You have to start from the notion that the "strength" is not a scalar value, but a relation between strain (creep) rate ep and stress s at a given temperature and microstructure for a given material. Consequently you get the best overview of strength in the (double or semi log) field of ep versus s at given conditions. The yield stress is normally measured at rates in the order of 1e-4/s. So you get only a single point (ep,s) from a single value of the yield stress. The creep strength usually refers to the minimal creep rate ep at a given stress s. So you get a series of (ep,s)-data for the state of maximum deformation resistance. It is generally located far away from the single point for the yield stress at high rate. As creep is plastic deformation at stresses, that by definition of plasticity must be larger than the yield stress at the creep rate, you get an idea of the field of plasticity, bounded by the two lines for the yield stress as function of ep and for the maximal strength (minimal creep rate as function of s). There is no guarantee that a material that is stronger with regard to yielding is also stronger with regard to maximal strength, but it is well possible and not unusual.

Thus, in short: plot the existing data for a given temperature T in the ep-s-field. This gives you the overview. 

large yield stress(ep)-    

yield strength is to mark the 

You could make an account to access the NIMS creep data which might have the data.http://smds.nims.go.jp/creep/index_en.html  

I believe when they talk about yield stress and fracture stress they talk about uniaxial quasistatic results.

Creep fracture stress is the stress that will lead to fracture after a given time of constant loading with three phases of strain accumulation: primary (logarithmic with time), secondaty (linear slope) and tertiary (exponential) leading to necking and rupture.

Is your load maintained with time? Then you should use creep rupture values. If you quickly unload the specimen at high temeperature then you should think of quasistatic values.

You can find some tests on my PhD Thesis http://meaagg.com/UBPHD/UBPHD.html

http://meaagg.com/UBPHD/UBPHD.html

Thanks for all your responses.

Let me inform that material data available in our organisation is in tabulated format similar to ASME BPVC section II. I am having 1e5 Hours creep rupture data at an interval of 100 deg C and similar data for yield strength.

@Andres...you pointed out correctly if load is maintained with time i should use creep rupture values and if  load is not constant with time one has to design for fatigue and take care of creep fatigue interaction...

but in this whole process where high yield strength at operating temperature is helping me.....

Till now few things are coming my mind....

first if fatigue is due to transient thermal stresses higher yield strength will shift the LCF curve in favorable direction.....second there are less chances of ratcheting...and third thermal shock resistance parameter will improve....and last better margin for short term excursions...

Am i thinking in a correct direction....

KINDLY SEE THE PAPERS BELOW: For plotting creep data against stress normalized with eitjher YS or tensile strength (sorry just I have written in spoken Englsih)

(i) William Harrison, Mark Whittaker, and Steve Williams; Materials 2013, 6, 1118-1137; doi:10.3390/ma6031118.

(ii) B.Wilshire and P. J. Scharning:Mater. Sci. Technol., 2008, 24, (1), 1–9.

(iii) B. Wilshire and P. J. Scharning: Int.Mater. Rev., 2008, 53, (2), 91–104.

(iv) M. EVANS, Mater. Sci. Technol., 2010, 26, (3), 309–317.

1) The Yield strength at any temperature is obtained at strain rates say 1e-4 s-1 by conducting tensile tests. But in the creep range, the creep strain rate varies and the minimum creep rates are much much lower say 1e-9 s-1 or 1e-10.  Let us take 304 SS: at 700 deg C, plot (log-log) the instantaneous strain vs stress plot (constant load creep tests at various stresses), then take the stress for 0.2% strain: I have observed this as around 80 MPa which is much lower than the YS (i.e., 250 MPa at 700 C) obtained by Uni-axial tensile tests.

2) It depends on the material: (a) for 304 or 316 SS: Yield strength (obtained by tensile test) is higher, then creep or creep - rupture strength also will be higher. (b) but for some other material, I suppose for Nimonic: it may not be so. (c) lets us give prior cold work and ambient temperature, then for 304 SS: creep rate (say at 700 C) will decrease with increasing prior cold work, but for Nimonic: it will be opposite. But for both cases: creep ductility will decrease with increasing prior cold work.

3) Hence one need to plot minimum creep rate with stress normalized with either YS or with ultimate tensile strength. Kindly see the papers given at the beginning.

Best wishes.

PLEASE SEE THE BOOK:  Steel Forging: E.G. Nisbett/A.S. Melilli (Editors) ASTM STP 903 (1986). I quote form the book: page 14.

In addition to my earlier answer, I further add these comments from the above mentioned STP book (but it would be better to consult someone working in EPRI).

Since it is a turbine rotor and is by forging: initially the Cr-Mo-V steels needed high temperature ductility and lower-temperature fracture toughness. To deal with larger ratings, larger forgings with hither tensile strength was required and Ni-Mo-V alloy was introduced. Even though the fracture toughness is important, most of the design acceptance criterion were based on yield and tensile strength criterion. 

@ Phaniraj C...

Sir thanks for suggesting the book....i saw it on google book....seems good...will ask our library to buy it.....

I am still searching some answers from mechanical design point of view....

Hi Naveen,

I agree with comments of Prof. Blum that at high temperature the "yield stress" value is not representative. Indeed, any stress value like 400 MPa should be related to the inelastic strain rate at current state of hardening, softening etc. However during start up or shut down regimes you may have stress peaks at lower temperatures, for example 300°C. The the yield stress is then more useful property: for stress peaks lower than the yield point you may expect less ratcheting, prolonged LCF life,.. etc.

