36 replies to this topic

[FUNMILAYO]

Good Day all, please i kindly request for help on how to size a gas line with the following operating conditions:

 

Length - 2km

Min Flowrate - 7.5 mmscfd

Design Flowrate - 70mmscfd

Temp in - 27deg c

Pressure in - 40 to 70 Barg

Pressure out 0 32.6 to 34.5 Barg

 

please i anticipate your response.. thanks a lot.

[ankur2061]

Funmilayo,

 

Some questions that you need to answer before your line sizing can be done:

 

1. Is the line running above ground or is it buried? If it is running buried then what is the soil temperature? If it is above ground then what is the maximum ambient temperature?

 

2. What is the molecular weight or specific gravity of your gas?

 

3. Can you provide the composition of your gas?

 

4. Please provide the inlet pressure which you want to use for sizing and not the range (40 barg or 70 barg)?

 

5. Please provide the end or termination pressure at the end of the pipeline for the purpose of sizing and not the range (32.6 barg or 34.5 barg)?

 

Regards,

Ankur.

Edited by ankur2061, 14 March 2013 - 06:40 AM.

[FUNMILAYO]

1. Some part of the line is buried and some is on the ground.

 

 

SOIL PARAMETERS

 

For the soil characteristic along the pipeline, the following assumptions were made;

Soil cover above pipeline top:                  900mm (onshore)

Soil temperature at 900mm depth:            27.9oC

Soil Thermal Conductivity:                        0.9 W/m/K

For the pipeline, the following parameters have been assumed:

•             Pipe Wall Conductivity:           45 W/m/C (Carbon Steel)

•             Pipeline coating:                     Polyethylene (PE)

•             PE Coating Thickness            3.5mm

•             PE Coating Conductivity         0.35 W/m/C

•             Equivalent Roughness:           10µm (internal lining)

•             Corrosion Allowance               3mm minimum

 

1. Molecular weight of gas -    0.7487

1. nlet Pressure -  70Barg
2. End Pressure – 34.5

Thanks a lot!

[FUNMILAYO]

soory about the way the table appeared

 

Gas Composition                       Minimum   Maximum

Methane                                              82               96

Ethane                                                 0                 10

Propane                                               0                  8

Butane + Paraffin (C4+)                       0                   7

Inert Gas                                        Not more than 15 mole percent

CO2                                               Not more than 8 mole percent

H2O                                              10 lbs/MMSCF

H2S                                                      0                   4ppm

Total Sulphur (by volume)                    0                    28ppm

Nitrogen (N2)                                        0                   3% by volume

Oxygen (O2)                                         0                   10ppm by volume

GCV   (Gross  Calorific  Value)          900 - 1150  BTU/SCF

Wobbe Index (HHV Basis)                    47                    52

[Pingue2008]

Dear,

 

I hope the attached spreadsheet helps. It is for a prelimary estimate. so based on the info you provided and the assumptions I made you need 14" pipe diamter (NPS).

 

Thanks,

Attached Files

•  gas line sizing.xlsx   24.27KB   329 downloads

[ankur2061]

funmilayo,

 

The value of molecular weight of gas 0.7487 seems to be incorrect. Is it the gas specific gravity instead of molecular weight?

 

 

Regards,

Ankur.

[FUNMILAYO]

tnks a lot  Ankur, sorry for the error 0.7487 is the the specific gravity not the mol wt.

 

 

Best Regards

 

Funmilayo

[FUNMILAYO]

Thanks a lot  Pingue2008

[ankur2061]

funmilayo,

 

When I calculate with the data you have provided I get the following results:

 

Input Data:

 

Inlet Pressure: 40 barg

Outllet Pressure: 34.5 barg

Molecular Weight: 21.68

Tin = 27 deg C

Tsoil = 27.9 deg C

Flow Rate: 1,981,728 Sm3 / day (70 MMSCFD)

Pipeline Efficiency Factor: 0.94

Base Equation: Panhandle B

 

Results:

 

Calculated minimum Pipe Internal Diameter: 197.3 mm (the nearest ASME/ANSI B36.10  wrought steel pipe is 8" Sch 40 = 202.7 mm ID)

Line Velocity: 18 m/s

 

Regards,

Ankur

[FUNMILAYO]

tnks a lot Ankur... i really appreciate this

[Steve Hall]

Using an isothermal pressure drop correlation for compressible flow, 8" pipeline (OD 219 mm, ID 202.7 mm, 2000 m length, roughness 0.000046 m), and flowing conditions:

Pressure upstream = 4,155 kPa absolute

Temperature = 27 deg C

MW = 21.68

Mass flowrate = 75,700 kg/h

Compressibility Factor = 1

 

I get:

Upstream velocity = 18.05 m/s

Pressure drop = 2,508 kPa

Pressure downstream = 1,647 kPa absolute

Downstream velocity = 45.5 m/s

 

I'm surprised my result differs so much from Panhandle B. Thoughts?

