Local (spot) PWHT after attachments welding (wet H2S service) 2

[teokal]

Hi to all!
The question is as follows:
We have constructed a Column (18mm thickness of SA516 Gr70) which is for H2S service, so PWHT was performed in our premises, in a furnace. Now that the Column is on site, the Client requires some extra internal supports for a new demister to be installed, on a specific elevation. Since the attachments will be welded, and since it is a Client's requirement, PWHT will be performed on site. My question is wether we can perform local (spot) PWHT on each attachment, or we should apply PWHT on the perimeter of the Column, at the specific elevation. One of my first experiences was that it was obligatory - by that time client - to perform PWHT on the perimeter, but I cannot find any references. In fact I cannot find any references to this, by pointing or excluding either ways. Any opinion or experience to this?
Thank you in advance for your replies.

Only best result in absolutely necessary time

 

[vesselfab]

Normally, an entire band is heated, but, if the NBIC AI and Customer approves, local bullseyes of heat pads may be able to be used for a couple of small areas of weld attachments.

If there are several areas of weld attachments, I would elect to to go with an entire circumferential band.

 

[Duwe6]

Respectfully disagree with vesselfab. When performing PWHT of welds for 'Process', heating an area larger than that of each attachment weld is not necessary, will add [a small ammount of]risk, and may require bracing.

The Column in question is now installed outdoors, and is subject to wind loading. I am leery of heating and, by definition, slightly softening a full circumferential portion of this vessel now that it is exposed to wind, and possibly carrying additional weight in the form of platforms & ladders, and possibly a load of trays or pall rings. "If its not broke, don't fix it".

 

[TGS4]

I will have to agree with vesselfab. The reason that a PWHT is performed for process reasons is that there is a corrosion mechanism that, when combined with a high level of stress, will cause problems. Doing nothing, or performing a bulls-eye PWHT will not likely solve your problem, and could even make the residual stress issue worse. I just saw a presentation at PVP-2011 that showed that a poorly-performed local PWHT can actually make the residual stress situation worse.

Follow the recommendations in WRC-452. Make sure that your PWHT contractor is VERY familiar with this document, and has used it before. If they haven't used it or aren't familiar with it, find a new PWHT contractor.

To Duwe6 - in this case, the column in indeed "broke". It MUST be fixed. Based on my experience with evaluating full-encirclement PWHTs on tall columns before, it is unlikely that additional reinforcement will be required. At worst, there may be a wind speed limit, which basically means don't try this in a hurricane.

 

[Duwe6]

Gotta disagree again, sorry. I've seen Amine service [very harsh on stressed material], in addition to NH3 small attachments and nozz's successfully spot-PWHT'd which delivered the expected service life. In other words, no stress-related problems.

If you weld in a 1"NPS nozz, the HAZ will have an OD of about 1.5", and the stressed area will be roughly 3-4" in diameter. For that, all I have ever seen or done is to PWHT a 12" square area. The remainder of the column circumference isn't broke, or stressed. Why would I want to PWHT all that area? $$$

Small welds only change the properties in small areas. Yes, the weld shrinkage from my hypothetical 1" nozz will put tension in an entire band around the column. After I relax the small section that is pulling on the column, there is no longer a stress band around the column.

If you weld in a small area, PWHT is only needed in a small area. To expand on the "Gotta do a full-encirclement" argument, what about field PWHT of circumferential welds? We weld all the way around a pipe or vessel, and the shrinkage is both circumferential and axial. The PWHT is only a band that extends a few inches on either side of the margins of that weld. If there was a significant need to do a full circ PWHT for a small attachment, then we would need to PWHT the entire length of the pipe or vessel to relieve the axial stress. PWHT isn't to relax where the stress goes, it is to relax the area of shrinkage that is actually causing the stress.

 

[metengr]

You might also want to look at temper bead welding for these attachments, as provided in Part 3 of the NBIC, under alternative welding methods. This is an ideal repair method for carbon steel in lieu of PWHT, although, I must admit I am a fan of the current local PWHT in the NBIC, which we patterned after AWS D10.10.

 

[Duwe6]

Personally, I am leery of using Temper Bead for items that require PWHT for process -- Amine, liquid NH3, wet H2S, etc. I havn't seen enough documentation of how they perform decades later.

If that documentation exists, could someone please point me in the right direction, please?

As far as Temper Bead in lieu of PWHT on material thick enough for the Code to require PWHT, Temper Bead works great and seems to have the same service life as having heat-treated that weld. I always use it -- faster, cheaper, equally good results.

