Pipe deflection, bending, etc.

[paulengr]

How to analyze to determine how much deflection should be allowed in a pipe.

Normally, obviously less is better. However I am looking at miles of pipe and trying to determine as it goes up and down over elevations when I'm exceeding design limits and need to look at providing support.

This is for steel pipe. Obviously with HDPE as long as it doesn't kink you can get away with almost anything.

 

[BigInch]

You bend the pipe (cold bend, or hot bend) so that it conforms to and fits the supporting surface of the trench. If pipe is allowed to elastically bend into the trench in vertical curves, or is pulled into position around horizontal curves, bending stresses are developed and when the pipe is buried, the bending stress remains. That bending stress effectively lessens the amount of strength that can be used to resist pressure stress and longitudinal stress, therefore the allowable pressure that the pipe can contain is also lessened. Not effective use of the material.

I hate Windowz 8!!!!

 

[LittleInch]

What exactly are you trying to do? Surface lay over an undulating surface or burial?

As a general rule anything much over 600 D as an elastic bend is probably too much. You risk buckling the pipe or overstress the pipe as BI says if you go much more. An experienced site engineer can establish when you need to bend by a few degrees and anything more should be able to be seen from a profile.

PE is much more forgiving as it has a much lower strength and also creeps under high stress gradually following the ground profile even if there is a span.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[paulengr]

Surface lay, not burial. My mining engineering friends ignore funny little details like design limits. If they screw it up, it's just another day at the office. If I screw it up from the engineering department, it's the end of the world.

Need to elevate it about 4 or 5 feet off the ground to accommodate some instrumentation (keep it from being flooded if/when they get a leak). It's slurry (pumping gravel) line so it has to be routinely rotated for wear reasons so putting bends or fittings in it is not desirable. So my thought was to gradually increase the elevation over a fairly long distance by berming it up a bit and leave it at that. The complication was in determining how to determine the proper slopes on the berms instead of just leaving it to heavy equipment operators.

600D sounds a bit excessive because with a 20" diameter pipe that works out to 1"/1000 feet whereas normally a very stiff beam is typically rated at around 1":200 to 1":300 feet in most structural specifications. The pipeline clearly goes through much more surface undulation than that.

"Pressure limits" is a relative word. Pressure is limited to 290 PSI due specifically to the using of Class 150 flanges and gaskets, NOT due to any structural limitation of the pipe structure itself. So there is a lot of room here.

I was finding lots of information on burying pipelines and calculating stresses as mentioned which seems to be a fairly popular topic. If I could just either bend it intentionally or add fittings, it would similarly be a nonissue. The fact that I'm dealing with a material that gets replaced every couple years, that has to be routinely rotated, and is nowhere it's design stress limit at least in the solid sections (not joints) seems to be something not in the typical pipe engineering literature.

It seems like my only real choice is to go back and design this as a beam loading problem and look at the stresses from that point of view.

 

[miningman]

Theres a really good reason why your mining engineering friends ignore the theoretical BS of design limits which piping proffessionals on a forum like this might be unaweare of. By your own admission these slurry lines DO leak. Leaks tend not to be catastrophic and are not in residential areas, at worst a significant clean up job , which is as you say , " just another day in the office" for mining professionals. Also as you allude to, this service is extremely abrasive and the pipes will wear thru unless they are rotated regularily. The number of manhours involved in this operation by far exceeds the installation manhours. As an engineer, your efforts should be devoted to providing the operations people a system that makes their lives easier, not spending time worrying about theoretical BS. Which also raises the question , what is an electrical engineer doing speccing out a tailings line application.? And again as you allude to, steel is not the optimum material for such a line. Hundreds of miles of HDPE tailings lines are in use across the world, from the arctic to the sahara. Is there a really good reason to use steel here??

 

[LittleInch]

Maybe I didn't explain properly, but 600D is the radius of the inside of your elastic curve - what did you think it was??. Thus for a 20" pipe, this is around 300m radius. For a 1000 foot length this equates to an overall 60 degree bend - a lot lot more than 1". This is a bit of a generic rule, but much below this - at about 400 to 500D, your pipe will start to span or lift off over sharper humps and you risk buckling the pipe. Draw this out in section using a radius of 300m and you can get a profile which a decent civil contractor can follow.

If you want to get technical, then you can work out the combined stress taking into account the bending stress and adding it or subtracting it from the axial stress caused by poissons effect and end cap forces plus hoop stress. ASME B31.4 (2012) now covers slurry pipelines and gives you the allowable stresses. If your pipe is much thicker than you need for pressure containment then you should be Ok.



My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[paulengr]

At the current time my role is an electrical engineer. However that means little. I've worked as a metallurgist (6 years), maintenance engineer (do everything, 5 years), maintenance manager (2 years, with dredgees and HDPE), and now electrical engineer (5 years). I'm dual degreed in process engineering and electrical. I've sized plenty of pumps and piping systems, and can fluently argue the nuances of Hazen vs. Darcy coefficients. But by and large everything I have been working with lays on the surface unless it's a chemical line or a water line strapped to the side of a tank, or it's running down a borehole or along a drift underground in which case it is technically buried.

