Hot buried Oil Lines 3

[Thomasmac]

I'm in the throws of designing a triple line buried pipline to run insulated and heat-traced at 120 C. The lines are only 1600m long, 2x4" and 1x6". I would like some information on how best to approach this line as there are probably a few schools of thought out there. We have the AutoPipe Plus software and intend to use this to determine stress levels at bends, although I'm not entirely confident that the results will reflect reality. The route has us crossing a few minor gullies (ie 4m deep) and a couple of horizontal bends along the way which would tend to suggest the pipe could expand into the bends, the insulation however (MDI) only has a compressive strenght of 300KPa so we risk crushing this with too much expansion. Im expecting 700mm of expansion per 500m.

 

[BigInch]

I assume you are past the point of thermal analysis and finding the thermal heat loss properties (heat loss shape factor) of the 3-pipe configuration within the ground and are now concentrating on limmiting stresses within the installed pipeline.

If the insulation is not rigid, it should be encased within a casing pipe. I have done uninsulated hot oil lines to 190ºF without any insulation other than that provided by the soil. If you can take the energy consumption hit, its the easiest, if not, better to encase it. If you cannot encase it, you must protect the insulation from crushing at the bends somehow. Honeycomb blocks added to the backfull?

Its tricky with a hot buried line. Yes, at that temperature you can/will have problems at both the horizontal and vertical bends if the alignment is not very carefully controlled. Virtual anchors (wherever a local pipeline segment has no relative movement with the soil when both are heated up) are set up at various locations along the pipeline. The pipeline tends to displace away from those points. They will be for example, at horizontal bends and at underbends at the bottoms of a slight valley in the alignment. In the vertical direction, the pipe will expand from each valley towards the overbend at the top of the peak inbetween valleys. At the peak, you may experience sufficient uplift to raise the pipe off the bedding. In extreme cases, it may be sufficient to pop the pipe up out of the ground. Generally I have found that you should limit vertical bends to a maximum of 5-7º to avoid problems. In some cases it will require extra cover above the pipe at the high points to maintain enough soil load above to hold the pipe down in contact with the bedding.

At horizontal bends, out-thrusts will tend to move the pipe into the side of the ditch and in the process, it will tend to climb up and out of the ditch there as well. Horizontal bends should allow sufficient movement to avoid such or risk crushing any uncased insulation and lifting the pipe out of the trench. Extra depth may be required at those bends to accquire enough soil load on top to hold the pipe down.

Pipe buckles are also a possibility as the pipe heats up, especially at overbends and underbends where the pipe has little lateral stability (only provided by soil load on top) and already has a bowed shape, the pipe will have a natural tendency to buckle upwards, if the soil load on top is not sufficient to maintain the profile as buried.

It should all be a careful balance between axial expansion loads, +/- uniformly distributed soil load on top, resisting soil load due to bedding below, critical buckling load and actual buckling load. Possible uplift off the bedding and span length must be limited at the overbends.

Be very careful at how you approach the "minor gullies". With a hot line, they are not minor. The need to control bends to within 5 to 7 degrees will require backing well off and beginning a slow drop to pass below the gullies, so that will probably be at least a 5 to 6 meter cut, if no scour is predicted. If you try to place the pipe in a "U "configuration, sidewall failure and pipe punch out at the gully wall may be possible, if not you could certainly overstress by bending at the bottom of the U with all the axial growth squeezing in at the top.





BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[BigInch]

Another bit,

Watch the melt temp of whatever insulation or coatings you might eventually decide to use. Many coatings won't reach that temp and some that say they can don't yet have the experience to prove it.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[Thomasmac]

Very helpful thanks BigInch,

We are now a little further down the track and have completed the soil modeling with Autopipe. We have come up with criteria for max bend angle and bend radius based on allowable pipe and soil stress levels and a full topographical survey is being carried out to enable us to acurately predict bend angles. Our final sticking point is how to protect the insulation from crushing as we fear that with the pipe runing at 120C and the outer casing rated to only 68 we may melt the outer casing if there is degradation of the insulation. Due to the pipe being restrained by the soil in the axial direction our modeling shows that we should be expecting around 25mm of displacment at the corners. Current school of thought is to wrap the pipe in a compressable closed cell foam at the corners (similar to your honey-comb suggestion), this should enable the pipe to move underground but not crush the insulation. Thoughts on this? .

 

[BigInch]

Be very careful with the calculations and the prediction of the insulation temperature. Don't forget that the maximum insulation temperature depends on the hottest temperature that the ground will ever experience, not the coldest. Offhand, I would not expect insulation to remain below 68, but perhaps I do not understand the exact layering system you are planning. You obviously have a carrier pipe, but do you have a casing with spacers keeping the carrier pipe centered within the casing, then insulation outside of that, then an exterior waterproof wrapping or coating for the insulation???

It also looks like to me that it is possible to have about 1mm/linear meter of pipe in expansion (assuming steel and 100ºC dT), so that a 25mm limit would say you have a virtual anchor at the center of 50 meters of pipe that expands 25mm in each direction from that. Soil shear values usually need about 200-250 meters of linear straight pipeline contact to develop a soil virtual anchor, so 25mm that may not be enough of an expansion allowance, unless the expansive directions are carefully controlled (with underbends placed every 50 meters with an overbend inbetween).

Since you have a rather short length of pipe run, not 100s of miles, you may be able to use foam type backups at bends, but also expect a maintenance price to come with that. Foam boards are commonly used to provide movement allowance for seismic action near fault lines, so I would say it will work.

The biggest problem I see is the maximum coating temperature allowable.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[Thomasmac]

Thanks again Biginch.

Our pipeline coating/insulation is comprised of the following. Pipeline coated with 300um high temp FBE for corrosion protection then a 40mm layer of closed cell MDI injected foam with an outer PE casing (the insulation is factory applied with the outer PE casing being centred around the pipe and the insulation injected in). The joints will be coated with the same system on-site and all bends will be field pulled to around 80D radius. You made a suggestion that we should concider going no more than 7 Degrees on overbends,is this per 12m length? Is this because soil could subside beneith the pipe if lifts off it's bedding?

 

[BigInch]

That's to minimize the uplift at the peak of the overbends. With more than 6 or 7 degrees (may be more or less for your temperature and pipe diam/wt) we found that lift off at the overbend would probably occur if we exceeded that figure. You could increase burial depth at the overbend to minimize that possibility, however the additional soil weight would also tend to increase the stress in the pipe. Since its not a long line, you don't have to be so worried about the extra cost of digging deeper, so increasing the bend and the depth could offer an alternative solution for your particular case.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com