mechanicalmark
Mechanical
Is it acceptable industry practice to weld slip on flanges to a long radius butt weld elbow? I am specifically talking about 8" and 16" 150# B16.5 A105 flanges. We typically use weld necks with butt weld elbows.
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations MintJulep on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Welding Slip on Flange to Long Radius Elbow 4
- Thread starter mechanicalmark
- Start date
- Status
- Thread starter
- #3
mechanicalmark
Mechanical
No codes are required. Piping is subjected to vacuum and is rated for 15 PSIG. However, would it be accepted in Code service also?
-
1
- #4
To answer your question it is not acceptable industry practice to weld a SO flange onto an elbow.
In the extreme some companies will not allow you to weld a BW fitting to a BW fitting. You have to have a short length of pipe between the fittings.
In the extreme some companies will not allow you to weld a BW fitting to a BW fitting. You have to have a short length of pipe between the fittings.
-
1
- #5
Theoretically you need a straight piece of pipe to weld the SO, true, an elbow section is unround and theoretically you can't slide the SO over the elbow. But in practice often you can. Then you don't have the (exact)required root openings but that is not a problem, not in strength and not in lining up. After all there are still a couple of mm (depending on the size of the piping system) left between the face of the elbow and the face of the flange.
I don't have a problem with welding a SO to a elbow.
Good luck
I don't have a problem with welding a SO to a elbow.
Good luck
SO flanges are welded to long radius elbows. It is a little more problematic to try to weld them to short radius elbows, because of the geometry, but that has been done as well. As to whether it is acceptable, it would depend upon the system you are doing this for. I would consider it a practice generally to be discouraged, but if the circumstances warrant it, and if the thermal expansion stresses are low,do it.
-
1
- #7
For a B31 code compliant calculation, how do you determine the flexibility factor k, and stress intensification factor i? The calculation is prescriptive in the code for a weld neck flange connected to a bend; however, I am not sure what you guys use when a slip on is welded to a bend.
If there was not enough room, I would consider using a short radius bend with a weld neck before a slip on welded to a long radius bend.
mechanicalmark, what is the motivation for wanting to use the slip on welded directly to a long radius bend?
If there was not enough room, I would consider using a short radius bend with a weld neck before a slip on welded to a long radius bend.
mechanicalmark, what is the motivation for wanting to use the slip on welded directly to a long radius bend?
- Thread starter
- #8
mechanicalmark
Mechanical
CRG, the motivation for using SO flange with LR BW Ell is our customer has it shown that way on their facility drawing. There is not enough room for the Ell and a WN. We were suppossed to supply this item, now the customer's A&E firm states they will supply this item.
Note, we perfer to use the long radius elbows to minimize pressure drop in this high velocity vacuum service.
Note, we perfer to use the long radius elbows to minimize pressure drop in this high velocity vacuum service.
For most piping systems, I have found very little impact specifying a short radius bend in a tight spot. Granted, they do have a cumulative effect on pressure drop (pressure rise in your case); however, I have never designed a piping system so critical to flow that the addition of one short radius bend made any real difference. Roughly speaking, the pressure drop of one short radius bend is equal to two long radius bends.
When corrosion/erosion is an issue, a short radius bend does accelerate the problem. So in these cases, I would not use a short radius bend. Instead, I would use a target tee with a weld neck flange.
Only after concluding that there were no reasonable alternatives, would I use a slip on flange welded directly to a bend. I have a problem giving you a specific reason why; however, I doubt that the bends have been tested by the manufacturer in that configuration. And, I avoid recommending equipment or components for applications they were not tested or designed for. It is just my cautious side reflecting the litigious nature of the world we live in.
Just my two cents worth.
When corrosion/erosion is an issue, a short radius bend does accelerate the problem. So in these cases, I would not use a short radius bend. Instead, I would use a target tee with a weld neck flange.
Only after concluding that there were no reasonable alternatives, would I use a slip on flange welded directly to a bend. I have a problem giving you a specific reason why; however, I doubt that the bends have been tested by the manufacturer in that configuration. And, I avoid recommending equipment or components for applications they were not tested or designed for. It is just my cautious side reflecting the litigious nature of the world we live in.
Just my two cents worth.
