T23 Boiler Tube - Curious Fracture 7

[Guest102023]

T23 Reheater Tube Fracture in Weld -- Anyone seen this fracture morphology before?


Several T23 reheater tubes failed in a similar manner after less than 6 months and ~250 cycles from commissioning of the HRSG. The circ. fracture originates in the weld, at the front (where you would expect bending force to be greater) then moves into the base metal as it propagates around to the back side. Meanwhile, a ~4" long tear happened on the front side. It seems like there is more than one thing wrong here.

Unfortunately I did not see the header side of the story (another case of repair first, ask questions later), so it is hard to confirm the lack of penetration that me and the repair contractor suspected.

Particulars: 2" T23, welded to P22 header with manual TIG, ER9018-B3, PWHT. Unknown tube origin, but they are significantly older than the plant. All zones appear to have normal microstructure (no ferrite at least). The correct filler was used. Weld hardness is 285HV1, HAZ up to 335HV1, base ~192HV1. I have found both tube surfaces decarburized. Scattered corrosion pits up to 3 mils deep on outer surface. A few very short secondary cracks and sometimes oxide along prior austenite GBs.

Any thoughts or suggestions? The hardness seems on the high side for a PWHT joint. I thought about reheat cracking, but did not think that can affect the weld also(?)

I am a newbie here so I hope this gets posted correctly ...

 

[unclesyd]

Can you post a picture/pictures of the fracture, say from about 12" away with a slight angle to the fracture.

 

[Guest102023]

Fig1 shows the general area. Fig2 shows a boroscope pic of the region before repairs - you can see part of the tube corresponding to the header at upper right (not the middle one). Fig3 shows the tube segment I am examining. At least two of the fractures have similar morphology - circ. fracture originating in the weld, and a long vertical tear, which is fairly square in profile.

I should add the HRSG is also experiencing chronic fractures of 1" drain lines below these headers, so we know thermal stress is the driver (surprise).

 

[Guest102023]

The link to the pics does not seem to work, don't know what to try next ...

 

[Guest102023]

Another attempt ...

 

[Guest102023]

Success ... here is another

 

[Guest102023]

... and finally

 

[Guest102023]

once again for Fig3 ...

 

[unclesyd]

Was the tubes welded to header using a socket?
just trying to understand the appearance of a fillet weld configuration on the tube or was this the previous attempt you mentioned to repair the leak?

The initial observation has the appearance of a rapid overheating but is tempered by the proximity to the header.

Very good pictures.

 

[Guest102023]

Initially I had the same impression, but evidently it is a full pen weld. That's how the macro section appears (attached), and that is how the contractor is repairing them. He reports that the OEM flattened the header surface at the joints and the tubes were just set on directly. Just wish I could have seen the headers before the repairs ...

Do you mean rapid overheating in service? There was no information given to indicate tube temperature excursion.

 

[unclesyd]

Overheating in a tube can occur without any general indication of overheating. One very common cause is steam blanketing where there is no appreciable scale to account for the overheating. There has been some high heat indicated by the surroundings, the very shinny black metal. it would be hard to tell if there had been steam impingement.

It also appears there is no shear lip on the longitudinal fracture surface while there is on the circumferential fracture surface is this true.

You mention failure of other components near this failure is it possible that this tube was quenched, or dry, for lack of a better word?

I'm looking for some very old pictures of 2 similar failures in a waste heat boiler. If my recall is correct it was quickly decided to plug the header and make no attempt at a repair.

 

[Guest102023]

OK, I understand what you mean. These tubes are near the hottest part of the flue gas (the hottest tubes are P-91). In full operation they carry steam only of course, but I am not knowledgable enough (yet!) about HRSG operation to know what exactly can happen during cycling up. I thought shiny and black (magnetite) was the preferred condition for tubes.

On the longitudinal fracture there are small shear lips, aided by the soft decarburized surface layers - hardness was as low as 129HV1 there. (Makes one wonder how they checked the hardness after N&T and after PWHT.) I am looking at this condition not so much as a possible cause but as an indication of a lower quality tube supplier. But the weld fracture appears fairly brittle; what you are probably looking at is the fillet on top of the groove weld. I am awaiting better micrographs of some of the fracture surface detail.