Edited by Steve Hall, 14 March 2013 - 11:28 AM.

[ankur2061]

Steve,

 

I did the calculations again, this time using the AGA Equation for fully turbulent flow and I get the following results:

 

Input Data:

 

Inlet Pressure: 40 barg

Outllet Pressure: 34.5 barg

Molecular Weight: 21.68

Tin = 27 deg C

Tsoil = 27.9 deg C

Flow Rate: 1,981,728 Sm3 / day (70 MMSCFD)

Pipeline Efficiency Factor: 0.95

Base Equation: AGA Equation for turbulent flow

 

Results:

 

Calculated minimum Pipe Internal Diameter: 221.6 mm (the nearest ASME/ANSI B36.10  wrought steel pipe is 10 Sch 140 = 222.2 mm ID)

Line Velocity: 14.3 m/s

 

Regards,

Ankur

[Steve Hall]

Interesting... My calculation is in good agreement with Ankur's AGA result in Post 12:

 

10", Schedule 140 (222.2 mm ID), downstream pressure = 28.8 psig

10", Schedule 30 (257.5 mm ID) which is standard weight for 10" pipe, downstream pressure = 36 psig

all other parameters same as Post 11

 

One of the takeaways here is that the old workhorse gas pipe formulas - Weymouth, Panhandle A, and Panhandle B - incorporate a generalized factor called e (for pipeline efficiency) that is intended to account for internal roughness, bends, and fittings. The "typical" choice for e is from 0.85 to 0.92, but it's a subjective choice that the pipeline designer chooses.. The formulas also have a built-in friction factor that is fixed (friction factor actually changes with Reynolds number, pipe diameter, and roughness). I don't know about the AGA equation.

Edited by Steve Hall, 14 March 2013 - 01:13 PM.

[Pingue2008]

FUNMILAYO, As you can tell as a process Engineer we are going to make the call on this and enjoy the consequenses.

OTHERS, my number on the attached spreadsheet is not near close to what you guys are suggested. could please look my cals and let me know what i am not accounting for? or overlooking.

 

Thanks,

Attached Files

•  gas line sizing.xlsx   24.83KB   156 downloads

[ankur2061]

Pingue2008,

 

Your viscosity value is way-off. It should be 0.011 cP instead of what you have considered as 0.049 cP.

 

In my calculation I had considered an absolute roughness of 46 micron (0.00015 ft) which is the standard roughness for new carbon steel pipe. However, even if I consider an absolute roughness of 10 micron (0.000033 ft) then I calculate an internal diameter of 209.5 mm (nearest equivalent ANSI/ASME B36.10 wrought steel pipe is 8" Sch 20 = 206.4 mm ID).

 

Regards,

Ankur

[Pingue2008]

Dear,

 

I ran a program called engineer's Aid and my pressure drop and velocity match the program results (see attached). I assumed a diameter and determined the pressure drop and the velocity. My question now is this one: what is the pressure drop contraint and velocity constraint for gas line sizing. I found a lot of materials on this forum about liquid line sizing constrains. I was told early on that the velocity requirement should be between 50-110 ft/sec. as one can see this range is very broad. I think this is where my  dilemma resides (what is the single target pressure drop/velocity allowed. I included different pipe sch for comparison.

 

Thanks,

Attached Files

•  gas line sizing.xlsx   46.51KB   129 downloads

Edited by Pingue2008, 14 March 2013 - 04:53 PM.

[katmar]

I have used the same assumptions as Steve Hall (i.e. isothermal compressible) and the results I get are very similar to Ankur's AGA calculation, but rather different from Ankur's Panhandle B calculation and also different from Steve's results.

 

The only difference I can see from Steve's calculation is that I took the compressibility as 0.92 at the flowing condition, but this is nowhere near enough to explain the difference.  I am a bit confused by Steve's comment in Post 13 that his results are in good agreement with Ankur's AGA results in Post 12.  For a pipe ID of 222 mm Steve gets a pressure drop of 11.2 bar while Ankur gets 5.5 bar.

