Definition of Steel Young Modulus 207Gpa 8

[antidio]

Dear all,

Could somebody tell me how was calculated/defined the "standard' value of 207Gpa as Yong Modulus of the steel.
In other words, if I have the stress strain curve as a results of a tensile test (ASTM A370) on a steel sample, then:
1) which part of the curve do I have to consider to calculate the YM. (I mean between which strain limits)
2) which mathematical method do I have to use? (linear regression, chord, etc)

what was historically done to get the value 207 Gpa?

http://www.eng-tips.com/

thanks a lot.

 

[MagBen]

1.linear portion, i.e. from origin to the yield point.
2. since it is linear, you can calculate YM simply by dividing stress by strain. practically, both engineering and true stress-strain curves for a steel could give you a similar result, however, in theory, you should use engineering curve.

 

[hokie66]

207 is a strange number. Most of us use 200.

 

[antidio]

Thanks MagBen,

what do you exactly mean for yield point? 0.5% strain? (absolute strain)

Analyzing more than 400 tensile test (ASTM A370) with this method I get an average YM of about 190Gpa with a stdev of 15Gpa. Pretty below the theoretical value of 207Gpa. How is it possible?

 

[metengr]

Depending on the reference, the value at ambient temperature reported for steel is typically 200 GPa. Your average valve and SD seem reasonable.

 

[MagBen]

207 GPa is not a theoretical value. while lots of steels have a YM of 200GPa, this value varies depending on what specific steel you are referring to. If you got 190 Gpa, the material could possbily be a stainless steel.

Since YM is related to elastic deformation. For the linear portion, the best regression is naturally linear regression, otherwise it is against the defination of YM. To roughly estimate the YM, one can just calculate one point at 50% yield. If you want to play with statistics, whle still have a low standard devaition, I would recommend using the portion of 20-70% of curve. The closer to yiled, the more linearity the curve could be. In addition, to evaluate the calculation using regression from a statistics viewpoint, S square and P value are more important than SD.

If the accuracy of YM is important to you, donot use an existing stress-strain curve, you may want to desgin an experiment taking the followings into account:
1. use a suitable stain rate
2. increase the length of the specimen at gage section.
3. load 50-60% of yield, unload to 20-30%...repeat couple of times
4. you may want to focus on loads at your interest


p.s. you can use .2% strain as the yiled point,especially when it is not easily define from the curve.

 

[MagBen]

p.p.s. Carpenter stainless customer 630 has a YM of 196Gpa, Carpenter stainless Type 420F has a YM of 207Gpa.

 

[CoryPad]

Young's Modulus does vary with preferred orientation in the material - what is your product form?

Young's Modulus determination requires very accurate strain measurement using a modern extensometer and an accurate force cell. Some force cells have poor accuracy near zero, so you may need to avoid that region. MagBen's suggestion of 20% to 70% of the yield point is a good one.

Section 13.2.1 of ASTM A370 specifically lists 207 GPa for carbon steels. For products like springs, the Young's Modulus is very important. The standard EN 10270-2 for spring wire lists E = 206 GPa.

 

[racookpe1978]

If your application is at anything higher than "room temperature" and if your application does require a very specific value for the Youngs Modulus, you need to either get a stress-strain curve for that alloy at that temperature, or repeat the Youngs modulus test at that temperature

 

[FennLane]

Young's Modulus of 207GPa is a good typical value for a medium carbon steel.

Pure iron would be slightly higher and alloy steels slightly lower.

The chances of measuring this value with any degree of certainty using a tensile test machine are slim to none.

The basic definition of Young's Modulus E = Ơ/Ɛ implies the use of True Stress and Strain as Engineering Stress is more correctly defined as S and Engineering Strain is e.

I would agree ate very low values of Strain there is little difference between the two values but True Stress needs to be accurately determined and the use of an engineering value will produce an error.

The next problem with using tensile test machines will be due to the Class of Extensometer. A typical extensometer that could be used fro the determination of proof stress just isn't good enough. The type of extensometer needed would be a BS 3846 Grade D and these types of device don't maintain their calibration accuracy for a few days (at least in my experience) Averaging extensometers may help but the class is still important.

The next issue would be machine and test piece alignment. Tensile testers are just not well aligned.

To determine Modulus with any degree of accuracy you would need to achieve levels of alignment around 10 times better that is typical and much more in keeping with machines used for LCF tests.

It would be possible to align a test piece but some high quality specimen holders which allow adjustment.

If you did adjust alignment and used strain gauges attached to the test piece you may get somewhere but the best method to accurately determine the modulus of steels use either resonance or ultrasonic measurement.

I would also question why it is a property that needs to be measured as the difference in modulus between to steels of similar composition will be virtually impossible to detect and are unlikely to change much apart from the influence of temperature (10% per 100 degC - approximately)

All of the tensile tester we used to sell had a check during the elastic loading portion of the tensile test.

If the modulus was not within 5% of the expected value we used to unload the machine prior to yield as it was likely there was a problem with the extensometer.

The idea of using Modulus measurement of part of routine QC just doesn't make sense for the majority of steels.

 

[EdStainless]

Unless you need values at a specific load or temp then using acoustic measurements is the easiest.
In the wire business we did exactly what Ben described.
Long samples, repeated load and unload.
It is important in wire because when you heavily cold work material you change the modulus.
We were drawing high strength steel wire (1.2%C) and we had a strength and modulus range to hit.

