Effect of hydrotest on component stress 3

[CWicker]

What is the effect of hydrotesting a steel vessel to pressures ranging from 1,2 to 1,35 times the design pressure in terms of approaching or exceeding the yield strength? Can limited yielding be expected? If a vessel has been hydrotested to 1,35x design pressure at commissioning and then periodically thereafter to 1,2x design pressure, will there be any adverse effects on total stresses, or will the periodic hydrotests have no effect at all?

 

[metengr]

davefitz;
I read your post above carefully. I am not sure I follow your rationale regarding "higher fraction of yield strength". If you review Section II Part D, Appendix 1 it is my understanding that the criteria for calculating the allowable stress value is based on the LOWER of the following in 1-100 a);

1. minimum tensile strength/3.5 (this is the change from 4 to 3.5)

2. tensile strength at temperature /3.5

3. 2/3rds of the yield strength (there is no change here, it has always remained at 2/3rds * YS).

4. 2/3rds of the yield strength at temperature.

The allowable stress will be governed by tensile strength/3.5 because this would be the lowest stress value up until creep deformation. At temperatures of creep deformation, the allowable stress value is obtained in part b) of 1-100.

Since the hydrotest does not permit exceeding 90% of the yield strength of the component AT AMBIENT TEMPERATURE, you can still safely hydrotest at 1.5X for Section I components using the revised allowablestress value because the 2/3rds yield strength of the material has not been changed.

Also, since Section I refers you to Tables 1A and 1B, there are no cautionary statements that require you to use a lower hydrostatic test pressure, as permitted in Section VIII. Section I still likes the conservatism of 1.5X.

In my view unless someone can explain to me otherwise, you can still take advantage of the higher allowable stress values for designing Section I components and still safely hydrotest at 1.5X because the yield strength criteria was revised.

 

[metengr]

The last statement in my post should be corrected to read - ... because the yield strength criteria was NOT revised.

 

[davefitz]

Meteng:
I had thought the allowable strength was revised due to a different fraction of the yield - I do not have a sect II handy, but if the latest wordings have revised allowables based on ultimate/3.5 , then you are correct and I am wrong-and I do not yet see the rationale for lowering hydrotest to 1.3 times allowable,unless it is based on exceeding ultimate at local stress raisers. In that situation notch toughness or RT ductility becomes decisive in avoiding failure during hydrotest.

For steels where the allowable is based on yield, then 1.5 times 2/3 yield at RT is of course 100% yield during hydrotest for simple shells without stress concentration. If there is no alloy for which the allowable is based on 2/3 yield, then it is curious that ASME would state it as a criteria. For local areas with nozzle or penetrations, the local concentration factor may be on the order of 3.0 ( can be as high as 5 at crack tips), but modern finite element analysis can give better estimates than these old estimates. With a local stress concentration above 2.3, one can exceed ultimate at the local stress riser with a 1.5 hydro.

 

[ggarciae]

As stated by metengr allowable stresses are higher because of minimum tensile strength safety factor.
But what happens when designing at room temperature and maximun allowable stress is minimun for 2/3rds of yield strength? (I really don't know if this may occur for any material) If test pressure is 1.5 times you would be reaching yield stenght.
European norm EN 13445 takes this criterium. Safety factors for carbon steals are tensile strength/2.4 and yield strength/1.5 for design conditions and yield strength/1.05 for testing conditions. Minimun required test pressure for formula not taking into account temperature is 1.43 times (there's another formula taking into account temperature 1.3 Sa/Sb). For a designed vessel for room temperature with just the needed thickness and where the safety factor for yield strength is minimun Pservice/Ptest(1.43*Pservice)=(S/1.5)/Stest => Stest= S/1.05


Regards from Barcelona
G. García

 

[jte]

Folks-

Interesting that nobody has discussed where the 1.3 value for the hydrotest multiplier came from. See thread794-94499 for a discussion of the hydrotest basis. The hydrotest multiplier was simply changed by the ultimate tensile ratio change from UTS/4 to UTS/3.5: (3.5/4)*1.5 = 1.313.

For materials governed by tensile this keeps the hydrotest stress the same. For a vessel fabricated to an S=17.5 ksi under the old allowables, and assuming no temperature adjustments for simplicity, the hoop stress at hydrotest conditions will be 26 ksi. The same vessel, but now thinner due to using the new 20 ksi allowable stress will see a hydrotest hoop stress of... 26 ksi. Simply put, while the allowable stresses were increased, the hydrotest stresses were kept the same.

jt

 

[GenB]

In this case the NBIC takes over.- When I work on used boilers or PV, I like to test atMAWP, my AI who isa Guru PE,ME and has other few U diplomas makes me test at original required test of the Code year and add when built.
So it ishard to do your own thing, I will not fight my AI for a little thing,even if my believe is different.
ER

 

[pipewelder1999]

I just left a project two days ago in which various components were replaced including roof tubes and superheater tube sections.

The areas we worked in the superheater were near the bottom of the loops. After the work was done the hydrostatic test pressure of 1400 PSI was reched on a 1250 PSI boiler.

The upper row of welds tying the tubes together consisted of S Straps and flare bevel welds at the tangent section of the tubes.

A leak developed adjacent to one of the welds at the heat affected zone. About two more days were added to the outage.

The superheaters had been in service prior to this outage with no problems in this area.

No work was performed in this area during the outage.

During rewelding of the tube tangents addtional cracking was observed in adjacent tubes which was removed and rewelded.

The cracks were typical underbead cracks based on what I observed while working on them and in my opinion existed in other locations.

The boiler was then tested to 1400 PSI again but with a MUCH slower build up of pressure.

As was stated by my welding partner the new tubes and welds were saying "Come on up, Come on Up" as the presure was buiding. The existing material was screaming "HOLD THAT!, HOLD THAT!"

My limited understanding of stresses that are present in circumferential butt welds compared to the hoop stress present in the base material makes me wonder why it is so important to exceed the operating pressure during testing.

Gerald Austin
Iuka, Mississippi

http://www.weldinginspectionsvcs.com