Welding P91 Thermal Power Plant 1

[doses]

Hello:

First post that I do here and I'm not english native speaker, so if my language is not precise, please be patient.
I'm a mechanical engineer working on the design of piping (engineering stage) of a Thermal Power Plant (Vapor Turbine).
I have to make specification for design and fabrication of transition pieces for piping of base material P91 (from Boiler to turbine).
I have read many threads here about the subject of "P91", and I'm very worried if I have understood all the variables that must be specified for elements of this materials and I've come here for some indications and answers.For example:
-When welding pipes or fitting with different thickness and same diameter ¿Is there a maximun difference of thickness before using transition piece?
-For some cases, we have to use Wye fittings for diameters of 14" or 16", and I'm not sure about the recommended manufacture process for this kind of elements, or the material specification for this (ASTM A234 or ASTM A335, etc).
-Finally, we are using ASME B31.1, but I have seen that ASME BPVC Section I have diffents requierments for PWHT of P91 welding. Then, two questions, ¿Why has B31.1-2007 not incorporate this requierments for P91? and ¿Why Ni and Mn affect the maximun temperature of tempering in PWHT?
I appreciate your help on this and thanks in advance.

David

 

[davefitz]

Doses:

It looks like you may have a lot of reading to do , to get up to speed on the issues with P91 welding and fabrication.

I'll try to followup with web links on relevant tech papers on the subject. Recent published papers by Jeff Henry may summarize the issues as they apply to code revisions that are being considered but are not yet incorporated in the ASME codes. But I think it is correct to state that the currently published ASME code minimum requirements are not considered to be sufficient for ensuring a long service life for a P91 component in high temperature service; the "designer" must have add'l training on the metallurgical issues to further amend the fabrication and welding spec to ensure a long ( 50 + yr) service life.

For a wye fitting , we have had failures of the weld seam of a fabricated P91 wye ( hot reheat transfer pipe) after 6 yr service life. It was replaced with a forged seamless wye, and many others will not now allow a welded fabricated P91 wye. But if you do use a welded fabricated wye, you should use a weld creep strength reduction factor WCSRF( about 0.75 at 1050 F, TBD) and/or N+T the fitting after the fabrication , and in this particular case where N+T is used to raise the WCSRF, a matching filler of 0.4% Ni may need to be used to ensure the filler creep strenght matches the parent metal- you need to furhter research this item- see recent japanese papers.

Higher Ni content weld fillers are used to prevent cracking of the weld prior to PWHT, but although this makes the weld more ductlie, it has a lower creep strength than the parent metal . This is OK for butt welds, but for seam welds on fabricated parts you will then need to add a WCSRF or insist on matching filler of 0.4% Ni. See jeff hendry papers on the effect of high levels of (Ni + Mg) on PWHT temp .

out of time .

 

[davefitz]

We have had multiple failures of the transition weld between the P91 HP main steam line and the 1.25 Cr- 1Mo -V turbine stop valve; the 3" thick valve weld end was directly welded to a 1.5" thick P91 pipe without a transition piece,and the weld failed in 16 mos , at 3 duplicate plants . The provided weld did not meet ASME requirements for a thickness transition slope of not greater than 2:1 ( ie 30 degree exterior slope) within 1.5 thickness lengths from the weld line. European PED recommends slope not greater than 3:1 ( 18 degree) in thermal fatigue service, and this weld certainly qualifies for sever thermal stress service.

Further, it appears that there is a loophole in ASME code that allows the weld zone thickness to be less than the min wall required thickness of the weaker alloy in the case of buttwelding 2 dissimialr metals. In our case, the wall thickness at the low Cr valve was only 50% of the required min wall based on published B31.1 allowable stresses. I do not recommenend using this loophole; instead, ensure that a F91 transition piece be used at this critical weld, thickness to match the min required thickness of the weaker low Cr material at the valve weld end and tape the P91 thickness to the thinner requried P91 thickness at a slope not greater than 3:1 ( 18 degrees).

