Welding material of different thickness

[MTPipeliner]

We've been having some discussions about welding materials of different thicknesses and can't come to a conclusion, so maybe someone here can help.

The instance we've been talking about is welding a sleeve onto a carrier pipe. The carrier pipe is .250" X-70. The Sleeve is .750 X-52.

The argument is whether the top edge of the sleeve needs to be tied in or if the leg of the weld should only be as long as the carrier pipe is thick.





_______________________
/| Sleeve
/ |
/ |
/ |
____/________|________________________
Carrier Pipe
______________________________________




_________________________
| Sleeve
|
|
/|
_____________/__|_________________________
Carrier Pipe
__________________________________________

The argument for the first diagram is that the top edge of the sleeve should be tied in to get the full strength from the sleeve.

The argument for the second diagram is that the shorter leg of the weld causes less stress on the carrier pipe than the longer leg will.

Let me know what you think.

Pat

 

[metengr]

In general, fillet welds exhibit poor fatigue resistance because they act as stress risers. Also, adding reinforcement to increase the load carrying capability of fillet welds serves no real benefit, and will result in reduced fatigue properties.

For your particular application, sizing the fillet weld leg length to match only the thickness of the carrier pipe wall will result in added stress concentration to the weld joint because of a notch effect.

To avoid having a notch, I would extend the fillet length to match the thickness of the sleeve to provide for a tapered transition. You can offset the fillet leg length on the carrier pipe side to reduce the amount of weld metal needed to form the fillet.

 

[metengr]

Just one clarification, the recommendation for extending the fillet leg, mentioned above, assumes cyclic service conditions (pressure and temperature).

For static loading or non-rotating components (no fatigue loading) you could use a fillet leg size that matches the thickness of the carrier pipe.

 

[GRoberts]

A properly tied in fillet weld will not create much of a stress riser in the thicker sleeve, so I would go for as little weld as possible to match the strength of the pipe. The main stress riser in fillet welds is at the root of the weld, not the toe.

 

[metengr]

GRoberts;
Sorry, but I have to disagree with your post. Fillet welds have typically higher strength reduction factors because of non-uniform stresses versus full penetration butt welds.

The source of fatigue crack initiation and propagation may or may not occur at the weld root versus weld toes. This is the reason why the NBIC does NOT allow the use of patch plates externally fillet welded to pressure retaining items as a repair.

 

[GRoberts]

I guess I could have worded it a little different, such as "A properly tied in fillet weld will not create much of a stress riser in the thicker sleeve at the toe as compared to the root." Any stress riser created at the toe of the weld will be small as compared to the stress riser at the root of the weld. If NBIC does not allow fillet welded patches, it applies to any fillet welded patch, whether the fillet weld extends to the full thickness of the patch material or not. Thus even if the fillet weld leg does extend to the top of the patch material (and thus eliminate any chance of a stress riser on that toe of the fillet weld), the stress riser at the root of the weld is significant enough for NBIC to dissalllow them.

I would never disagree that a butt weld has much better fatigue and strength factors than a fillet weld.

 

[sayee1]

A few questions:
1. Why is the sleeve of X52 while the carrier is of X70 material(Basically, this would be the main contributor to the big diff in the thicknesses)? Is the sleeve a load bearing member or for say anode fixing?
2. What is the application and what is the code for fabrication (is it pipelines?)
3. Is it possible to taper the ends of the X52 sleeve with a 1:4 transition to meet the minimum leg length/throat requirements of the fillet weld as per the strength calculations?
4. What is the minimum throat of fillet weld required as per the design calculations and what is the consumable proposed for welding the materials of diff strengths?

Thanks and regards
Sayee Prasad R
Ph: 0097143968906
Mob: 00971507682668
email: sayee_prasad@yahoo.com
If it moves, train it...if it doesn't move, calibrate it...if it isn't written down, it never happened!

 

[MTPipeliner]

1. The Pipe is X-70 because it is an existing pipeline. The fitting is X-52 because that is what the vendor could supply with out having to make a custom fitting. Load bearing in the sence that it will contain the full pressure of the pipeline.

2. The sleeve is for a hot tap. It has a 8" 600 ANSI flange that an 8" 600 ANSI ball valve bolts up to. The Design code is DOT 192, ASME B31.8

3. When I ordered the fitting I didn't realize that it wouldn't be beveled at all, but it may be possible to order them with a bevel in the future.

4. Don't know off hand.

 

[sayee1]

As a first pass comment, based on the replies given to the queries, since the sleeve is load bearing and for a hot tap, the full strength of the sleeve is required for meeting the requirements of the pipeline.

The point is that you will surely have had to do an engineering calculation for the sleeve to have obtained the minimum thickness of the sleeve. Exterpolating the same calculatons, you should be able to arrive at the minimum throat calculations for the fillet and work out the best possible leg lengths in this case. The consumable you will be using for the fillet weld will surely be LH with a strength in between the 52 and 70 grades. So you also should be able to use the actual strength of the weld metal/consumable to your benefit to arrive at a lower leg length if the client/code does not restrict the same.

Thanks and regards
Sayee Prasad R
Ph: 0097143968906
Mob: 00971507682668
email: sayee_prasad@yahoo.com
If it moves, train it...if it doesn't move, calibrate it...if it isn't written down, it never happened!

 

[unclesyd]

Just curosity what is the process conditions inside the pipe?

 

[MTPipeliner]

The welding rod in this case is a 70 grade LH rod.

The pipeline is flowing natural gas at about 55 deg F and a velocity of about 7-8 ft/s.

Thus far, I haven't seen anything in the code that would prohibit a weld that didn't tie into the top of the sleeve.

It appears that there are different points of view on this page just as here in the office.

If the throat of the weld meets the minimum requirements, is there any benefit to tying in the top edge of the sleeve?

Can the full strength of the sleeve be achieved without tying in the top edge?

Pat

 

[stanweld]

The minimum fillet size should be calculated because the X52 material is fully load bearing. It is doubtful that the fillet weld size need be as large as the thickness of the X52 because the sleeve/fitting has obviously been reinforced. Based on the diameters and ratio of yield strengths, I suspect that the fillet size should be 3/8" to 1/2".

 

[sayee1]

"Thus far, I haven't seen anything in the code that would prohibit a weld that didn't tie into the top of the sleeve.
If the throat of the weld meets the minimum requirements, is there any benefit to tying in the top edge of the sleeve?

Can the full strength of the sleeve be achieved without tying in the top edge? "

Not tieing into the top of the sleeve would mean you have a sudden transition and this could be detrimental from stress distribution point of view. Hence, as a minimum, even if you wanted to avoid this, I would as a good practice ensure a taper on the sleeve end (1:4) to ensure a smooth transisiton, provided the design calculations allow me that change in the thickness of the sleeve.

Thanks and regards
Sayee Prasad R
Ph: 0097143968906
Mob: 00971507682668
email: sayee_prasad@yahoo.com
If it moves, train it...if it doesn't move, calibrate it...if it isn't written down, it never happened!