Sunstealer,
Do you have a curve handy for strenght/temperature of steel?
(I know there are thousands of types, is there a common/average one)
Yes, sure, in the NIST NCSTAR 1-3 - they quote literature data too, which is something that any metallurgist would do. It gives correlation (or not!) between your own testing data and that which is available.
Chapter 7, page 104, Fig 7-1 (modulus), Fig 7-7 (normalised YTS), Fig 7-8 (normalised UTS). *normalised is not something that's intuitive so if anyone has any questions I (and I suspect lots of others) can answer regarding that.
http://www.fire.nist.gov/bfrlpubs/fire05/PDF/f05130.pdf
(Don't be swayed by the visual colours which enhances data points and tricks you into thinking that the average (or whatever) should be lower or higher than it is. Curve fitting is not done by eye now).
There is other stuff out there, but it's hard to find freely. Usually you find textbooks etc that will have data taken from available papers. eg: AISC - Steel Design Guide - Fire Resistance of Structural Steel Framing. (Disclaimer - not sure what type of steel this is - I haven't looked closer - so it's upto the reader to look at the source).
Table X.1 Steel Modulus of Elasticity and
Yield Strength Reduction at Elevated Temperatures
Steel Temperature
°F [°C] Em/E Fym/Fy
68 [20] 1.00 1.00
200 [93] 1.00 1.00
400 [204] 0.90 1.00
600 [316] 0.78 1.00
750 [399] 0.70 1.00
800 [427] 0.67 0.94
1,000 [538] 0.49 0.66
1,200 [649] 0.22 0.35
1,400 [760] 0.11 0.16
1,600 [871] 0.07 0.07
1,800 [982] 0.05 0.04
2,000 [1,090] 0.02 0.02
2,200 [1,200] 0.00 0.00
You'll also see the picture of the wooden beam with steel girders drooped over it often accompanied by a "strength/temperature graph" - bear in mind that modulus, YTS and UTS are all affected by temp.
And no, I'm not going to make my employer's comprehensive data on steel properties v temp available! (actually it wouldn't make any difference - a) we don't use plain carbon steels at any temperature, b) truthers don't like data/graphs/maths/science etc.
Is there a common/average one? Good question. The biggest problem with materials testing data is there are mountains and mountains of the stuff, but very little access to it, let alone public access.
The reason for this is quite simple: Data is generated by private companies, which means that private companies pay huge sums to generate that data through test programmes. These companies are often rivals so why would they give their expensive data away for free which would remove their technical edge? So average or common ones are taken from general text books etc,
who quote their sources. That's important, because not only is it peer reviewed (truthers look away, that's a dirty word), but if fortunate you can actually get the original "hard data" - the actual measurement. This helps make a judgement on the quality of the data and therefore you are able to "weight" it against certain criteria. This can form part of the data set that you finally use to ascertain your own material property data, but usually it's used as per NIST have done, i.e, include it on the graph, because it's all good reference material and helps the engineer view the overall picture and compare where his own generated data lies.
Papers and so forth, on different topics, used to be hoarded by individuals from university departments as part of a reference framework and through them (as an undergraduate/post-grad etc/affiliate/sponsor/or partner company) you could get access. Nowadays people/companies/universities etc sign up to web based sites to get access to research. Data is expensive.
I'm quite interested in those graphs - sadly I've had to go through the dirge of producing similar material myself aswell as integrating it into our company's database. It isn't a straight forward task and requires the use of statistical techniques as well as using experience and knowledge to "weight" data. (I can trace some things back to the early 50's! - it's all valid data, but how do you incorporate it into a database?).
