Published 20 August 2026.
How this came up
We published a page for 441 on 19 August 2026 carrying a proof strength of 250 MPa, taken from the Outokumpu Core range datasheet, cold rolled row. The figure looked high — higher than 304 in the same table, which is not what a ferritic usually does — so it went on the review list as a suspected discrepancy between the datasheet and the EN 10088-2 minima it cites.
It was not a discrepancy. A distributor datasheet for the same grade and the same product form gives both numbers in one row:
| Value | Orientation | Product form | Source |
|---|---|---|---|
| 230 MPa | longitudinal | cold rolled, up to 8 mm, annealed | thyssenkrupp Materials, E3 |
| 250 MPa | transverse | cold rolled, up to 8 mm, annealed | thyssenkrupp Materials, E3 |
| 250 MPa | not stated | cold rolled coil and sheet | Outokumpu Core range datasheet, E3 |
Sources: thyssenkrupp Materials (UK), 441/1.4509 and Outokumpu Core range datasheet, May 2015, both retrieved 20 August 2026.
The reading that fits is that the Outokumpu figure is the transverse one. That is an inference. Neither document says so, and the 441 page carries it as a note rather than as a condition, with the gap named explicitly: test orientation for the 250 MPa figure — the datasheet that prints it does not state one.
Why the steel cares which way you cut it
Rolling does not just make sheet thinner. It aligns grains and it aligns the crystallographic orientations inside them, so a rolled sheet is not the same material in every direction — it has a texture, and texture makes yield strength directional.
Ferritics carry more of it than austenitics for a structural reason. The body-centred cubic lattice of a ferritic has fewer slip systems available and develops a stronger rolling texture, and the stabilising elements in 1.4509 — niobium and titanium — pin the grain structure that texture lives in rather than letting recrystallisation wash it out. Face-centred cubic austenitics like 304 and 316L recrystallise more completely and end up closer to isotropic, which is why an eight per cent spread of this kind does not show up on their pages.
Eight per cent is not a rounding error. On a section sized to yield, it is the difference between a design margin you have and one you believe you have.
The fourth coordinate
This site already argues that a property value is not a property of a material but a property of the material as described by a document. The standard is one coordinate. Product form is a second. Specified minimum against typical value is a third.
Test orientation is a fourth, and it is the least often printed of the four. A yield strength quoted without its standard looks incomplete to a careful reader. A yield strength quoted without its orientation looks finished, and that is what makes it worse.
The practical rule is narrow enough to be usable: on rolled ferritic and duplex flat product, ask which way the specimen was cut before you use a proof strength. On austenitic flat product the question is worth asking and rarely changes the answer.
What we changed
The 441 data file now carries both figures as separate rows, each with its orientation in the condition column, and a fifth gap that says the orientation of the higher number is not established by any source we hold. Under the E-ladder both sit at E3 — real values from resolvable supplier documents, with no method, no specimen detail and no traceable chain behind them.
We did not average them. Averaging two numbers that differ for a knowable reason is how the reason gets lost.
What this note does not cover
- Whether the 250 MPa figure really is the transverse one. Two documents, one number each way, neither states the orientation of the third. The reading is ours.
- How large the effect is on other grades. We checked one. The mechanism generalises; the magnitude does not, and we have not measured it.
- Which orientation a standard requires for acceptance testing. EN 10088-2 is paywalled and was not opened, so we cannot say what it specifies, only what two datasheets print.
- Through-thickness properties, which are a different problem again and are worse.
- What your mill certificate says. It will name an orientation, and for your actual heat it supersedes everything on this page.