 

[TGS4]

Duwe6 - I have nothing against a local (bullseye) PWHT, as long as it is done correctly - properly sized soak bands, appropriate gradient control bands, etc. Not everything NEEDS a full encirclement band.

Also, you would be very surprised at the effect that weld residual stresses would have on adjacent regions. Likewise for poorly executed PWHTs.

My recommendation is do the PWHT - follow WRC-452. Here's my rationale: if, say 10 years down the road, this vessel develops excessive corrosion-related damage that causes either an unplanned outage ($$$) or an injury, and that damage can be, even tangentially related to either a PWHT that wasn't done, or an improperly-performed PWHT, would you be comfortable arguing in front of a judge that what you did was OK? Was it following good practice? Was it following standard industry practice? Or was the one-time cost years before not worth doing something right? My answer to all of the above is, that I would do it right, and the good practice/industry standard practice is that the vessel was called out to be PWHT-ed, for process reasons, therefore, the vessel should operate in the completely PWHT-ed condition. Maybe temper bead could work - I don't know. WRC-452 has been out for more than 10 years: it's the definition of an industry good practice - follow it.

 

[Duwe6]

TGS4, I absolutely agree with you on poorly-done or no PWHT. It is indefensible in court, and/or in your customer's boardroom. This is a small part of why I am leery of Temper Bead in lieu of PWHT-for-Process.

Do it right, do it once, and everybody can go home safe & happy, with a proudction unit purring along and making $$$$.

 

[vesselfab]

Guys....didn't mean to stir a hornets nest up.

I answered based upon not knowing the amount of welding involved or the size of tower or whether it was erected or not and that is was PWHT for service and not for thickness. Iand i figured that a mist eliminator is generally just below top head so nt a big load factor.

I have never been a fan of temper bead to replace a PWHT for service.

Ya know, we can fuss all we want, but it's all up to the user and the NBIC AI

 

[Duwe6]

Not even the NBIC & AI. This PWHT is for 'Service', and is not a Code requirement. The only thing the AI will do is to assure himself that the heat-treat does not harm the vessel. That is one of the problems with PWHT-for-Service. Thus the Hornet's Nest.

Personally, I use an API-510 to verify the adequacy of this work. An experienced [field experience, not design experience] engineer also is a good choice.

 

[stanweld]

teocal,
It is up to the Owner or his Engineer to determine the PWHT methods for Service. If I were the owner and you were welding additional internal attachments, I would not permit temper bead welding.

Ask the Owner or his Engineer for direction.

 

[teokal]

Thank you all, guys, for the replies.
The owner had anyway desided from the beginning that no temper bead would be allowed, and my question was whether heat treating entire band, or local (bull's eye)... Following your comments and reading some WRCs and API's we have written down a very nice procedure, pointing all the problems for either ways, but the main concern was the fact that the column was already erected with the platforms and ladders on it and most important all the piping connected to the column. So, supporting (from the top using a crane) was very necessary for the column if entire band was heated. This - the cost of it I mean - has lead the owner to allow local (bull's eye), despite his initial will to perform it on entire band... So, colleagues, I believe (or at least I hope) that everybody has learned something from this post, so I thank you very much for your replies and sharing of knowledge.

Only best result in absolutely necessary time

 

[Duwe6]

teokal, thank you for being a gentleman/woman and giving the Forum the final resolution of your problem. I, and probably all members, appreciate someone "closing the loop".

Again, Thank you, sir/madam.

 

[teokal]

Thank you too, Duwe6, and to all of you once again. Apparently, a lot of technical questions in our field have financial solutions... Oh, and it's a "man".

Only best result in absolutely necessary time

 

[Guest102023]

metengr,

I have my doubts as to the suitability of temper bead welding when avoidance of SCC is the issue. When done well (a big if, because I see local mom-and-pop fab shops attempting DIY temper bead welding) it will enhance the mechanical properties and mitigate residual stresses.

In some cases it is residual tensile weld stress that is the critical factor, and this can only be effectively minimized by thermal methods.

 

[metengr]

brimstoner;
Selection of an alterative method to PWHT requires very careful consideration of material and service. There is work being performed by ASME to assign a maximum hardness for temper bead application. Stress relief requires thermal treatment.

For power boiler steam service, I have found and demonstrated by 20+ years of successful experience that temper bead works very well.

 

[Guest102023]

metengr,
I don't think we disagree here; TB is perfectly appropriate for many situations. As you well know it needs thorough procedure development and training by any organization attempting it.

 

[hafnihs]

I need information about where we get information the manufacture of the unit HEATER TREATER LIKE the pic.