ASTM G48A - length of test period 1

[bruv]

ASTM G48 method A suggests that 72hrs is a reasonable test period. All the specs that we have previously worked to have a test period of only 24hrs (Norsok, Shell, ExxonMobil etc)

Does anyone know of a spec that insists on a 72hr test period? We have a client who has an ongoing internal debate on whether to change their data sheets.

 

[EdStainless]

I have always used 72hr as it makes the test less sensitive to variations in preparation.
For well conducted tests the 24hr and 72hr results are identical. This has been proved for methods C, D, E, F, but I have never seen such data for A and B.

Of course I would also suggest the use of Method C. The results from A are much more erratic.

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

 

[SJones]

What material are they wishing to test, and what are they wishing to test it for?

Steve Jones
Materials & Corrosion Engineer

http://www.linkedin.com/pub/8/83b/b04

 

[bruv]

EdStainless - we will be conducting tests on our own material before replying to the client.

SJones - At the moment, cast super-duplex stainless (ASME SA995 6A), but it may be extended to other materials.

 

[SJones]

Then why go against ASTM A923?

Steve Jones
Materials & Corrosion Engineer

http://www.linkedin.com/pub/8/83b/b04

 

[EdStainless]

G48 is only a method. You have to specify sample prep (is any at all allowed?), test temperature, test time, and acceptance criteria.

I really dislike using the original non-acidified test solution. It can vary considerably from one batch to another.



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

 

[SJones]

The standards that bruv quoted do deal with ancillary test parameters. Essentially, they follow on from the TWI developed standard test method for duplex welds.

Steve Jones
Materials & Corrosion Engineer

http://www.linkedin.com/pub/8/83b/b04

 

[bruv]

Then why go against ASTM A923?

Because our clients ask for G48 method A plus charpy-impact testing, usually at -46C, plus micro-examination & ferrite count to ASTM E562. They, in turn, are driven by the requirements of their clients.

 

[EdStainless]

A923 and G48 are VERY different tests done for VERY different reasons. You cannot sub one for the other, or use on for both. The only similarity is the test media.

The A923 requires a freshly ground surface to reveal any secondary phases that may be present. The temperatures are well below the critical pitting temp for the given alloys. The goal is to see if there is corrosion due to secondary phases in the structure.

G48 is used as a guide to determining or verifying the critical pitting (or crevice) corrosion temperature of a sample. We use G48C on finished tubing, no surface prep of any kind is allowed (strictly as shipped), and we fix time and temp. We expect 100% to pass this test as it is about 10C below our historically determined CPT.

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

 

[SJones]

Indeed, couldn't agree more, but the NORSOKs and Shells of this world are working on the A923 objectives and that's why they are all aligned on a 24 hour exposure. Unfortunately, they call up G48 in doing so instead of asking for 923 (probably because TWI developed the test procedure way before the advent of 923 when there was only G48)

Steve Jones
Materials & Corrosion Engineer

http://www.linkedin.com/pub/8/83b/b04

 

[EdStainless]

The grinding of the surface is the critical issue with testing duplex grades. Selecting the correct temp for a specific alloy is also important.

A923 is being continually reviewed and revised. New alloys are being added and errors fixed. Someone trying to work with duplex grades could not ask for a better spec to work with.

Writing a spec does not end your work, it only begins it. A spec that hasn't been revised in five years is not worth using. I have been known to quote jobs with separate price adders for various testing. I can send a message that way.

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