 

[unclesyd]

The shiny black as you posted is nominally magnetite as you stated. This is the wrong scale for a normally operating boiler. As some would say this is burnt metal, not totally wrong.
Has anyone recently altered the hot gas flow path in this area?

Two reasons I ask, one is the possible overheating and the other is a remote possibility of fatigue as the cause of the original failure. The fatigue aspect comes from your mention of brittle looking failure. We had a fatigue failure in a screen tube on another waste heat boiler right at the header.
If you can get some thickness measurements of the failed section along with the data you have already have it might give some indication of the time frame.

The additional photomicrographs you mentioned will be interesting.

 

[metengr]

In addition to the above, have you verified by chemical analysis that the failed tubes are indeed Grade T23?

The axial-oriented fracture surface looks like it exhibits chevron markings that point back to the circumferential weld, and the fracture looks rather flat and brittle-like in appearance(hard to tell with the picture).

I will tell you this; I have seen similar failures on conventional Grade T22 material where we had a serious water hammer event during a hot start condition in of our HRSG’s. The water hammer event was so localized and severe we sheared tubes from the header and several split axially. This does not appear to me to be the result of overheating - just a guess and based on what I have seen and read thus far.

 

[Guest102023]

I can understand overheating on startup, but how/why by steam blanketing when there should be no water present? I will ask some questions about the set-up.

Fatigue is an interesting suggestion, and would explain a few observations. Certainly the cyclic thermal strain is substantial. I just have found no evidence to justify the lack of penetration I had expected to find. It could also be both, of course. The metallographic section suggests part of the fracture surface was exposed for a period of time.

 

[Guest102023]

The tube chemistry conforms to T23 (minus the B & N which were not done). EDX done on the weld confirms -B3 filler.

No reports of water hammer; I think someone would have noticed. As for the vertical tear, I think that should not have happened unless something was very wrong - either insufficient tube properties or excessive loading (such as water hammer). The tube appears normal, decarb notwithstanding ... makes the question of which part of the fracture came first a little trickier. Another pic attached.

 

[metengr]

No reports of water hammer; I think someone would have noticed

Not necessarily a true statement for many of the plants I have dealt with over the years. Keep in mind that if your header drains are not routed or vented correctly and condensate pools in one location of the header and you start-up in this condition, you stand a high chance of a local water hammer event or subsequent shock wave that will fail thinner wall tubes versus a header. I have seen this before.

Unfortunately decarb is typical for most boiler tubing unless it is dealt with by customer specification. Second, your close up view of the axial rupture does not appear to be from exposure to elevated temperature service. The Grade T23 materials have high allowable stresses which mean they have good high temperature strength and creep strength, if heat treated correctly.

I believe your failure initiated from the axial rupture and the only type of operating event that can cause this in service is shock loading from a water hammer or over temperature excursion. I don't see any evidence that this tube saw excessive metal temperature short or long.

 

[Guest102023]

I am coming around to your thinking, because I have to think this violent rupture was caused by an unusual event. I started out thinking weld LofP, but have no evidence.

Even fatigue does not explain the vertical tear (no evidence by SEM fractography, although there was much post-failure damage).

 

[Guest102023]

Have just discussed this with the plant engineer, who advised they were already concerned about water hammer isues. Good news, because I had pretty much ruled out metallurgical factors.

Thanks to all - I will reciprocate where possible. My specialities are welding engineering and failure analysis (mainly refinery). I just started career #3, working for myself.

 

[unclesyd]

Glad your getting a closer to a resolution even though I may have taken you the long way around. I can always say that I kept you talking until metengr came along.

Just out of curiosity you mention two things that If possible a little more information would rev up my windmills.
The first was that was the manufacturer flattened the header and set-on the tubes and welded. Was that his method of attaching the tubes to the header?

The second was that they start up the boiler dry?

If you get any more information from observation or analysis please post.