Pipeline Valve Change Out

[PKEngineer]

We are in a situation of replacing multiple 16” 600# gate valves with ball valves installed on natural gas pipeline. Both type of valves has same face to face dimensions. Problem we are facing is that this pipeline is in a seismic fault zone and we have feeling that pipeline is under stress and there is a risk of pipe move when flanges are unbolted for this change out. Does anybody have recommendation/experience for this valve replacement issue? Response will be much appreciated.

 

[pmover]

expose the pipeline 200-300-ft from the ball valve in both directions. remove the valve and fastened one flange. align the other flange by maneuvering the pipeline to mate/properly line up the valve/flange. use sand bags or rock-free spoil to keep pipeline in place. once fastened and any testing is completed, carefully back-fill the exposed pipeline.

 

[PKEngineer]

Thank you pmover for your early response. I am little confused need of exposing the 200-300 ft of pipe. This line is buried but valves are above ground at valve stations which are planned to be replaced.

 

[gerhardl]

You have not described exactly how the the valve and pipeline layout is constructed above ground. Ideally the total construction should be solid enough not to crate skewness in alignment and stresses not allowed for bolts, valve and pipelines both in present situation, during change or afterwards under foreseeable earthquakes and earth movements

Normally valves of this weight and dimensions, and in this pressure class, should each be supported and anchored independently. Stresses in the pipeline should be handled separately as such.

See to that each valve has a separate and adjustabel foundation and anchoring. Additionally: How is the weight and layout difference between new and old valves? Operation and repair access? Adaptions necessary?

The pipeline must then be adjusted, if necessary, to avoid stresses from pipeline transferred to the valves. Possible means of adjustment: adapted sealings, distance rings, adjustment pieces, lockable and not lockable piping joints, support for pipe-ends and mechanical adjustments etc.

 

[PKEngineer]

Thank you gerhardl for a comprehensive response. Photo for above ground valve attached. Pipe on both hands of the valve is sitting on supports but looks like need better anchor and support with possibly shoe welded to the pipe on both ends in order to restrain pipe from moving during the valve change out. What is your thought?

 

[LittleInch]

That's quite a neat layout, but there is insufficient pipe to allow for any flexibility.

As noted above, you really need to excavate probably about 20m min ( allow for up to 30) back in either direction to allow you sufficient wriggle room to both allow you to remove the valve and re-fit a new one without huge amounts of force to get the flange faces aligned. Sometimes the forces will spring the pipe apart, sometimes they won't let you force the flanges apart enough to remove the valve without flange spreaders which will let you remove the valv, but are more difficult when it comes to re-fitting it.

The forces involved in preventing any movement could be in the range of hundreds of tonnes.

If you look closely the left hand side looks like it has lifted off the support already indicting some levels of stress/ movement / settlement.

This famous video gives you some idea of possible movements and forces....

https://www.youtube.com/watch?v=0z3FmgzTeK8

Just get a digger out there - trying to fix a stressed pipeline in position such that it doesn't move a mm in any direction is futile. IMHO.





Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[brimmer]

There isn't much you can do except provide as much stress relief as you can by excavating the pipe on either side. It will help. You can often see the pipeline physically move when you do this. You may want to be in the "safe" work mode when you do this if you really expect a lot of movement. I would expect if there was movement from seismic you may have already observed it, as the valve is the anchor point & this is where there will be issues, not in the unanchored underground pipe.

 

[PKEngineer]

Thank you LI. Do you know any vendor name who may provide flange spreader for 16" 600# flanges, if you happen to be in US. I understand risk of using spreader when pipe can sprung like your shared video, asking for info only.

 

[PKEngineer]

Thank you all for providing great responses.

We were trying to avoid excavation, if possible, in order to avoid permitting requirement as mandated for the area. But based on most of your folks recommendation, looks like better option would be excavating some pipe on both end in order to relieve stresses on the pipe which will also provide some wiggle room to change the valve in course. Any further thought?

 

[GD2]

PKEngineer,
It’s extremely difficult to make empirical assumption on how much forces and moments will be involved when the valves are released. It’s usual to see pipe lift off in such pipeline layout regardless of how good Stress is done. Is the knife gate valve also have a lift off from the ground?
One way to figure out excessive load could be to find out the historical data (if there is any) on the ease of operation of the valve. With huge piping load, torque/current would be in very high order. Comparing data’s from different valve stations, you will have an idea about which valves will be under more stress than the others.



GDD
Canada