The current project is to redesign the flow and density meter mountings to stop wiping them out every time we have a pipeline leak in the local area, and if you listen to the instrument guys, they all want a vertical pipe run. With the flow meter, it's just plain silly. Full pipe is an argument for people with short pipelines. All I need to do is elevate it for a short distance. The issue here is just to avoid belly aching about whether or not the berm is feathered out properly to achieve that while pleasing the instrument premadonnas about whether or not they are going to get their instruments or their feet wet working on it.

The density gauge crowd (Ohmart in this case) are claiming that they think they need 5 diameters up and downstream. With a 20" pipe and figuring in long radius elbows, I end up with a structure that is close to 30 feet tall, and to put on my electrical hat for a moment, I've only really set up the power lines to accomodate 23 foot tall equipment. So in a way this is even more ridiculous than the flow meter crowd. I can't find a shred of documentation supporting these requirements so I'm all ears for alternatives or some sort of reason why 10 diameters makes any sense whatsoever. I can see justification with the fact that I effectively can't avoid settling at all even with 12+ feet/second line velocities but I can't imagine what the straight pipe argument is for with density gauges, so right now I'm mostly leaning towards a slightly elevated horizontal section.

I agree that wear on slurry lines is the #1 issue and I'd also agree that if you can possibly avoid it, go with HDPE. It just plain lasts so much longer as long as you don't do stupid things with it that there is almost no comparison. 99 out of 100 mines can't be wrong, right? However there are three major gotchas with HDPE:
1. Pressure limitations. If someone for some reason decided to run with a pipeline that is way too small (hint, hint), then pressure limitations get to the point where you first spread your booster pumps way out and then start converting to steel to double the pressure limit, then eventually the higher class flanges start to look really good. Eventually someone just has to face reality and spend the money on bigger pipe but they've managed to dodge that bullet for a long time here.
2. Velocity limits. If you don't screen in the pit then the only alternative to sanding up is run very high line velocities...12-15 feet/second. Which means small pipe.
3. Buying HDPE is the same thing as buying solidified oil, and you need a lot more plastic than steel. So right now it looks awful expensive at least at first, especially when the pesky pressure problem also pushes for even larger pipe than what has been used in the past.

So as of right now I'm really trying hard to avoid building some gigantic hairpin thing to satisfy some kind of theoretical concern while also trying to answer concerns about how long to make a sloping berm to support the pipe while making it tall enough to silence the instrument premadonnas.

I eliminated all these details in the first post though because I tried to distill the basic problem down to something that I thought would be fairly simple, straightforward, and probably out in some ASTM standard or ancient engineering guideline somewhere that everyone has followed since the dawn of time but unless you live, sleep, eat, and breathe, pipe, you wouldn't know about it, similar to the Crane guides that are so very popular for hydraulics with some engineers.

 

[LittleInch]

That's some experience. For the initial point see fairly basic sketch to make my original point, i.e. just make your very long radius bends 600D radius to get up to your additional 3-4 m or whatever height you need to avoid the occasional flood(!). You can get this more accurate of course on CAD to generate a profile for the civil guys to follow. This will lie on the ground no problem without spanning or overstressing.

I tend to agree that for a long line at these velocities your pipe will be full and you don't need to be vertical for the instruments. The 5D issue is normally that to get good measurement, the flow really needs to be stable and without swirl or undue turbulence which normally occurs after elbows, tees etc which is normally satisfied by having a straight length so the flow becomes relatively stable. Very common requirement for meters and such like, including nuclear density gauges which is what I think your density guage is.

In terms of PE, have you looked seriously at lining a steel pipe with PE? With a straight line in particular, you can insert something like 1000 to 1500m in one go , especially if you bend it into a U shape and then break the tapes and revert to a circle. You essentially gets the best of both steel (higher pressure but thinner as you don't need to allow for the erosion) and PE (again thinner as it doesn't need to have material for pressure. For smaller lines there is also now a variety of steel and fibre re-inforced plastic pipe, normally up to about 8" to 10" OD, which is reelable and does similar things - flex steel is an example but there are others



My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[paulengr]

PE linings: Yep, been there, done that. A 40+ mile long line in central Georgia was being attacked by cyanobacteria which is a long word for a bug that lives in the soil and eats pyrites and poops sulfuric acid. If nothing is done about it within a few years they will perforate the pipe walls. By pigging the lines routinely to knock the colonies off the pipe walls it can extend pipe life out to 10-20 years. Even better is once the pipeline is getting to the point where reliability is a real concern, go back and insert PE liners. This eliminates the problem, improves flow coefficients, and extends pipe life to at least decades. However again there's that pesky wear thing that may destroy the advantages over havying a big thick wall such as SDR 11.

You forgot to mention that you can also achieve something very similar by using epoxy linings. These days you can use FBE (fusion bonded epoxy) to achieve nearly any coating thickness and are no longer limited to the 5-10 mils of gunned on powder coating. Although it was liquid applied by a contractor at my former employer, this does almost the same thing for steel and ductile iron pipe and is supposed to be cheaper for initial construction.

The 5 years as maintenance engineer was for the oldest cast (now ductile) iron pipe plant in the U.S. Their ideas about pipe life are 700+ years projected service life, and even non-ductile cast iron pipe from over 200 years ago is just now being replaced. It's the only place I've ever worked that actually offered a 100 year warranty on their product. That particular site had been manufacturing pipe of some kind for over 200 years. If you are installing something that will last that long, then clearly there is a value n worrying about "theoretical" concerns about pipe installations.