ASK MTI - PUBLIC DISCUSSION

Recently, we had some failures on the tube to tubesheet welds. And we have reasons to believe the failed welds were due to the undesired microstructure (i.e. having too much ferrite due to high cooling rate). Generally, we don't do any ferrite testing here on such weld, and are thinking to incorporate this requirement into our standard. So, just to check around if anyone practices doing this. 

Thanks, 

Recently, we had some failures on the tube to tubesheet welds. And we have reasons to believe the failed welds were due to the undesired microstructure (i.e. having too much ferrite due to high cooling rate). Generally, we don't do any ferrite testing here on such weld, and are thinking to incorporate this requirement into our standard. So, just to check around if anyone practices doing this. 

Thanks, 

Guessing this is duplex SS? Or? If so, what alloy of tube and tubesheet (clad or solid?) and filler metal.

Recently, we had some failures on the tube to tubesheet welds. And we have reasons to believe the failed welds were due to the undesired microstructure (i.e. having too much ferrite due to high cooling rate). Generally, we don't do any ferrite testing here on such weld, and are thinking to incorporate this requirement into our standard. So, just to check around if anyone practices doing this. 

Thanks, 

@Gary: yes it is DSS.  The tube, tubesheet and fillers are 2205 or equivalent. 

@ Charles: thanks for sharing. After those tests being completed and found satisfactory, do you do the ferrite check or any other test (beside PT) on the production welds?

Guessing this is duplex SS? Or? If so, what alloy of tube and tubesheet (clad or solid?) and filler metal.

Recently, we had some failures on the tube to tubesheet welds. And we have reasons to believe the failed welds were due to the undesired microstructure (i.e. having too much ferrite due to high cooling rate). Generally, we don't do any ferrite testing here on such weld, and are thinking to incorporate this requirement into our standard. So, just to check around if anyone practices doing this. 

Thanks, 

Recently, we had some failures on the tube to tubesheet welds. And we have reasons to believe the failed welds were due to the undesired microstructure (i.e. having too much ferrite due to high cooling rate). Generally, we don't do any ferrite testing here on such weld, and are thinking to incorporate this requirement into our standard. So, just to check around if anyone practices doing this. 

Thanks, 

In addition to Gary's material question, what was the failure mechanism? Was ferrite checked on your WPS development? Are you accelerating the cooling any or is it cooling in still air?

Recently, we had some failures on the tube to tubesheet welds. And we have reasons to believe the failed welds were due to the undesired microstructure (i.e. having too much ferrite due to high cooling rate). Generally, we don't do any ferrite testing here on such weld, and are thinking to incorporate this requirement into our standard. So, just to check around if anyone practices doing this. 

Thanks, 

I would like to hear from someone how to check the ferrite in that production weld.  Sounds difficult.  But for me, assuming 2209 filler metal, the most sensitive area would be the tubesheet HAZ.   Any pre-production tests should be done on the actual thickness and chemistry of the tubesheet.  The thicker the tubesheet, the quicker the cooling, and nitrogen content of the tubesheet is key.   Of course the ER2209 filler should also have a reasonably high N content - many producers will not supply 0.14% min. N on wire filler metal, and 0.08% min. is the AWS standard.   (Unlike the S32205 wrought product with 0.14% min. N, the A5.9 wire has no such modification.)  The higher nickel of the ER2209 most often compensates for quick cooling, but there are limits to that.

In addition to Gary's material question, what was the failure mechanism? Was ferrite checked on your WPS development? Are you accelerating the cooling any or is it cooling in still air?

Recently, we had some failures on the tube to tubesheet welds. And we have reasons to believe the failed welds were due to the undesired microstructure (i.e. having too much ferrite due to high cooling rate). Generally, we don't do any ferrite testing here on such weld, and are thinking to incorporate this requirement into our standard. So, just to check around if anyone practices doing this. 

Thanks, 

In addition to Gary's material question, what was the failure mechanism? Was ferrite checked on your WPS development? Are you accelerating the cooling any or is it cooling in still air?

