Solid CRA Nozzle Welded to CS Vessel

[Angsi]

Code: ASME
Mat'l: Predominantly A516 Gr 70 + 3mm 316L
Location: Offshore Platform
========================================================
Due to wet CO2 corrosion, most of the vessels require 316L internal cladding. The reservoir temperatures are high and in some cases up to 140 deg C.

After speaking to some vessel fabricators, they admitted that when it comes to smaller nozzles, it is almost impossible to get a less than 8% ferrite content at the weld overlay fusion line. Thus these smaller nozzles will fail the chemical analysis required.

The best and easiest solution to this problem will be to use solid CRA.

If solid CRA nozzles are used, the temperature then becomes an important factor in selecting the right material. As the reservoir temperature is high, the temperature could vary between 40-140 deg C depending upon the service this vessel is in. The CRA material could be a choice between stainless steel, duplex and superduplex.

Question:
=========
Is there any problem welding stainless steel, duplex or super duplex nozzles onto a internally cladded carbon steel vessel? Is welding a concern? Is there anyting else regarding welding of dismilar material that I need understand before taking this approach?

 

[metengr]

Yes. If the nozzle is a solid CRA, special precautions for welding to carbon steel substrate are in order because this is a dissimilar metal weld (DMW). You will also need to take into account the differences in thermal expansion and mechanical properties between the solid CRA nozzle material and the carbon steel substrate as part of your vessel design evaluation for service conditions. I don't believe the use of superduplex ss or even duplex ss may be optimal, as well.

Regarding DMW, the cladding will need to be stripped back on the carbon steel substrate to facilitate DMW along with a proper selection of filler metal. The filler metal could be 309L or 309MoL for joining either an austenitic stainless steel or duplex stainless steel nozzle to the carbon steel vessel wall.

Post weld heat treatment (PWHT) requirement per the Code of Construction needs to be reviewed and will only apply to the carbon steel vessel wall. If you elect to use duplex ss as the nozzle material, PWHT should NOT be performed. In this case, the carbon steel weld prep along the opening of the nozzle penetration would need to be buttered first using 309L and locally subjected to post weld heat treatment before welding in the duplex ss nozzle. This is why I would use a 316L nozzle material in lieu of a duplex or superduplex ss. Once this DMW operation is completed, a weld overlay of 309L will be necessary to protect the nozzle attachment weld up to the solid CRA nozzle and surrounding exposed base metal that was stripped of cladding.

 

[Angsi]

metengr, is this a common practise when it comes to 2" nozzles? It is just that I am not willing to try something that is out of the ordinary. What has your experience been on this matter? Anyone?

 

[unclesyd]

We use this approach all the time on CRA lined vessels and we also used this approach on CS vessels to resolve some steam condensate corrosion problems.

We have 14, 100,000 gal 304/316 SS clad reactors and all nozzles are solid CRA alloys. There are no clad nozzles in these vessels.

When we have to replace or increase a nozzle in size we use essentially the same procedure as described by metengr. We stay with the 309LMo if the clad is 316/316L SS.

 

[Angsi]

Many thanks. A couple more fundamental questions. Though it has been an industry practise, what is the basis for the chemical analysis? I recall part UCL recommends chemical analysis is performed to verify quality of overlays but which international code/standard addresses the chemical analysis of weld overlays in specific (i.e. one layer, two layer techniques, ferrite content etc?)

Another silly question. Why should the chemical composition at 1mm above fusion line matter when the other 2mm has the right composition (i.e 316L)?

Thanks.

 

[metengr]

The requirement for chemical analysis and ferrite content is driven by the customer engineering specification, more than anything else.

Regarding your second question, I agree. Dilution of the weld overlay with the base metal will be a factor at 1mm from the weld fusion line.

 

[unclesyd]

Adding a little to the above post. I think a lot of specifications you see for clad and clad penetrations weld procedures still carry over from the days when you didn't have the variety of welding electrodes available you have today. At times it was almost impossible to overmatch the chemistry to help mitigate the dilution effect. Like 310Mo was hard to come by and an L grade was impossible to get. Every time you asked for it the answer was always that you were welding on CS, why do you need this.