Distortion when welding Slip On Flanges 1

[iam42]

Hi Guys,

I was wondering what novel ways you guys use to reduce angular weld distortion from welding a slip on flange. I have attached a pretty crappy sketch but hopefully it will give you a rough idea of what I am talking about.

The problem is mainly when welding Stainless Steel flanges that have a reasonably thick pipe.

http://files.engineering.com/getfil...ae7d-c9c12631c156&file=angular_distortion.JPG

Thanks

 

[gtaw]

I suspect the angular distortion should be minimal. Considering the sketch shows the angular deflection in the opposite direction of what will actually occur, the problem should be less than you expect.

Best regards - Al

 

[moltenmetal]

First off, your slip-on flange sketch looks remarkably like a weld neck flange. The hub on a slip-on flange is much less than the thickness of the flange through the stud holes.

Secondly: why are you using slip-on flanges with heavy wall pipe? If the heavy wall is required for pressure/temperature resistance, you probably do really want a weld neck.

The exterior fillet weld will just plain NOT distort the flange in any meaningful way. If the wall thickness of the pipe requires a massive interior fillet, as long as you do not attempt to do it in a single pass and allow the correct interpass temperatures, you should be fine.

 

[DekDee]

Agree with the previous comments.
What you can do is weld the inside first and then before welding the outside bolt the flange to another flange.
Leave fully tightened until after cooling and then see if it has made any difference.
Cheers,
Shane

 

[JAYDEE23]

I have experienced the same issue. I am interested in processes used or comments in this area also. I have attached a to scale code slip on process and result. This would be a worst case scenario for us.

 

[gtaw]

Consider changing the sequence of welding. Complete the fillet weld against the hub first, then complete the fillet weld on the inside diameter.

As per the sketch provided, welding the inside diameter first causes the weld to act as a fulcrum point. The second weld contracts once it is completed and causes the flange to rotate on the initial weld (the fulcrum point), thus the outer edge of the flange is "rotated" toward the fillet deposited against the hub.

The proposed sequence uses the strength of the flange to resist rotation around the initial fillet weld deposited against the hub. The second weld the fillet on the ID is going to contract as it cools and pull "inward". However, the mass of the flange will resist. The weak link is the relatively thin pipe wall. The pipe wall should be distorted rather than the flange. While there will be some distortion, there always is, it should be within tolerable limits.

Some times a sketch or two tells the story better than words alone. Let us know what your findings are.

Best regards - Al

 

[moltenmetal]

Absolutely you would weld the exterior fillet first! That's where you're going to tack it during fit-up, and it seems natural to me that the welder would start welding at the end their tacks were on. Didn't even occur to me that people would try to weld the interior fillet first. Perhaps this isn't on piping, but rather on a vessel?

 

[DekDee]

moltenmetal,
Welding the inside of the flange first is not only standard practice it is the most logical way.
Spirit level on the flange face vertically.
Tack top and bottom on the outside once flange is level with the pipe.
Rotate flange / pipe 90 degrees.
Spirit level on the flange face vertically.
Tack top and bottom on the outside once flange is level with the pipe.
You now have 4 x tacks on the outside and the flange is (hopefully) square in both planes.
Weld the inside first and then remove all four tacks on the outside.
Fully weld the outside without interruption.

The welder wants to weld bottom to top without interruption (stopping midway at a tack) or running over an incompletely removed tack that will leave a pronounced hump.
The other thing is if you are removing tacks whilst welding the outside first will it be affecting the squareness of the flange ?

In my experience distortion comes from the pipefitter being too lazy to "centre" the pipe inside the flange.
Easier to let the flange "hang" on the pipe which when tacked up leaves a big gap on the bottom (between the pipe OD and flange ID) and no gap on top.
You are then faced with unequal amounts of weld being applied.
Regards,
Shane

 

[moltenmetal]

Aha, I see. It depends on whether you're spooling pipe or doing field fits I guess. Your order of operation makes perfect sense if you're attaching a flange to a piece of pipe on a bench, whereas we're most often field fitting a spool and tacking in place. I hear you about the tendency not to center the pipe in the bore- there is some slop between the bore and the pipe OD which is handy for providing some fit-up tolerance, but it can be sloppily used by a poor fitter.

I've seen RFSOs warped before, but it really takes some serious heat input on the ID weld to do it. Never seen one of our own welders do it, and our subs are pretty good too- but I have seen some over the years so it's definitely possible.