ASME B31.1 weld type question 2

[xjboonie]

Quick question: what type of weld is this (girth, longitudinal, branch, attachment, etc.)?

Carbon steel 1” semicircular plate, full penetration weld, inside of carbon steel tee, 2” wall thickness. The code is ASME B31.1. Process fluid is steam.

Thanks!

 

[davefitz]

Not sure, but fluid eleastic vibrations will eventually rip the plate off and it will plug downstream equipment. Initial crack will be from thermal stress, and vibrations will complete the failure. Several famous accidents were caused by similar constructions ( eg, LMFBR ).

"...when logic, and proportion, have fallen, sloppy dead..." Grace Slick

 

[MJCronin]

Boonie

If we can assume that your device is meant to modify flow within a steam system, why do it this way ?

Are there no steam valves available in the area where you live ?...... New or used ???

Cheapest solution would be an RO contained within a new set of flanges ???.... Perhaps ?

Steam systems do not operate the same way as liquid systems ....


MJCronin
Sr. Process Engineer

 

[Dhurjati Sen]

double sided fillet weld

DHURJATI SEN
Kolkata, India

 

[gtaw]

The original post said it was complete joint penetration, so it can't be a double fillet weld. Fillet weld must be fused to the root, but they are not required to penetrate beyond the root into the joint.

I don't believe the post was asking whether the weld is a fillet or a groove weld, but whether it was circumferential or longitudinal or girth, etc. Circumferential and girth welds are typically butt joint. This application appears to be simply a groove welded T-joint with reinforcing fillet welds using AWS terminology. Still, I don't think that was the real question in the original post.

Best regards - Al

 

[weldstan]

It's a full penetration non pressure attachment weld. Of all the numerous B31.1 power piping designs I've seen, this one is definitely new to me.

 

[KevinNZ]

Putting the weld prep in to the wall of the tee does not seem right. Why not prep the edges of the plate?

 

[xjboonie]

Several famous accidents were caused by similar constructions ( eg, LMFBR ).

can you provide any specifics on those accidents?

If we can assume that your device is meant to modify flow within a steam system, why do it this way ?

yes, it is to straighten steam flow and reduce pressure fluctuations. This is one tee in a header with 3 inlets and 2 outlets. It is an existing system.

Putting the weld prep in to the wall of the tee does not seem right. Why not prep the edges of the plate?

that's my bad on the sketch part - the plate is prepped with a double bevel grind, while the interior of the tee is just cleaned up. My poor attempt to show the full penetration weld.

 

[MJCronin]

yes, it is to straighten steam flow and reduce pressure fluctuations.

Does anybody else out there in "eng-tips" land buy this ??

No, Steam does not act that way... The piece you describe will not "straighten flow" or reduce pressure fluctuations.

How do you know that your flow is not straight ?

What are the steam operating pressures and temperatures ? What is your steam velocity at the tee and is it within normally accepted limits ?

MJCronin
Sr. Process Engineer

 

[davefitz]

KevinNZ:
the Fermi I LMFBR had a failure caused by a fluid distribution plate break free and plug a liquid sodium cooling channel. Likewise the Monju LMFBR had a sodium coolant leakage failure when a thermocouple cracked at a partial penetration weld and broke free due to fluid induced vibrations. In both cases the initial crack was due to thermal stress, and high cycle fatigue from fluid-elastic vibrations led to complete failure.

More common is the failure of fossil fuel power plant spray attemporators, usually interstage attemporators between the 2 final superheaters. Both the spray nozzles and the header liners are vulnerable to thermal shock cracking if they are manufactured of stainless steel. The correct design that does not fail is to use inconel or P91 nozzles and liners , and the liner should be kinetically formed to the header ID ( "detna formed") using explosive detenation cords.

For the monju thermocouple, a full penetration weld would have extended its fatigue life by a factor of at least 20.

One systemic problem with major power engineering desings is the lack of awareness of the issues related to thermal stress, fatigue , and fluid elastic vibrations. Exactly zero colleges on the planet called earth teach these areas during undergraduate studies, and thus the design engineers are clueless until after failure of the part and then read the forensic metallurgical analysis .

"...when logic, and proportion, have fallen, sloppy dead..." Grace Slick

 

[MJCronin]

Amen,....davefitz.....

I have been involved with the design, analysis, specification and testing of steam and other high energy systems for over 40 years

I have never seen such an internal device on a steam piping system before ......

MJCronin
Sr. Process Engineer

 

[KevinNZ]

Hi Davefit

I was questioning the design of the weld of the plate to the inside of the tee. Seems not a good idea to cut a groove in the fitting and then fill this with weld and then weld the plate. Why nor prep both sides of the plate only.

 

[Mateeng]

Quick question: what type of weld is this (girth, longitudinal, branch, attachment, etc.)?
Answer: Tee Joint weld /Non pressure part attachment joint

 

[DekDee]

Mateeng,
How can it be a " non pressure part " when it is welded directly to the pressure retaining wall of the Tee ?