-
1
- #10
As far as SIF or k values goes, most of the ones you see in the B31 codebooks came from experimental work done by Markl and his compadres. You would have to review their material to see how they got those values and to make a judgment as to whether you could apply them in this case.
Another thought - B31.3 only gives you 30% (I think) of the fatigue strength for a SO flange vs a WN. This is because a SO flange is welded using a fillet weld and not a full pen groove weld ("buttweld"). Also, there is less NDE available for this type of joint vs a buttweld, so you can't verify the integrity of the joint, and correspondingly the codebook kicks you in the butt for using a SO flange. If you have any loads transferred to this joint in the form of reactive moments or forces due to pipe supports, equipment nozzles, seismic loads, wind, etc., I would be VERY careful when using this setup. Make sure you have identified all your load cases and anticipated operating scenarios, e.g. startup, hydrotest, seismic event, steam-out, operator standing on the pipe, etc. Any one of those may overload this connection. If I were doing this myself, when setting the allowable for this connection, I would probably derate the allowable by a goodly factor relative to the allowable stress for a standard WN to BW elbow.
Most of the owners of chemical plants/refineries/oil fields that I've worked with will not allow the use of SO flanges, period, in any service, even one such as this, without a rigorous analysis showing that the use of a SO in any proposed application is OK, and that usually means an FEA.
So in this case, you have a SO flange welded directly on a LR elbow. During fit-up, due to the curvature of the ell you will have a larger gaposis between the SO flange and the ell on the ID of the ell vs the OD of the bend, this being caused by the need to square the bevel on the ell with the groove on the bore of the SO flange, i.e. align the CLs of the two fittings. They have to be squared (1) so the CL doesn't get screwed up and (2) so the end of the ell doesn't interfere with the gasket on the companion flange. This means more weld metal has to be deposited by the welder on the OD side than on the ID side, which provides more opportunity for weld defects since the welder is filling a larger gap on the OD. The additional weld metal may also distort the elbow but I don't think that will add up to be a problem in this case. It might look funky though.
You might talk to an experienced welder/pipefitter to get his/her opinion on this.
Lastly, you can buy LR ells with straight tangents on them and that will get you out of this problem. Or you can have them formed (bent) from straight pipe.
Thanks!
Pete
Another thought - B31.3 only gives you 30% (I think) of the fatigue strength for a SO flange vs a WN. This is because a SO flange is welded using a fillet weld and not a full pen groove weld ("buttweld"). Also, there is less NDE available for this type of joint vs a buttweld, so you can't verify the integrity of the joint, and correspondingly the codebook kicks you in the butt for using a SO flange. If you have any loads transferred to this joint in the form of reactive moments or forces due to pipe supports, equipment nozzles, seismic loads, wind, etc., I would be VERY careful when using this setup. Make sure you have identified all your load cases and anticipated operating scenarios, e.g. startup, hydrotest, seismic event, steam-out, operator standing on the pipe, etc. Any one of those may overload this connection. If I were doing this myself, when setting the allowable for this connection, I would probably derate the allowable by a goodly factor relative to the allowable stress for a standard WN to BW elbow.
Most of the owners of chemical plants/refineries/oil fields that I've worked with will not allow the use of SO flanges, period, in any service, even one such as this, without a rigorous analysis showing that the use of a SO in any proposed application is OK, and that usually means an FEA.
So in this case, you have a SO flange welded directly on a LR elbow. During fit-up, due to the curvature of the ell you will have a larger gaposis between the SO flange and the ell on the ID of the ell vs the OD of the bend, this being caused by the need to square the bevel on the ell with the groove on the bore of the SO flange, i.e. align the CLs of the two fittings. They have to be squared (1) so the CL doesn't get screwed up and (2) so the end of the ell doesn't interfere with the gasket on the companion flange. This means more weld metal has to be deposited by the welder on the OD side than on the ID side, which provides more opportunity for weld defects since the welder is filling a larger gap on the OD. The additional weld metal may also distort the elbow but I don't think that will add up to be a problem in this case. It might look funky though.
You might talk to an experienced welder/pipefitter to get his/her opinion on this.
Lastly, you can buy LR ells with straight tangents on them and that will get you out of this problem. Or you can have them formed (bent) from straight pipe.
Thanks!
Pete
- Status
Similar threads
- Locked
- Question
- Locked
- Question