I expect the ASME code to eventually recognize this issue , about the same time they remove the steam locomotive paragraphs, in about another 75 yrs.

 

[unclesyd]

As posted above one has to read, read and read some more to keep up with the developments in welding T/P 91.


Here is a very good paper on the welding of T/P 91 from Sperko Engineering. In addition there is an address to get additional information from VM Tubes.

http://www.sperkoengineering.com/html/Grade 91 R 10-05.pdf

 

[davefitz]

See below link to a 2005 article by Jeff Henry on P91 issues. The increase of (Ni+Mg)above 1.0% will require reducing the PWHT to below 1450F, due to suppression of AC1 and AC2 temps. From "Combined cycle journal" first quarter 2005.

<

http://www.fnal.gov/directorate/OQB... Industry Alert P91T91 code changes asme.pdf>

 

[doses]

Davefitz and unclesyd:

Thanks for the answers. I download the articles above, and I have added all the requierments mentioned to the specification of wye fittings in the project.

I undestood why Ni+Mn are so important fo temper temperature. As they are "gammagenos" (I don´t know the english word) elements, they will lower AC1 temperature. However, above is written Ni+Mg, is it correct?

Again, thanks a lot

 

[stanweld]

Ni, Mn & N all lower the transition temperature. Because Nitrogen is esential to produce the metal carbo-nitrides essential for high temp creep properties, Ni + Mn are the elements controlled to assure that the transformation temperature will not be exceeded during PWHT. Exceeding the transformation temperature essentially renders the material equivalent to P9 alloy steel )approximately 1/2 the design allowable stress of P91).

Weld fabricated Wyes should be avoided; if used, they should be normalized and tempered after manufacture. They must be fully evaluated with calculatios supporting their use. In one plant's service failures the weld reinforcement required in the highest strained zone had effectively been removed on the ID side of the fitting by the manufacture for flow reasons. Also type
IV creep cracks were observed in the HAZ in the longitudinal portion of the weld.

 

[doses]

Hi, again:

Stanweld: I have seen some industries and manufacture processes (My previous job was performing welding inspection and NDT, PT and MT mainly). But, I had never seen how fittings are made. ¿how is a wye made? ¿Can you describe the process in a simple way?.

By the way, if my questions seems basics for this forum, is because I graduated 3 years ago, and in my actual job in Piping department this is my 5th month. So, as you can see, I'm a very new "piping engineer". But, I read and learn everything I can, that's why I'm here asking.

 

[rmw]

This is the place for asking.

rmw

 

[MJCronin]

doses...

Forging is certainly the way to go for high pressure piping components like "wyes". A "weld fabricated" tee or wye is a component that is classified as an "un-reinforced fitting in ASME B31.1 and should only be considered in low pressure, low stress level applications.

Additionally, there is an ASME publication that addresses some of the issues stated above.

http://scitation.aip.org/getabs/ser...0127000002000197000001&idtype=cvips&gifs=yes.

Mr Jeffrey Henry (of Alstom Power, TN) seem to be the most knowledgable in this particular field.

My personal favorite shop is Ram Forge out of Navasota Texas. They have been in business a long time and have specialty forge the specific type of wye you seem to be interested in.

Let us know what you finally decide...

-MJC

 

[unclesyd]

doses,

Here is a paper on welding the creep resisting alloys though not specifically about T/P91 there is some good information about welding same. One point taken in particular is the requirement that the weld has to be cooled below 100C to complete the transformation of Martensite.

http://www.msm.cam.ac.uk/phase-trans/2005/LINK/182.pdf

 

[stanweld]

doses,
Wyes can be weld fabricated from pipe and forging components or fully forged and machined. Full contour machining can be done or partial machining can be done. F91 a forgings are normalized and tempered prior to machining. As stated previously, forged and machined wyes should be purchased.

doses/MJCronin,
Jeff does not work for Alstom. He now works for Structural Integrity Associates (

http://www.structint.com);

he's still in Chattanooga - visited with him a couple of weeks ago. Jeff/SIA provides seminars on enhanced P/T91, P/T92, P/T23.