 

In converting the flow in SCFD to mass I have assumed STP is 60 F and 14.696 psia.  This gives me a standard density of 0.915 kg/m³ (0.0571 lb/ft³) and therefore a flowrate of 21.0 kg/s (46.3 lb/s) which is the "same" as Steve's 75,700 kg/h.

 

My calculation (attached) gives a required diameter of 219.7 mm to achieve a pressure drop of 5.5 bar, and I would regard this as fairly close agreement with Ankur's AGA value of 221.6 mm.  I would expect the old style equations to be slightly conservative and comparing 221.6 with 219.7 this seems to be true - but the Panhandle B at 202.7 mm is not conservative.

 

The reason Pingue2008's results are so different is that he has used the minimum flow rate and not the maximum.  This makes a difference of almost 10x in the flow rate. Also his density of 2.22 lb/ft³ is (close to) the flowing density and not the standard density so he has got the wrong flowrate in mass terms.  The difference in viscosity is negligible.  It only affects the Reynolds number and in the regime we are working the friction factor is not sensitive to the Reynolds number.

Attached Files

•  funmilayo.pdf   10.64KB   210 downloads

[breizh]

I've added an other xcelsheet , which is matching quite well Harvey's results.

 

Breizh

[Steve Hall]

Thanks, all. Katmar, my error was in the viscosity (mistakenly entered mPa-s instead of Pa-s in my program). Also, my approach is different: instead of calculating a diameter to match a desired flowrate, I use look-up tables with actual pipe diameters. By making the adjustments, my results match yours.

[Art Montemayor]

Wow!  What a great thread and topic!

 

The topic is not only important and interesting, it has been handled, discussed, and responded to by some of our top members - and all in ONE DAY!

 

I have really enjoyed this thread and want to thank you all that contributed.  I hope our other members realize the value here and take the opportunity to read these valuable comments as well as download and study the excellent calculations submitted.

 

Kudos to all.

[Bobby Strain]

Art,

   So, how many engineers are required to do a value conversion from one set of units to a different set, without benifit of Katmar software? I haven't checked the gas flow hydraulics yet. I was out selling work to Valero. But I will give it a check. And, with all these comparisons, does anyone have the pipeline calculation software for comparison?

 

Bobby

[katmar]

Bobby - this is a very important point that you raise here and one that is probably a more important one for the junior engineers to learn than the technical aspects. And that lesson is that all engineering work should (must!) be checked. It is standard practice in consulting and construction companies that all calculations should be checked by a different person and ideally using different methodology.  When I have had young engineers come onto a project and told them of the procedures involved to get their calculations checked they have often had the atitutude of "don't you trust me?".  I have always tried to show them that it is more for their own protection than for the company's protection.  It is the engineer's right to have someone check their calcs. We all make mistakes at times and having a backstop is essential. Nobody, no matter how senior or experienced, should ever object to having their calcs checked. 

 

Even if the methods are applied perfectly, it is a very instructive process to see the different results that are obtained with different methods - as we have seen here with the comparison of the Panhandle B, AGA and the integrated isothermal compressible gas equation.

[FUNMILAYO]

Thanks All for these enlighten information shared...

[Pingue2008]

Katmart and Others,

 

I am trying to understand the difference.  Either the program I use is out of date or there is something that I don't grasp. I ran my program again with the number Katmart has in his program (21kg/sec, 38.7kg/m3, total pressure drop 5.5 bar,viscosity 0.011Cp) and my number is still quite different. please see the output from the program and comment please. the output here is close to my excel. I did not attach the excel.

 

Thank you,

Attached Files

•  Gas.JPG   69.22KB   29 downloads

[katmar]

Pingue,  You must use the upstream flowing density (i.e. 2.42 lb/ft³) to calculate the pressure drop.  The density you have used is the standard density (i.e. 0.0571 lb/ft³).  You need to use the standard density to convert the flow in standard volumetric units to a mass flow, but to calculate the pressure drop you need the actual density.  If I key in the values in your screenshot I get much the same results as you, so I don't think there is anything wrong with your software.

 

If you scroll down to page 4 there was a topic that was last posted to on 24 Feb and titled "Line Sizing for Compressible Fluid" where the actual and standard conditions were discussed.