Pipe stress/loads from installation 1

[NewENG102]

When a pipe line is installed to a pump flange there can be some due stress put on the pipe as wells the pump flange if there are misalignment in the pipe line. Can some expert here point me in the right direction on how I can come up with the load that will be put on the pump flange.

 

[KoachCSR]

It's not worth evaluating - have the subcontractor correct the misalignment at their cost.

 

[NewENG102]

@KoachCSR this is for an exercise- If you can point me in the right direction that would be great.

 

[RVAmeche]

/This could be done using pipe stress analysis software. As KoachCSR mentioned, it's kind of a trivial exercise if that's your only goal. To mitigate the issue, that's what flexible connectors (bellows type connectors) are used for.

Also note, your 25' section of straight run between the pump suction and pipe anchor (from your other thread) would still cause significant load on the pump suction nozzle even if it wasn't 1/8" misaligned. That 50F expansion with nowhere to go (anchor, pump nozzle, straight pipe) will always cause very high forces.

 

[NewENG102]

@RVAmeche Thank you for the information. I had a feeling best approach would be a stress analysis software.

 

[JJPellin]

I have not seen any software or other tool that can evaluate the forces resulting from all forms of piping misalignment. We have a specification for how parallel the pipe flange needs to be with the pump flange. If these are not parallel and are bolted up, a massive stress can be imposed on the pump. I don't know how to estimate that stress. But, ultimately, the answer is as KoachCSR stated. Have a spec for pipe alignment and make your pipe fitters adhere to the spec.

Johnny Pellin

 

[KevinNZ]

Pipe stress programmes can model the effect of pulling a misaligned pipe back to alignment. For example put an anchor at the other end of the pipe with displacement or rotation = the misalignment. Of course the movements need to be in the elastic range.

 

[XL83NL]

I have not seen any software or other tool that can evaluate the forces resulting from all forms of piping misalignment.

You could try to simulate such a misalignment using a cut short, but Im not sure how that would work, as it only allows displacement in 1 direction. At one time, we had to pull two flanges together with massive resultant forces. This was unacceptable. To determine the resulting stress to some extent (i.e. getting an order of magnitude to determine a go or no-go), I simulated it with AutoPIPE using the Imposed Support Displacement option (short ISD). I modelled an anchor at the flange that was misaligned. The ISD then allowed my to specify the 6 DOFs at the point under consideration (i.e. the flange), and specify which load case it has to be combined with. It may not have the exact result, but it gave sufficient information to determine a go or no-go (in this case, a no-go).

 

[saplanti]

From my pipe stress experience the pump flanges are very delicate against very small rotations, and of course displacements, due to the thermal loads, and the allowable loads on the pump flanges are not big either by comparing the pipe side flange allowable.

Therefore, you need to eliminate the misalignment at the flange connection, don't even think about how much rotation can be handled by the connection.

 

[DSB123]

XL83NL,
Any decent Pipestress program will allow "Cold Pull/Push" in three directions and also rotations as input.

 

[saplanti]

Pump side allowable loads (directional, and combinations) can be obtained from the pump manufacturer.
Manufacturers want to protect the shaft from additional deflections and loads which are induced by external forces and moments. Therefore they limit the external load. Some may be obtained from the applicable pump code.

As it mentioned above you may be able to use one of the pipe stress analysis software, or an FEA software to play with it to make the loads under the allowable.
1. You may implement, in the model, the available rotation and displacement to obtain the induced flange forces on the flange face to compare with allowable.
2. However the exact opposite of Option 1 is too difficult since you have almost infinite load options in 6 directions to meet the allowable.

The option 1 above will probably tell you to stop playing or it may be helping you what to do next in the second option.

This was the reason I would not recommend to find the rotation and displacement options can satisfy the directional and rotational allowable and allowable combinations.

Good luck.

 

[Riley15]

The link below may provide some insight.
Link