must cite AWS A4.2M (or ISO 8249) and a current calibration date. A metallographic result must cite ASTM E562 or ISO 17781.
- Check where it was measured. Weld metal, heat-affected zone (HAZ), and base metal can have very different ferrite levels. API TR 938-C gives different windows for each zone.
- Check how many readings. A single spot reading is not representative; codes usually require several readings averaged across the weld face and root.
- Compare in the same unit. If the purchase order says "40–60% ferrite" and the report says "52 FN," do not pass it automatically. Request a metallographic volume-percent result or a documented, grade-specific conversion.
- Watch the edges. Results within about 5% of the acceptance limit are candidates for referee metallography, because both methods carry measurement uncertainty of several percent relative.
- Check the certificate trail. The ferritescope calibration certificate, the welder qualification (WPQR), and the WPS ferrite target should all be internally consistent.