------------------------------
Justin Bekker
Engineer
Calgary AB
(587) 323-9042

Original Message:
Sent: 02-06-2024 12:52 PM
From: Victor Doan
Subject: Ferrite check on tube to tubesheet weld ?

Recently, we had some failures on the tube to tubesheet welds. And we have reasons to believe the failed welds were due to the undesired microstructure (i.e. having too much ferrite due to high cooling rate). Generally, we don't do any ferrite testing here on such weld, and are thinking to incorporate this requirement into our standard. So, just to check around if anyone practices doing this. 

Thanks, 

------------------------------
Victor Doan
LyondellBasell
Houston TX
(908) 992-5647
------------------------------

@Justin: the only corrosive side is cooling water on the tubeside. Few pits were formed and eventually created the leak path.

No, the WPS followed the QW 193 (Section IX), so no ferrite was checked. 

Just air-cooled (at least there was no requirement from us to do otherwise).

------------------------------
Victor Doan
LyondellBasell
Houston TX
(908) 992-5647

Original Message:
Sent: 02-07-2024 10:12 AM
From: Justin Bekker
Subject: Ferrite check on tube to tubesheet weld ?

In addition to Gary's material question, what was the failure mechanism? Was ferrite checked on your WPS development? Are you accelerating the cooling any or is it cooling in still air?

------------------------------
Justin Bekker
Engineer
Calgary AB
(587) 323-9042

Original Message:
Sent: 02-06-2024 12:52 PM
From: Victor Doan
Subject: Ferrite check on tube to tubesheet weld ?

Recently, we had some failures on the tube to tubesheet welds. And we have reasons to believe the failed welds were due to the undesired microstructure (i.e. having too much ferrite due to high cooling rate). Generally, we don't do any ferrite testing here on such weld, and are thinking to incorporate this requirement into our standard. So, just to check around if anyone practices doing this. 

Thanks, 

------------------------------
Victor Doan
LyondellBasell
Houston TX
(908) 992-5647
------------------------------

Hi Victor,

I would suggest to conduct a "verification weld" to supplement your current WPS. This weld's testing would include ferrite testing and possibly a microstructural evaluation to check for any detrimental phases/precipitates. Based on this verification weld, you may want to restrict your heat input to a certain range. ASME IX doesn't address some of the best practices for welding DSS. Similar to what Charles has stated, I always recommend ferrite and microstructural checks when developing DSS WPSs. ASME IX doesn't address it, but often end user specifications require it. 

@Justin: the only corrosive side is cooling water on the tubeside. Few pits were formed and eventually created the leak path.

No, the WPS followed the QW 193 (Section IX), so no ferrite was checked. 

Just air-cooled (at least there was no requirement from us to do otherwise).

------------------------------
Victor Doan
LyondellBasell
Houston TX
(908) 992-5647

Original Message:
Sent: 02-07-2024 10:12 AM
From: Justin Bekker
Subject: Ferrite check on tube to tubesheet weld ?

In addition to Gary's material question, what was the failure mechanism? Was ferrite checked on your WPS development? Are you accelerating the cooling any or is it cooling in still air?

------------------------------
Justin Bekker
Engineer
Calgary AB
(587) 323-9042

Original Message:
Sent: 02-06-2024 12:52 PM
From: Victor Doan
Subject: Ferrite check on tube to tubesheet weld ?

Recently, we had some failures on the tube to tubesheet welds. And we have reasons to believe the failed welds were due to the undesired microstructure (i.e. having too much ferrite due to high cooling rate). Generally, we don't do any ferrite testing here on such weld, and are thinking to incorporate this requirement into our standard. So, just to check around if anyone practices doing this. 

Thanks, 

------------------------------
Victor Doan
LyondellBasell
Houston TX
(908) 992-5647
------------------------------

Recently, we had some failures on the tube to tubesheet welds. And we have reasons to believe the failed welds were due to the undesired microstructure (i.e. having too much ferrite due to high cooling rate). Generally, we don't do any ferrite testing here on such weld, and are thinking to incorporate this requirement into our standard. So, just to check around if anyone practices doing this. 

