Raw gas pipeline sizing!!! 1

[hot2000]

How can ý calculate the pressure drop ýn raw gas pýpelýne?
I use Reneuard equation for sizig and ASME 31.3 for determining the wall thickness? Is ýt rýght or wrong?
Could you please recommend me?

 

[TD2K]

If the gas is dry with few liquids you can simply calculate the pressure drop using Darcy Weisbach corrected for the gas expansion if that is critical. I suggest, if you don't have it, you get a copy of Crane's Technical Paper 410, a very worthwhile investment.

If you have liquids (which is likely since you said raw gas), that becomes an other issue. To get some answers (and I'm not very familiar with multiphase pressure drop correlations), you'll need to give us an idea of the pipeline size, length, operating conditions, liquid/gas ratios, etc to get a reasonable answer.

I'm not familar with the Reneuard equation and Google doesn't find any information on it so I can't help you there.

You would use B31.3 for selecting the wall thickness if you are designing the pipeline to B31.3 OR that is what you want to use to calculate the wall thickness even if the pipeline is being designed under another code (and assuming the other code is less strict than B31.3).

Your question whether B31.3 is the right code is a bit of 'how big is big', it's difficult to answer. Gas pipelines are normally not designed to B31.3 (B31.8 and other regulations in the US typically apply) but the owner can elect to go to a more conservative design if they want to (and I believe B31.3 will give you the thickest wall but I don't work with the piping codes to any great depth so I could be wrong). However, you really haven't posted enough information where this pipeline is or the situation to give a complete answer and I don't want to mislead you.

 

[hot2000]

Dear TD2K,
Flow rate is 3MMSCF/day.
Sp. Gr . is 0.65,
water coming out with gas from well is 2 bbl/day,
condansate coming out from well is 24 bbl/day,
well head pressure is 1650psig, pipeline length is 1650 ft,there is no elevation difference....
so,
i did the calculations based on Reneuard equation and found 2 3/8" OD 4.9 mm thickness is adequate...If you make your own calculations that would be great to compre the resuls...
thanks...

 

[TD2K]

For your gas, water and condensate flow, the gas is over 95% by weight of the total (and even more on a volumetric basis) so it's essentially a gas system.

I calculated you would need less than a 30 psi pressure drop over this length to get your required flow rate so your pipe size seems reasonable assuming 100F gas and 2" sch 80 pipe (pending your final decision on what corrosion allowance you need). 2" is also a reasonable minimum size, smaller pipe needs more support and is more suspectible to damage so I wouldn't use them.

The velocities are fairly low, about 15 feet/sec, so you will get some liquids accumulating in any piping pockets or low points (for example, where you go into the piperack), you'll need to take that into account in your design.

You'll need to get someone else to check your wall thickness as it's not my area of expertise.

Is the 4.9 mm a calculated wall thickness as 4.9 mm doesn't match standard pipe wall thickness?

I don't think you have included a corrosion allowance for your piping. For piping designed to B31.3 handling wet fluids, I seen anywhere from 1/16" to 1/8" with 1/8" being more common. If you don't include a CA and there is any corrosion, you'll need to reduce the MAWP of the piping to reflect the reduced remaining wall thickness.

 

[1969grad]

What in the world is the Reneuard equation? Been doing hydraulics a long time and never heard of it.

If you use a dry gas flow equation you should use a flow efficiency factor assuming that the liquid loading is not more than 10% by volume. Otherwise, use a two phase flow equation such as Biggs and Brill.

Suggest you make a PIPEPHASE run to confirm your line sizing.

If the pipe is not within the battery limits of a facility then the appropreiate code is B31.8.

For a corrosion allowance you need to consider how much CO2 and H2S is in the gas stream as the free water will combine with these two components and make an acid requiring that you include chemical injection with a corrosion inhibitor.

 

[hot2000]

Dear 1969grad,

The Renouard Equation is,
(P1)^2-(P2)^2= 48.6xdxLx(O)^1.82x(Di)^-4.82
where ,
P1 inlet pressure,bar
P2 outlet pressure,bar
d=gas fictive density app. 0.56
L=length of pipe,mt
Q=gas flow rate,mt3/hr
Di=inside pipe diameter,
and the criteria is V, velocity
V=(353.9xQ)/(P2xDi^2)

if the velocity is below 25 mt/sec the design is true...
It's a trial error method..You know the inlet pressure(P1), length(L), flow rate(Q) and assume a diameter(Di) and calculate a outlet pressure(P2)and compare with the velocity(V)...if the velocity is lower than 25mt/sec the diameter is true..if higher try another diameter...
SO I did not heard about BIGGS&BRILL Equation could you describe me?
ANOTHER,
I haven't sour gas in here so Can I use the B31.8? Breou Veritas is supervising our pipelines and they say we should use B31.3...How can I convince them? Because it's loss of money , pay for thicker pipes..
ANOTHER,
How can select my CA(corrosion allowance) ? What is the criteria for choosing the 1/4", 1/8", 1/16" CA?
thanks for your advises and help gentelmen!

 

[TD2K]

I would go back to Breou Veritas and ask them why they want to use B31.3 as the pipeline design code over other codes. You haven't got anywhere near the experience to be making a recommendation/judgement like this (sorry, not trying to be mean, just upfront with you). You likely hired them to supervise your pipelines based on their experience, use it.

Take this as an opportunity to learn engineering at this point in your career. For example,

http://www.google.com

is a great way to get some general and even specific information such as on Beggs and Brill. Consider getting some fluid flow books, Beggs and Brill is a very common multiphase flow correlation and there are several of them. The industry trade journals, if you can get them, are also very good for a new engineer in my experience.

The Renouard Equation appears to be somewhat similar to the Panhandle or Weymouth formulas in the US. They are an empirical formula for large diameter pipelines and I think using it for 2" piping is pushing it.

Setting the CA has a lot of judgement call to it. How corrosive is the fluid, how errosive is it? What is the piping material? How long do you want the pipeline to last (a pipeline with a short life doesn't need as much CA as one that will be in service for many years). If it's a corrosive service, how succesful will your corrosion prevention program be?

 

[1969grad]

Well said TD2K.

Might add that anytime you use an flow equation with other than Darcy friction factors (Modified Colebrook-White) or the AGA equations then you allways have to check the Re No. to make sure that you are within the valid range of the formula. Even then, you have to make sure that you have turbulent flow.

For example, there are several versions of the Panhandle flow equations. If you use them in their valid Re No range then they should give you good results.

By the way, Biggs and Brill have a text book on 2 phase flow. Also, check the PIPEPHASE manual for other equations.