Requirements for Impact testing A36 or A992 at -40¦C ?

[MrStohler]

In the area near Edmonton Alberta:
* We are designing a non-building steel structure that could be exposed to temperatures as low as -40°C.
* It is not subject to dynamic loading (fatigue is not a design concern)
* It is not in a high seismic region.

The materials:
* A992 for the wide flange shapes
* A36 for angle braces (it is braced frame),
* A36 for plates and other connection material.

There appears to be no specific requirement to perform notch toughness tests on the material, since it is specified to meet ASTM A36 or A992 not type W and not type WT.

The draft version of the material purchasing specification we have written does not require notch toughness tests.

Is there a code requirement to conduct the test under these conditions that we have missed?

The more interesting question that has been presented regarding the specification is: "What is the reasoning behind NOT requiring the tests? Please elaborate why these conditions would not require this type of test"

 

[metengr]

You need to check with the local building codes.

 

[metengr]

Just to add to your last question - A36 can be purchased to impact requirements IF the purchaser (you) specified impact test testing requirements. See Supplementary Requirement S5.

If the local building codes provide no additional information, I would recommend at -40 deg C, a minimum of 20J for CVN impact energy at this temperature just to have some margin of notch impact resistance in the fabricated structure.

 

[MrStohler]

Thank you

The Alberta Building Code and the Handbook of Steel Construction CAN/CSA-S16-01 by CISA/ICCA are the governing design codes.

Both of these appear to be silent with respect to the issue above. This is not to say they are silent on notch toughness, it is required for Type WT, Type AT and Type QT steel (pg 6-3 of CAN-S16) BUT A992 and A36 do not appear to fit these conditions.

With regard to the recommendation of a notch toughness of 20J, would you kind enough to point out the source document that would allow me do do more research.

 

[metengr]

MrStohler;
The 20J CVN impact energy that I suggested above is based on what I would use (engineering judgement) provided you have no construction code application or local building code design requirements. I believe in what you stated above you have a design code with suggested impact requirements and material requirements, and I quote;

The Alberta Building Code and the Handbook of Steel Construction CAN/CSA-S16-01 by CISA/ICCA are the governing design codes.

Both of these appear to be silent with respect to the issue above. This is not to say they are silent on notch toughness, it is required for Type WT, Type AT and Type QT steel (pg 6-3 of CAN-S16) BUT A992 and A36 do not appear to fit these conditions.

My question back to you is this based on the above information, did you select the correct material for this design?

 

[MrStohler]

I have no qualms about the material. We have used this same material in more severe conditions in the US. But the governing codes did not require notch toughness testing.

Let me further clarify the code requirements for notch toughness: The code does not explicitly require testing for Type W steel. It has been our firm's contention that A36 and A992 correlate with this category and not the WT category.

This project is somewhat unusual in that although is is in Canada the owner is not from North America. The owner is does not believe that the code's silence on this issue is sufficient to conclude it is not required.

I am trying to dig into the since behind the requirments for the testing.

I greatly appreciate your input. Thank you.

 

[metengr]

MrStohler;
I can give you my perspective based on past experience in working with code and standards in the US. The science behind impact testing requirements for codes and standards organizations is to assure minimal risk of brittle fracture after fabrication. Many code and standards organizations provide either specific requirements for a minimum CVN impact value at a designated service temperature or provide exemption curves so that CVN impact testing does not have to be repeated because the material specification requirements will assure that same material will meet the minimum CVN impact value.

So, as I stated above, A36 and A992 are not demonstrated 'low temperature' service steels unless supplementary CVN impact testing is invoked by the Purchaser to assure that the supplier of the steel can meet code and standards specified CVN impact requirements.

 

[MrStohler]

metengr

Thank you, I have a follow up question based on your reply.

..unless supplementary CVN impact testing is invoked by the Purchaser to assure that the supplier of the steel can meet code and standards specified...

Could you provide a specific code references?
Which code and section would lead to this testing being required?

 

[metengr]

I don't have copy of the Handbook of Steel Construction CAN/CSA-S16-01 by CISA/ICCA to evaluate minimum CVN requirements. As far as an example of codes and standards that follow what I mentioned above regarding rules for CVN impact testing is the ASME Boiler and Pressure Vessel Code (Section VIII, unfired pressure vessels for low temperature service).

 

[greentank]

MrStohler,
Just out of curiosity on my part, how did you come to the conclusion that A36 was good for -40C? Everyone I have gone to for materials selection advice has said that A36/44W (which I'm also being told is now the same thing, hence A36 is classified as grade W) is no good for below
-29C.. I currently have a project for Fort Mac and am in the process of material selection for a design temp of
-45C.. Your explanation would be greatly appreciated.

 

[MrStohler]

It depends on how the material will be used.

In this instance it is being used in a structural application that is not subject to fatigue nor is it subject to high inelastic demands for seismic events. The governing structural steel codes (see post above) do not prohibit it's use and A36 has been used under many similar (or worse) conditions in the US.

It could be how one views the codes. Should one only do that which the code explicitly allows or merely avoid that which it explicitly prohibits. The 2 approaches would yield very different codes. I subscribe to the later point of view.

It is my understanding that there are piping and pressure vessel codes that do prohibit it's use at -40°C. What conditions led these codes to prohibit A36 under these circumstances? Is it a desired factor of safety, high inelastic demand, fatigue loading?

 

[metengr]

MrStohler;
In a nutshell, one of the reasons that A36 is not recommended for "low temperature service" is the inherent risk of brittle fracture because of the low toughness of this material (below -20 deg C) supplied under ASTM/ASME A/SA 36 with no impact requirements.

By the way, you don't need dynamic loads to induce brittle fracture in a structure. All you need is low toughness material coupled with significant residual tensile stresses in the presence of a flaw or crack(s) = a sudden structural failure.

 

[MrStohler]

Please advise, A36 is not recommended by whom?

 

[metengr]

ASME Boiler and Pressure Vessel Code (B&PVC). Recognizing that the ASME B&PV Code code does not directly apply to your fabricated structure, one can still use this code as a guide for material behavior related to service conditions.

Section II, Part D of the ASME B&PV Code limits this material to -20 deg F with no impact testing.