Theoretical A333-6 vs A106B question. 1

[robsalv]

Folks,

Theoretical situation/question, which could be based on flawed interpretation of codes...

In a pressure vessel, if I want to use A106B just below its appropriate material curve, I need to do Charpy impacts and they need to meet a certain minimum value. No drama there.

To avoid the impacts, I could use A333-6 instead. Cool, no dramas there, that seems straightforward.

However, what has piqued my interest is that when I have a look at the ASTM spec for A333-6, it appears that the minimum impact value at -45degC is LOWER than what would have to be used to test A106B.

Does that mean that as supplied A333-6 is less safe than an impact tested A106b???

Comments?

Thanks in advance.

Rob

 

[metengr]

No. ASME SA-106 is a Standard Specification for Carbon Steel Pipe for High-Temperature Service, this specification does not require impact testing.

ASME SA-333 is a Standard Specification for Seamless and Welded Steel Pipe for Low Temperature Service, this specification requires impact testing. The specification requires 10 ft-lbs or 14J at -45 deg C (-50 deg F).

I am not sure about your theoretical question because SA 106 is not produced for low temperature service, thus you cannot compare with SA 106 Grade B with SA 333 Grade 6.

For a pressure vessel application, ASME Section VIII Div 1, Part UCS, impact exemption curves are listed in Figure UCS-66. For carbon steel material like SA-106 Grade B and SA-333 Grade 6, Curve A would apply for both material, which is based on component thickness and minimum design metal temperature.

Thus, I don’t see the validity of your question. If anything, SA-333 Grade 6 would be a more appropriate material for use at lower service temperatures in comparison to SA-106 Grade B material.

 

[metengr]

One minor correction is that both SA-106 Grade B and SA-333 Grade 6 material, being ASME/ASTM pipe specifications, would fall under Curve B, not Curve A, in Figure UCS-66.

 

[robsalv]

Metengr,

Value your contribution.

I wasn't querying the designated service focus of both A106 and A333. That's clearly understood.

I'm working on the assumption that A333-6 is a curve B steel due to fine grained practice. I'm happy to be shown to be wrong though!

If it is a curve B steel, then my posed question is valid.

Given A106B is curve A, if slightly below the exemption curve, then the code requires impacts to prove suitability. I can avoid the impacts by using A333-6 instead though.

However, that's where things get interesting. The minimum impact values that are required to be met for A106B (because it's below the curve) are HIGHER than the minimum "as supplied" impacts for A333-6 [14J per your post].

If the A106B passes the higher impacts, could you consider it safer than A333-6 for the same proposed application?

The other working assumption I have is that A106B and A333-6 are basically the same bit of steel, but one comes certified for low temps to -45DegC.

Interesting stuff.

Cheers

Rob

 

[robsalv]

I posted too soon... correction noted. Metengr is absolutley right, both steels are curve B... and I believe that proves my basic premise wrong...

I still have a dilemma though.

Practical experience has shown that for temperatures between -29DegC to -45degC, A106B pipe would require impacts but A333-6 doesn't since it's certified. But A106B would need to meet 27J minimums [per fig UG 84.1] whereas A333-6 is supplied with 14J minimums.

A106B, if it passed impacts, might actually be tougher than the "as supplied" A333-6...

Rob

 

[metengr]

No, SA 106 Grade B would not have a higher impact value than SA-333 Grade B because of General note (c) referenced in UG-84. If you refer to Figure UG-84.1 in Part UG of ASME Section VIII, Div 1, the curve applicable for SA-106 Grade B is the lower one because of the lower specified yield strength of 35 Ksi, which would require a minimum CVN of 15 ft-lbs (20.4 J). Thus, SA-333 Grade 6 material by default would easily meet this minimum requirement.

 

[weldtek]

Metengr,
I'm confused. Per Sec II, both 106 GrB and 333 Gr 6 are 35K yield. Again per Sec II, 333 Gr 6, full size specimens require 13/10ft lbs @ -50F, but, if you were to test 106GrB at -50F, it seems per UG 84 it would have to meet 15/12 ft lbs, for full size specimens.

 

[metengr]

weldtek;
I reviewed UG-84. In UG-84 (c), it lists an exception to materials that are produced and impact tested in accordance with general note (c) in Figure UG-84.1.

This general note pertains to SA-333 and other listed materials, and mentions that these materials do not have to satisfy the minimum energy value curves in Figure UG-84.1 provided they are used at minimum design metal temperatures which are not colder then the test temperature in the specification.

 

[weldtek]

metengr,
I follow that, so, would it be correct to say that if 106Gr B were to be used at -50F, it would need to meet 15/12 ft lbs?
thanks,

 

[metengr]

Yes, it would have to meet 15 ft-lbs (average of three specimens) CVN.

 

[robsalv]

Thanks Metengr. Star to you.

I see the error of my ways. General note C has been spotted. Also I erroneously thought A106B had a higher minimum yield, hence me going erroneously to the higher curve in figure UG 84.1...

Thanks Weldtek for spelling out that A333-6 is supplied with minimum 13/10 ft-lbs impacts [table4 of ASTM A333 spec] and if using A106B instead of A333-6, ASME VIII UG84 would specify minimum 15/12 ft-lbs impacts.

So technically, A106B would need to meet slightly more conservative minimum impact values meaning in theory, A106B could be tougher than "as supplied" [and UG 84 exempted] A333-6... Any feel for whether A333-6 would generally romp in the 15/12 ft-lbs anyway...?

Thanks for helping clear this up.

 

[metengr]

Yes, A-333 Grade 6 should easily meet 15 ft-lbs at -50 deg F. As far as A-106, this steel would probably be too close to call at -50 deg F (if I were a betting man, I would say it would not generally meet 15 ft-lbs at -50 deg F).

 

[codeeng]

In any event if you can't get Code impact test exemption for 106-B you would be wise to use LT material for a LT application. I went down this road once before to save a few bucks - never again!