Stainless Steel Selection High Temp Environment 1

[rmillercwi]

I am looking for some assistance deciding the correct filler metal to use to weld non-code attachments in a high temp, high velocity location inside an HRSG. I have experienced weld cracking of 304 to 409 duct liner components and 304 to 304 & 310 to 310 header restraints. Max temp 1200 degrees in the inlet duct and up to 1500 degrees in front of the duct burner. Typicly a 309 filler would be used for the 304 to 409, 308 for 304 to 304 and 310 for the 310 to 310. I have wittnessed several weld failures in these locations with several different fabricators. Most failures are occuring within two years of service. Thinking maybe of going with a 304,310 with an L or H designation might be appropriate. I have read the addition of carbon will assist with temperature resistance to eliminate the cold cracking. Welds breaking off at the toe of the weld on the liner items and welds breaking in the center of the weld on the header restraints. Or maybe an Inconel or 347 filler would be the permanent fix?

Any help would be appreciated

 

[EdStainless]

If you make the welds stronger you are likely to break more things. These are geometry issues. You have the right fillers, you need to figure out how to put less load on the welds.

= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection

http://www.trent-tube.com/contact/Tech_Assist.cfm

 

[metengr]

Switch to Inconel filler metal for the header restraint welds because the centerline cracks could be indicative of undersized or inadequate strength fillet welds.

For the weld toe cracks in the duct liner, this could be low cycle (thermal) fatigue crack propagation. You might need to add additional bracing to strengthen (reduce flexing)the duct liners.

 

[rmillercwi]

Thank you for the responses. The non code attachments in the inlet duct dont really have any load on them other than the high velocity they experience. Allthough, they do have the potential to rattle if not installed correctly. One side of the liner section (10-20gauge) is attached to something solid (beit a corner angle or another section of liner) with the other side z-clipped (2"wide,1/4"thk) every 12-18" or so to allow thermal expansion and yet keep the component tight. Whats odd is that its typicly the field welds that are failing. Identical materials welded in the shop are having a noticeably higher success rate.(thinking a more researched procedure/installation practice is being used). Most of the failures are visible just downstream of the diffuser on the floor and then again apparent just upstream of the first harp on the floor and the walls. If you inspected just one unit you would think for sure it was a welder error. But after finding similar issues in similar locations in several different OEMs, I am thinking its a filler metal/procedure issue. The completed welds look fine but appear to have been welded too cold as there is very little to no penetration into the basemetal on the failed side. Typicly the thinner of the two sections being welded but not always. The non failed side of the weld appears to have adequate penetration. My theory is that a filler metal with more strength, penetrability and resistance to heat and velocity is the fix. I visited a couple filler metal supplier websites and was where I learned of the reasoning behind the carbon L and H designations. Apparently adding higher levels of carbon to stainless promotes higher temperature resistance. At least thats how I percieve it.

The header restraints have been redesigned to allow better heat transfer. The higher ups believe they have the fix nailed down (as do I) but I am still concerned this issue and above are somehow related. I also am in favor of using inconel perhaps an erincr3? Would the same inconel filler be acceptable for joining both 304 to 304 (inlet duct restraints) as well as the 310 to 310 upstream of the duct burner? Would you suggest a gtaw or a smaw process?

Sorry for the long post and all the questions. Your thoughts appreciated.

 

[metengr]

Whats odd is that its typically the field welds that are failing.

This is a very important observation, and implies that what is being done in the field is not the same approach that is done in the shop. You need to research this item and determine what changes are being done in the field versus the shop.

The completed welds look fine but appear to have been welded too cold as there is very little to no penetration into the base metal on the failed side. Typically the thinner of the two sections being welded but not always.

This is another important observation. If what you are reporting is correct, this could be a dissimilar metal weld failure or poor welding technique. Normally, the dissimilar metal weld failure would occur on a ferritic to austenitic stainless steel joint because of carbon migration (ferritic side). Switch from 309 to Inconel (ERNiCr-3).

Go with the GTAW process. For the 304 to 304, I would consider ER308H and for the 310 to 310, I would use ER310.

 

[CalvinKelly]

In Rolled Alloys' experience, most weld failures at high temperature are the result of incomplete weld penetration. One simply cannot get the same penetration in a stainless weld as in a carbon steel weld. Any plate 3/16" or greater needs to be beveled to get the weld bead deep into the joint. Unwelded areas act like cracks, and grow outward through the weld bead under thermal or mechanical cycling, until final failure.
The ferrite in stainless weld metals will turn to sigma in over the 1200-1500F temperature range you mentioned. On occasion flux cored wire can have very high ferrite, or at least not be what you think it is. I would suggest making all of these dissimilar metal welds using alloy 82 (ERNiCr-3) wire or 182 (ENiCrFe-3) covered electrodes.
If you wish illustrations of appropriate weld joints, email me jkellymetal@gmail.com
James Kelly

James Kelly

http://www.rolledalloys.com

 

[EdStainless]

Hi Jim. Good to see that you are still working.
Will I see you in Nashville?
Ed B.

= = = = = = = = = = = = = = = = = = = =
Rust never sleeps
Neither should your protection

http://www.trent-tube.com/contact/Tech_Assist.cfm