Hi Naveen,

Why is a yield strength of 400 MPa at operating temperature mentioned as a material selection criterion for turbine rotor ? Normally at operating temperatures in the creep regime, the main phenomenon at origin of defects would be creep, so the yield strength should not be a material selection criterion. But, I agree with Konstantin that temperature excursions, low-cycle fatigue etc…. could have negative effects because the yield strength has been passed beyond.

Another aspect could be that using a low yield strength material would become redhibitory for the turbine (damage, breakdown,  …. ) for manufacturing reasons. For instance, a sufficiently high yield strength at operating temperatures could be necessary to warrant a safe fixing of impellers on rotor shafts. An heavy fit shrink is e.g. not possible with austenitic stainless steels because of their too low yield strength (high temperatures to be avoided). Also, the use of such materials imposes limits on the tip speed … And maybe other similar issues.

But, maybe these problems do not exist anymore in present time manufacturing or for the materials used nowadays ?

Kind regards

I would like to add something to the excellent replies that you get from the others. Creep affect materials even if the applied load is not over the yield producing damage. If you consider also fatigue in your scenario, the importance of the load level becomes crucial in identifying the most likely failure mechanism. For  a high level of stress, fatigue should be dominating ( larger total strain range), and the damage should develop from the surface. However, if the load applied is small and the temperature is very high the damage produced by fatigue will be progressively reduced. A good work, for temperature up to 565 on CrMoV has been done from Dr Holdsworth.

Dear sirs,

thank you very much for your replies.....

yield strength (YS) is a resistance against instantaneous failure (not time dependent) so definitely a higher value of it will provide a safety against load excursions...also, during transients like startup/shutdown for same peak stress value (elastic) total strain will be less in a high yield strenght material (assuming same young's modulus) and hence high fatigue (LCF) life....

@Guibert Crevecoeur...your reply made me to think in some other direction (dont know how much correct I am)...steam turbine designs (unlike pipes and pressure vessels) are governed by thermal design (targeted for highest efficiency). Blades/flowpath could be designed more compact (low length rotor) with high YS material and will result in high efficiency. In addition vibration aspects also require compactness of design....so a better design (higher efficiency and low vibration) may be possible with high YS material..

Regards

> Naveen,

Yes, indeed you might be right. Compactness of design could also inspire using higher yield strength material. There might be a combination of design and operational reasons going in the same direction.

However, to have the last word on it you should try to find out some design details (material of rotor, dimensions, turbine and rotor design, ...) corresponding to the cases where you encountered this requirement. Also, to which generation do (did) these turbines belong ? And, which are (were) the operating and maintenance policies of the customers of these turbines ? Were they foreseen to replace older units, corresponding to which technical specifications, etc... If this is possible of course and the information is available .... (?)

Regards

@Guibert Crevecoeur..

Dear sir,

This requirement is for next generation turbine (AUSC - operating temperature>700 deg C). In this link  (http://link.springer.com/article/10.1007/s11837-012-0241-4 ) it is written "A minimum 0.2% yield strength greater than 400 MPa in the radial direction (or transverse for bar stock) at 760°C is required for highly stressed regions, although higher strength would be desirable " Similar type of requirement is shown in slide 22 of this presentation (https://www.netl.doe.gov/File%20Library/Events/2010/24th%20Annual%20Conference%20on%20Fossil%20Energy%20Materials/FE-2010_Strength-Concepts_Hawk_NETL.pdf) 

Even if you go through the document "A review of the properties of 9-12 % Cr steels for use as HP/IP rotors in advanced steam turbines" (open access - check on google) on page no 11 it can be seen that when researchers tried to raise the temperature of steam turbines from 565 to 600 deg C the properties of proven materials at 565 deg C were targeted for the new rotor material at 600 deg C. In which 0.2% PS at 600 deg C >350MPa was also there. So it seems to me a minimum operating temperature yield strength requirement was always there for all generation of steam turbines.

But the thing is why so high number of 350 or 400 MPa YS against the creep rupture (1E5 hours) of 100MPa. Even if I consider at the support of blade, stresses are very high at some fillet and in case of a 20% overspeed the stresses due to centrifugal will increase by 1.44 times, still number look very high to me.

Regards 

Dear Naveen,

"So it seems to me a minimum operating temperature yield strength requirement was always there for all generation of steam turbines".

Yes, that's it ! It's an old habit of designers. And to be sure, they put a safety factor on the maximum stress. For instance, take 200 MPa as a "normal YS" for the material at the operating temperature. You mention stress increases by 1.44 times during operation. This brings us to 288 MPa. And now put a 1.25 or 1.4 safety factor to be sure. You get 360 MPa or 400 MPa.

I lived this several times in discussions about standardization or defining design rules (as a representative of a licensed agency or of a Ministry). There is always a respected fellow who has performed more calculations than the others and makes a nice presentation of the results. Then there is a discussion and he or somebody else proposes a safety factor of say 1.25 or 1.4 "to be sure". And all agree. And so a custom is created. And everybody is happy, because with the added experience, the numerous damage assessments, the unforeseen shut downs encountered by the users, etc... they all have acquired a deep feeling of what safety must be.

Then the manufacturers intervene : they think first of all in terms of gains, saving money, lower cost materials, adjusted dimensions, etc. and find after a while that the requirements in the standards and in the codes are exaggerated. As a consequence, there is sometimes a revision (of that point in the standard or the code), sometimes not. It depends on the risks and the arguments of all sides. Standards and codes are anyway periodically revised.

But I must say, with the steady technical improvement, the whole process often ends up with compromises. And these hold until the next revision. Where the participants to the discussions have acquired some more experience and the state-of-art has evolved.  

That's the game you are probably playing in !