I have seen acoustic measurements done using a cell phone with a frequency analyzer app on it.
They were getting values within 3% that way. Using a precision microphone and analyzer you can get <0.5% error.
Steels are nice because the modulus is the same in all directions (unless heavily cold worked).

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

 

[MagBen]

FennLane, thanks for correction. By definition, the true stress-strain curve, instead of engineering one, should be used. I meant to true curve, but my fingers failed to follow my brain.

 

[antidio]

Dear all,

Thanks a lot for your valuable contribution.

Just some clarification because I understand that I was not very self-explanatory:

1) I am talking about C-Mn steel, Grade X65, the one used for pipeline construction (API 5L or similar)
2) The YM that I refer to was calculated cutting samples in both longitudinal and transversal position. To be honest I didn't recognize a significant difference between the two directions (Long and Transv).
3) Tensile tests were conducted with "standard" extensometers. Tests were performed at room temperature.
4) YM was generally calculated considering the range 10%-50% SMYS and using engineering values.
5) My impression was that the "test procedure" and "test equipment" were not suitable to calculate with a good grade of accuracy the YM
6) Most of the time the operator didn't look to the Stress Strain curve (during test execution) to understand if the test was progressing well or not

We performed similar test on compressive direction, following the ASTM E9 guideline and the test results were 10 times more accurate using the same "standard extensometers", the same calculation method and so on. Therefore I think that 90% of the reason of the reason of the higher spread in the YM (following tensile results) is depending to the test specimen machining and alignment. As normal we take great care for the machining and test piece alignment during compressive test but not during tensile.

Thanks again

 

[antidio]

Just one more detail about the sample direction and its influence on the YM:

samples were machined form fabricated pipes in longitudinal and transversal direction.
Pipe was SAW, it means that:
1)we start with a laminated plate (defined direction of lamination=longitudinal respect to pipe axis. Plate was obtained via TMCP from a slab even 10 times thicker than the plate.
2)then we cold worked up the plate to a pipe with UOE or JCOE method (cold expansion ratio max 2% for no more than 3 times.

I didn't noticed significant differences in the YM in long or transversal direction even if based on my limited knowledge if you had significant cold working in one direction the YM in this direction should be slightly higher.

 

[EdStainless]

In a steel, unless it is heavily cold worked, you should not see a diff in Modulus based on direction. The tensile properties maybe be very different, but not the modulus.

A good tensile machine with self aligning (gimble mounted) jaws and good extensometers will actually yield good data. But you must use long samples, with long range extensometers, or optical ones.
The load and un-load cycles are important to take any 'slack' out of the system.
Temp control is critical, for measurement of load and extension as well as the actual properties.

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

 

[MagBen]

Ed's statement "tensile maybe different, while Mudulus does not, based on testing directions" appears inconsistent. The inconsistency maybe easily reconciled by noting the emphasis of load and un-load for YM measurement process. Taking away the mechanical "slack" would help the material behavior more isotropically. Remember the tensile strength could be manipulated by changing the strain rate (e.g. the tenisle could increase with increasing strain rate).

I am imagining if the Mudulus could vary a bit at different directions when the sample is a single crystal?

 

[metengr]

http://www.nist.gov/mml/acmd/struct...Steel-elasticity-variation-study-21012_v2.pdf

 

[EdStainless]

In Ti (CP) you have material with a hex crystal structure that develops a texture when rolled.
So Then it gets a final stress relief (CP Ti is not annealed).
The modulus in the three directions in plate or sheet are significantly different. The average is about 15.5kkpsi. The low is about 12.5, and the high is about 17.
All the while the difference in T and L tensile strength can be +/-50%.

There is no dependance of Modulus on strength, they are independent parameters.

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

 

[MagBen]

Not trying to beat to death, but for a specific material, could it exist a tendency that the higher the yield, the higher the mudulus? If assuming no plastic deformation between zero load and 100% load of yield, the modulus must be proportional to yield strength, i.e. modulus = yield/.2% (perfect linear). The assumption fails, because you can not use the yield point to calculate modulus, the tendency may still exist: the higher the yield, the higher the slopes for most of loads, and so the higher the modulus. The change ratio of yields seems to be higher than that of modulus when the status of mateiral is changed, so even people see a big change in yield, they only see a small change in modulus.

Not to pretend to be an expert, myself actually doesnot have first hand experience in measuring modulus, it is to me a learning expereince! I really appreciated and enjoyed reading different opinion from all different perspectives.

 

[EdStainless]

With stainless alloy steels low amounts of cold work (<25%) don't change the modulus a detectable amount.
In a SS this level of cold work could raise the tensile 50% and double the yield strength.
At high amounts of cold work (>50%) the metals have no distinct yield as they have only a few % elongation at failure. These cases can have distinctly elevated Modulus, though the way that you do stress relief is important also. You don't want to be measuring residual stress and calling something else.

You can't use the engineering yield point for modulus calculation.
Once the load/strain line deviates from straight you are probably at 0.1% or less elongation.
With a 10" sample this would be 0.010" total elong.
Our optical extensometers would measure 0.0005" change on a 10" sample.
Going clear to 0.2% is almost always way beyond true elastic behavior.

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