What are the differences between Charpy and Izod impact tests?

[snowJT]

Is the only difference between these two impact tests how the specimen is held in position for the test? Charpy holds it like a simply supported beam, and Izod does it as if it was a cantilever?

 

[Gokul43201]

Almost.

With a standard, notched sample, which surface would be notched for the two tests?

 

[snowJT]

that would be the charpy, I've only done charpy impact tests.. so I was wondering if that was really the only difference...

in that case, wouldn't charpy be more of a better method to measure impact because without the notch, the material could break in different ways, and with only one support some materials will tear more before it breaks?

 

[Gokul43201]

To my knowledge, they both use notched samples (I haven't looked up the ASTM specs). With the Izod test, the notch is on the upper (impacted) surface, while with Charpy, the notch is on the lower (non-impacted) surface.

I would wait for someone else here to second this - I'm repeating something I was taught a decade ago, and things may have changed.

 

[snowJT]

no, both of them do have a notch, either a v-notch, or a keyhole notch, I found a book that talks about it... but I can't find things that compare them against each other, I haven't used or seen an izod, so I'd like to know if there's disadvantages to using it.

 

[Gokul43201]

We had a prehistoric Izod pendulum in my old department, and a much newer Charpy tester. I never actually used them much, except for one class where we had to find the impact strength of something from the Izod test (I guess it's obvious why they would let us inept students handle that one).

I would wait for Astronuc, Perennial, Fred or Q_G to chip in. Some of them may actually have reasonable experience with these beasts.

 

[Astronuc]

I haven't done much in the way of hands on with Charpy or Izod. Back in unversity we did some testing as part of class, but I don't really remember the details.

As I recall the stress states are different in the vicinty of the V-notch, and IIRC, there is more shear (in the plane of the centered through the V-notch) in the Izod test as opposed to normal stresses compression/tension. The results are consequently different, and that relationship will vary according to the material (i.e. depends on elastic and the shear modulus).

Also where the impact occurs there will be an acoustic shock traveling throught the specimen that will effect the fracture somewhat. I have seen such discussions, but that was in some text which I don't have - and it was many years ago.

I think there have been comparisons with computational methods, and probably comparative experiments, but at the moment I can remember any sources. Probably there are articles in materials journals where comparisons have been made.

Here - http://nvl.nist.gov/pub/nistpubs/jres/104/6/j46sie.pdf
Compare corresponding materials in Table 1 and 2, or look at the results in Table 3. See also section 3.5 Charpy vs Izod

Here is a nice comparison of the two methods.
http://civilx.unm.edu/laboratories_ss/mechmat/charpy.html

It's best to get a copy of ASTM E23.

Here is one comment -

HOWEVER, the numbers from those two test methods are not comparable. There is no accepted method of comparing CVN and IZOD numbers, but it is commonly known that the IZOD test produces higher numbers than a CVN test on an identical material sample, ie, CVN is more conservative.

This statement would indicate more resistance to fracture using the Izod test.
http://epi-eng.com/RW-EPI-ShaftTuff.htm

Nice background - http://www.matsci.ucdavis.edu/MatSciLT/EMS-174L/Files/DuctileBrittleTransition.pdf

Of course, as a modeler, I have to question whether the test actually replicates a real loading as would be encountered in the actual environment of the material.

Many tests are useful for certification/acceptance, but not for actual proof of porformance in the real environment.

 

[PerennialII]

There are lots of correlations available which enable usage of impact energy / ductility based measures of toughness in a quantified analysis (basically correlate for example CVN transition temperature to fracture toughness transition temperature, CVN upper shelf energy to J-integral based fracture toughness and so on), which is the way to attain consistent quantitative material properties out of these tests (with some scatter of course, and different correlations work better than others, some just don't at all).

There are also a number of correlations build between different types of impact tests (CVN, CUN, the keyhole variant, Pellini, DWTT, DT tests,...), but for some reason Izod at least usually isn't mentioned at all. The understanding I've is that people involved with fracture of "traditional metallic materials" have usually used CVN and Izod in principle doesn't bring anything new to the table ... so it's not used. Could check ASTM E23, both are contained within the same standard but haven't ever been interested in the Izod part.

We're doing lots of modeling work trying to improve the existing correlations from CVN T27/28J to fracture toughness T0, really cool simulations computing the dynamic fracture of the CVN specimen when it's "going all over the place" and trying to figure out what it really measures.

 

[Astronuc]

Just adding to what PerennialII mentioned, I was thinking about this last night, and I recalled that most fracture toughness measurements these days seem to use compact test (CT) specimens.

One issue with Charpy or Izod is the mixed mode in the cracking region - and I believe Izod introduces a slightly higher Mode II component.

See - http://www.ndt-ed.org/EducationResources/CommunityCollege/Materials/Mechanical/FractureToughness.htm - which gives a good introduction and overview. Toward the bottom of the page is an important consideration - that of orientation of the crack and specimen. Some materials have 'texture' and are more anisotropic than others.

The issue we've had with tests of fracture toughness is how close the stress field in the test matches the stress field for in-service loading.

With respect to an impact load, the impacting object would put the immediate volume under compression (under contact area) and shear and tension to the side. Ahead of the compressive field, the material would be in tension.

 

[PerennialII]

...

The issue we've had with tests of fracture toughness is how close the stress field in the test matches the stress field for in-service loading.

...

Yeah, 1st it's the C(T)s or SEN(B)s for lower bound values (or if only have indirect data correlations), nowadays quite often followed by a constraint analysis (the number of these has increased quite rapidly during the last few years), analysis of biaxiality & membrane/bending effects, coupling to what fracture type really to expect (changes in "fracture appearance"), residual stresses ... the list is often pretty long really. Especially if want to / need to aim for 'best estimates'.

 

[rdt2]

http://www.boulder.nist.gov/div853/Charpy website/Documents/CVN history.doc has a good summary of the development of impact testing.

Another good article is http://www.instron.co.uk/wa/products/impact/charpy_izod.aspx?ref=http://www.google.co.uk/search

IIRC, Izod and Charpy chose test configurations that reproduced common failure modes. That would explain why Izod, who apparently wanted to develop better tool steels for metal cutting, chose a short cantilevered beam.