Repair of pipelines under crossings 1

[LittleInch]

I've got a peculiar situation which I would appreciate some input / experience.

We have a pretty thin (7.8mm) pipeline with a D/t of about 100, some 40yrs old with some identified corrosion defects under a road crossing.

Fairly uniquely in my experience, we are able to open cut the road to effect repairs as the owner is very reluctant to shut down and clear the pipeline to allow replacement / diversion.

The planned method was to use composite repairs on the multiple locations within the crossing where it requires it with some other non critical defects being subject to coating repair.

The picture is complicated by the pipe previously utilising a concrete sleeve which needs to be broken up to gain access to the pipe. We would like to then simply compact the back fill and do away with the casing or any other protective device (slab/half sleeve etc). According to API 1102, we're at the ragged edge (~0.95 of allowable stress) of acceptance, based on the original nominal wall thickness, even with a reduced MAOP from the original design.

My issues and concerns are:
1) API 1102 does not cover anything to do with defects to the pipe
2) With the number of repair sleeves in a relatively short section, I'm concerned about possible stress concentrations due to the additional loadings from the crossing
3) It is not clear whether composite types repairs, or indeed other repairs as listed in B 31.4, sec 451.6 are valid for crossings

My thoughts have swung towards recommending a single split sleeve with epoxy infill, but would greatly welcome thoughts and experience on the points above.

If you want any more info please ask, but I need to be circumspect in what I can put out there.

Thanks

LI

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

 

[bimr]

Not much information to go on. What type of street, traffic loads, construction restraints, etc. are missing.

Many areas now require no cut construction techniques such as HDD for a project like this. Based on the age of the pipeline, why ian't replacement instead of repair being evaluated?

 

[LittleInch]

It's a road with normal loads. Precise details not relevant here.

I agree replacement is a better bet, but owner doesn't want to lose throughout.

What I'm trying to find out is if the normal repair options are still valid for crossings or not.

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

 

[BigInch]

we're at the ragged edge of acceptance, based on the original nominal wall thickness, even with a reduced MAOP from the original design.

Cyclic stresses and somewhat unknown loading conditions.
Are you happy with being on the ragged edge?
I think you know what to do.

If that was my pipeline, I'd be
1) making you case it, even though I would normally avoid casings, or
2) hottapping, blocking and diverting to a new crossing.

 

[LittleInch]

BI,

I do know what I'd like to do which is chop it and but a thicker pipe in, but client says no.

Next best is to close sleeve the whole thing.

The point of my question is to see if anyone out there has actually done anything similar (repairs under crossings in an uncased situation) or has any good pointers.

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

 

[brimmer]

All crossing calculators I have come across do not take into account corrosion (wall loss). These calculators are all conservative by nature to start with though, you could try to using a reduced wall thickness (equal to corrosion depth) in the calculation assuming its not too deep, see if it passes as the entire area is reduced (although I guess as you mention you are already on the edge), or do for specialized engineering assessment that would take the corrosion into account.

No specific information or limitations are present for repairs under crossings, can use same repair type regardless of location in all the codes I see. I would not recommend clocksprings, given that I have seen installation go wrong, and these loosen over time. I would go for the split reinforcing or compression sleeve given that you will only have one opportunity to make a repair at this location and you want it to hold.

Not sure you are going to get enough stress concentration at the ends of these types of sleeves (fillet welded ends would be much worse), I guess it depends how much you think the ground & pipe will move in this area from the road traffic. I have made this type of repair under roads and used concrete slabs over top to distribute the load, with soil backfill/compaction requirements at the crossing. I have heard of modeling this scenario, doing FEA with sleeve configurations assuming you have enough info on the loading from the road crossing, but I have not done this type of assessment.

 

[LittleInch]

brimmer,

many thanks for the input - it more or less mirrors my current thinking. As soon as we take any of the defects if we took those as the min wt it fails 1102.

I hear what you say about the clock spring type repairs. the codes don't say anything, one way or the other about repairs under crossing, in part I guess because its rarely done?

The split sleeve is now under serious consideration, but practical issues and schedule are now coming into play.

I've seen FEA type work done on similar systems, but the time and cost is a lot more than just biting the bullet and replacing the thing. Compaction of the ground both above and below is reasonably straightforward, but the extra load during installation when the support isn't there is now starting to worry me. Short lengths mean a longer repair schedule. The road is a private one, but there are limits on how long they can have it open and financial penalties come into play if the repair takes too long.

