Variation of mechanical properties for metals with specifications?

[thauptman]

I want to get an idea of how much variation there is in actual mechanical properties of metals (yield strengths most often) with specifications (6061-T6 or AISI 440, etc).

Is there an existing source for this type of information? For instance do metal mills compile and publish material test reports from all the lots going out the door into one big histogram?

I'm a mechanical design engineer, doing one-off and low volume mechanism design. Usually I need to get it right the first time, with no prototypes, and not much room (volume/weight constraints) for a big safety factor.

Presently I grab a reference book (or matweb) for a value. Because of the volumes I don't do, testing is not an option and the associated risks are accepted. But it would be nice to have a better understanding of where I could push the boundaries.

-Traveler

 

[swall]

There are spec minimum guaranteed properties and then there are "typical properties", which usually are higher.For design purposes, you need to stay with the guaranteed minimums. With no opportunities for extensive testing, you have no other alternatives.

 

[CoryPad]

MIL-HDBK-5 has various values that are derived statistically.

 

[israelkk]

According to MIL-H-6875H HEAT TREATMENT OF STEEL, PROCESS FOR para 3.3.2.1 says:

3.3.2.1 Permlsslble variations of Classes A and B steel from design ultlmate
strength. When a minimum acceptable strength level and no maximum strength
level Is speclfled by design or the applicable material speciflcatlon, the
maximum strength shail be 20 ksi above’ the mlnlmum. except for Hy-Tuf and H-11
steels for which a maximum strength of 30 ksl above the mlnlmum Is acceptable.
For 300 M steel, a maximum strength of 30 ksi above the mlnlmum is acceptable,
provided the maximum tensile strength does not exceed 305 ksl.

1.2 Classification. Steels covered by this specification are classified
into the followlng four classes. Unless otherwise specified, the process and
equipment requirements In this specification refer to all classes of steel
tabulated In Tables IA, IB, IC and ID, respectively.
Class A - Carbon and low alloy steel
Class B - Martensltic corrosion-resistant steel
Class C - Austenitlc corrosion-resistant steel
Class D - Prectpitatlon-hardening and maraging steel

 

[thauptman]

CoryPad, the MIL-HDBK-5 (Now MMPDS-01) is great. I didn't know about it and it gives me a better sense of the various properties and how much confidence I can put in them.

Too bad they didn't throw in the weibell parameters but it makes sense given it's purpose.

Cheers!

 

[EdStainless]

I understand your desire to take advantage of actual properties, but if your devices need to work then you have to stick with traditional safety factors.
One reason is that different mills use different processes and see different properties. One guy may always be 30ksi over the minimum, but another may consistently only clear it by 5ksi.
If you really feel the need to shave then buy the metal first and you will know what properties to design with.

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[thauptman]

Part of my motivation is to gain a better understanding of the materials rather than just shave things closer.

If I had a better idea of the potential spread, there are occasional situation where I might try to cherry pick based on the property. That's something I hadn't considered before, thanks.

All my work is for complex electro-mechanical mechanisms in an unregulated industry (University R&D) so I make up my own safety factors... Pulling from industries that make sense when I can.

 

[plasgears]

When you order the matl specify certified place of mfr.