Thanks, 

Recently, we had some failures on the tube to tubesheet welds. And we have reasons to believe the failed welds were due to the undesired microstructure (i.e. having too much ferrite due to high cooling rate). Generally, we don't do any ferrite testing here on such weld, and are thinking to incorporate this requirement into our standard. So, just to check around if anyone practices doing this. 

Thanks, 

Recently, we had some failures on the tube to tubesheet welds. And we have reasons to believe the failed welds were due to the undesired microstructure (i.e. having too much ferrite due to high cooling rate). Generally, we don't do any ferrite testing here on such weld, and are thinking to incorporate this requirement into our standard. So, just to check around if anyone practices doing this. 

Thanks, 

Recently, we had some failures on the tube to tubesheet welds. And we have reasons to believe the failed welds were due to the undesired microstructure (i.e. having too much ferrite due to high cooling rate). Generally, we don't do any ferrite testing here on such weld, and are thinking to incorporate this requirement into our standard. So, just to check around if anyone practices doing this. 

Thanks, 

Hi Victor

if you are going to apply ferrite measurement as part of the PQR check out NACE paper 10634 from the 2018 conference written by JCG.

they show quantitative metallography results in welds depend heavily on how you do the testing.

Also, the phase balance of tube/ tube sheet welds in DSS/SDSS can be improved by the use of argon nitrogen shielding gas mixes.

yours

Glenn

Recently, we had some failures on the tube to tubesheet welds. And we have reasons to believe the failed welds were due to the undesired microstructure (i.e. having too much ferrite due to high cooling rate). Generally, we don't do any ferrite testing here on such weld, and are thinking to incorporate this requirement into our standard. So, just to check around if anyone practices doing this. 

Thanks, 

Victor,

To follow-up on Glenn's suggestion to reference NACE Paper No. 1010634 and Gary's comment that "...checking the ferrite in that production weld. Sounds difficult.", of the three variables studied in the NACE Paper, the measurement locations by far had the largest range of ferrite measurement values; the other two variables being: grid size and number of fields. 

So it's not surprising that the most significant conclusion in this study was: 

"...ferrite content varies greatly across the fusion line (weld interface) so that the ferrite measurement method for HAZ of DSS welds commonly employed in welding qualification procedure i.e. calculating mean ferrite content from a set of measurement locations may not be appropriate for evaluating ferrite content in HAZ."

Similarly, the most significant suggestion in this study was:

"As it has become clear that measured ferrite content is greatly affected by the measurement
location, PQR prepared by fabricators should at least include the micrographs showing where in
the weld ferrite measurements were taken from" (i.e., the location).

So rather than trying to measure the ferrite content in production weld, you would be better off improving your existing WPS with ferrite measurements in the most appropriate locations during further PQR development to establish the welding variables most influential at those locations.

The metal doesn't care how or when you measure ferrite content because the metal's performance only cares where it's corrosion resistance is weakest.

If you focus on refining your WPS as suggested in this study and strictly follow your WPS in production welding, you'll optimize the weld's performance.

Lastly, I found it very interesting that this study pitting was found in the weld metal, not in the HAZ, using ASTM G48, Method A.

Keep asking questions in your search to find answers to problems.

Hi Victor

if you are going to apply ferrite measurement as part of the PQR check out NACE paper 10634 from the 2018 conference written by JCG.

they show quantitative metallography results in welds depend heavily on how you do the testing.

Also, the phase balance of tube/ tube sheet welds in DSS/SDSS can be improved by the use of argon nitrogen shielding gas mixes.

yours

Glenn

Recently, we had some failures on the tube to tubesheet welds. And we have reasons to believe the failed welds were due to the undesired microstructure (i.e. having too much ferrite due to high cooling rate). Generally, we don't do any ferrite testing here on such weld, and are thinking to incorporate this requirement into our standard. So, just to check around if anyone practices doing this. 

Thanks,