It's not an easy one and if there are any more thoughts, please let me know.

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

 

[bimr]

Sometimes one needs to man up and terminate the client.

 

[BigInch]

Best advice yet.

 

[brimmer]

Well, if you can get your hands on slabs with the full footings so they don't touch the pipe and distribute all the load fast enough, then I guess you could feel more comfortable with a clockspring, though I've never had issues getting compression sleeves installed in a pinch, typically in my neck of woods I can get these installed faster than clocksprings

 

[EdStainless]

So he can't afford to shut it down to fix it but he'll shut it down in a few years when it fails.
Typical.
For every problem there is an answer that is fast, cheap, and wrong.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[LittleInch]

Brimmer,

When you say compression sleeve what exactly do you mean. I've got a fair idea, but I'm not clear how in practice you actually get The tightness that is required. I can't risk direct welding on this pipe.

Thanks for input so far to everyone.

I'm going to make a good play for replacement, but need to see what else is out there and how others have addressed this issue.

LI

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

 

[BigInch]

Probably like this one,

http://www.ogj.com/articles/print/v...-compression-sleeve-for-seam-weld-repair.html

 

[LittleInch]

Thanks BI - That looks very interesting and may be just what we need.

Is there any issue with SCF at the end of these sleeves if there is any bending stress?

Matthew

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

 

[BigInch]

Theoretically there are stress concentrations set up at any change of cross-section. The sleeves should terminate in regions of low stress to ensure that concentrations also remain lower than limits. Conservatively you might assume a continuous unsupported pipeline beam passing over multiple supports, in which case the bending moment at each support would be approximately -W*L^2/10 where Length L is distance between "hard" support points. Bending at each L/2 center span between supports would be about W*L^2/8 - W*L^2/10. As the moment diagram goes from positive at centerspan to negative at the supports, the region of very low to zero bending moment is at about 1/4 span from supports. If the sleeve terminated in that region, bending stresses there would be minimal.

 

[BigInch]

Of course if you had continuously good support, there would be very little bending stress at all anywhere.

 

[rconner]

If you are planning on breaking up concrete around a known weakened critical service pipe in a crossing location while operating at a very high % of YS, please use GREAT care. Have a good weekend.

 

[brimmer]

Like this one

http://www.tremcopipeline.com.au/pdf/petroline/petrosleeve_presentation.pdf

See detailed presentation. No shut down required. Tighten down with chains, you may have to lower line pressure temporarily depending on what they operate the line at now during sleeve installation. There have been no issues with high SCF that I am aware of with this type of sleeve (i.e. failures, reported issues), but you would expect some increased stress wherever a sleeve ends.

 

[lilliput1]

The pipe should be replaced entirely. Consider ramming pipe sleeve underground across the roadway then pushing in new pipe in sleeve from one end. There would be minimal downtime to connect to remaining good portion of piping at each end. Check youtube.com for video of pipe sleeve ramming.

 

[dhengr]

LittleInch:
Sounds risky to me that you would try to break a conc. sleeve around a weakened pipe away from that pipe without doing any further damage to the weakened pipe. Someone is going to run a breaking hammer a little to hard or a little to deep, or actually pull a bulge in pipe as they pull/break a chunk of conc. away from the pipe.

Instead of a new conc. sleeve, or some other more elaborate casing to protect the repaired pipe, consider what is called lean conc. as a trench fill. It is basically a weak grade of conc. about 1000-1500 psi compressive strength. Strong enough to be a very good fill when it comes to distributing concentrated compressive loads on the pipe; weak enough so it can easily be broken out at a later time, and quite economical in ready-mix form vs. real, strong structural conc.

You guys know much more than I do about ASME & API pipe codes and what you can and can’t do under these conditions. I’d want to replace this section of pipe so I didn’t have to block the road again in a few years because we found another defect in the same deteriorated section of pipe. When you have a number of defects to repair in a fairly short distance I would wonder that you might not create a new defect in an only slightly less defective area right next to one of your fixes. I’ve seen this a number of times in both structural and mechanical systems, where you need to fix this two feet of something, and in the process the fairly weak next two foot section goes to hell during the process.