27 replies to this topic

[ankur2061]

Dear All,

In the GPSA engineering databook, 12th edition FPS version, there is a chart (Fig. 20-4) for water content of NG. The chart was originally published by "McKetta" & "Wehe" in Hydrocarbon Processing 1958 issue.

However, this chart is not a very easy chart to read & for values in between the values given in the chart, it is really difficult to read the water content.

I am looking for an empirical equation for calculating the water content of NG at a given saturation temperature & pressure & specific gravity which would give fairly accurate values.

Would appreciate any help from our forum members.

Regards,
Ankur.

[daryon]

Hi Ankur,

Sharma and Campbell developed a semi empirical correlation for calculating water content of natural gas, it can be done by hand but apparently it is complicated, involving water fugacity at saturation and system fugacity at operating conditions. Its applicable from 1 to 690 barg and 15 to 238 degC and is reported to be accurate to +/- 5% which is comparable to the McKetta-Wehe Chart. I have never used it, the source of this info is Natural gas hydrates: a guide for engineers by John J. Carroll

There are charts in Campbells Gas Conditioning and Processing, Volume 1, that appear easier to read than the McKetta-Wehe Chart in the GPSA handbook. They are describe as a composite based on all earlier charts from McKetta-Wehe, McCarthy et. al., and the institute of gas technology.

Robinson et al made computer generated charts using SRK EOS for sour gas water content:
Title Estimation of the Water Content of Sour Natural Gases
Authors Robinson, J.N., Wichert, E., Petrofina Canada Ltd.; Moore, R.G., Heideman, R.A., U. of Calgary
Journal SPE Journal
Volume Volume 17, Number 4
Date August 1977

There are rigorous 3-phase computer based flash calculations methods that predict the distribution in all phases, which include composition and are more accurate e.g. Aqua*sim (ref. Oilfield Processing of Petroleum: Natural gas by Francis S. Manning).

I've written to Aspentech to ask how HYSYS calculates the water content of gas, I think it is based on McKetta-Wehe. I'll let you know when I get an answer. I am not aware of an reliable equation that you seek.

[ankur2061]

Daryon,

Many thanks for your response. It is highly informative. I needed this information because I have developed a spreadsheet for molecular sieve bed sizing for dehydration of NG based on the GPSA method but the only missing part is the calculation of the water content which one has to refer from the McKetta-Wehe chart which can only be embedded as an image file in the spreadsheet & is not visual friendly.

Again my sincere thanks for your informative response.

Regards,
Ankur.

[breizh]

Hi Ankur ,
Let you try this book ,

http://books.google....s water&f=false

Hope this helps

Breizh

[daryon]

Hi Ankur,
I heard back from AspenTech... "The gas composition is determined by the equilibrium calculation by the Property method selected in the Fluid Package. Empirical methods are not used." This makes perfect sense, I've no idea why I thought HYSYS would use McKetta-Wehe seeing it is already preforming 3-phase flash calculations to determine the distribution of each component with in each phase.

Perhaps you could add a note to your hand calculation suggesting that the gas water content (estimated from McKetta-Wehe chart) is checked against simulation software such as HYSYS, UniSim e.t.c. I think this software will provide slightly more accurate figures for water content as gas composition is taken into account.

Regards,
Daryon

[Rock_Dock]

There is a John Campbell article that suggests using ProMax for predicting the water content. Here it is:

http://www.jmcampbel...m/december-2007

I have checked ProMax values against GPSA figure 20-4 and found they matched very well, with an average error of < 5%. When I compare them to Hysys I get an average error of 276%. I took into account various feeds and such. I would expect UniSim to be about the same as Hysys.

[ankur2061]

Rock,

Fig 20-4 in GPSA is for sweet gases. The article you have posted as a link is for sour gases. I have not checked against HYSYS or any other software.

Regards,
Ankur.

[ankur2061]

Dear All,

The McKetta-Wehe chart given as Fig. 20-4 in GPSA engineering databook, 12th edition is also available as a table in the following book:

Table 8.1, Water Content Readings from the Chart by McKetta and Wehe (1958) (lbm H2O/MMcf @60 0F, 14.7 psia), Chapter 8,
"Dehydration", "Natural Gas Engineering Handbook" by Dr. Boyun Guo and Dr. Ali Ghalambor

From this table one can directly read the water content of natural gas for the given saturation temperature & pressure.

Intermediate values can be linearly interpolated from the tabular values.

Regards,
Ankur.

[ankur2061]

Dear All,

New developments in determination of water content have resulted in some very accurate empirical correlations for water content determination. In pursuance of this subject I have traced a new article which gives a very accurate correlation for water content of both sweet & sour gases. I have even programmed the equation in an excel spreadsheet "Hydrate Inhibitor Injection". Following is the link:

http://www.jstage.js...i/52/5/270/_pdf

Hope this helps.

Regards,
Ankur.

Edited by ankur2061, 03 June 2010 - 12:24 PM.

[Robert Montoya]

In the HYSYS you can to find in UTILITIES the diffrent dew point for several pressure and temperature of the gas. In the GPSA and Cambell all the values is refer to 60 0F, 14.7 psia for the conditions of saturations

Regards

[ankur2061]

Dear All,

I tried to program the equations in the above mentioned article as an excel sheet & I was getting erroneous results. I wrote to the author & he has responded promptly with some corrections. Following are the corrections:

In eqn 6, the variable T should be in degree Kelvin & not P as mentioned in the article.

In eqns 7 to 10 the variable P should be in kPa(abs) and not T as mentioned in the article.

With these corrections implemented the results are very much close to experimentally measured results of water content of sour natural gases. The eqns 1 to 5 are correct as given in the article, which provide the water content of sweet natural gas.

Hope this helps.

Regards,
Ankur.

[ankur2061]

Dear All,

A document by the title:

ASTM D1142 - 95(2006) Standard Test Method for Water Vapor Content of Gaseous Fuels by Measurement of Dew-Point Temperature

has both a calculation method as well as a table for water content of natural gas as a function of various temperatures & pressures of the natural gas. The table can be plotted as a graph and a regression correlation developed from it.

Regards,
Ankur.

Edited by ankur2061, 09 July 2010 - 09:25 AM.

[ankur2061]

Dear All,

Please find attached a spreadsheet for "Water Content of Sweet Natural Gas". The attached spreadsheet is valid for NG pressures upto 500 psia. Comments are welcome.

Regards,
Ankur.

Attached Files

•  Water_Content_Sweet_Natural_Gas.xls   14.5KB   1472 downloads

[Propacket]

The spreadsheet is based on Ideal Model for calculation of water content. This method assumes negligible solubility of gases in liquid water phase. That is why this method is suitable for low pressures only. Per spreadsheet, it is mentioned that it is applicable to pressures less than 500 psia. I remember that a technical paper on different methods for measuring the water content of gases revealed that this model is accurate only at low pressures less than 50 psia. At pressures greater than 50 psia upto 200 psia, it gives 25% error. At pressures greater than 200 psia, use of this model is not recommended. I will try to find a copy of the said paper.

Thanks

[ankur2061]

P. Engr.

The attached spreadsheet has been checked against the "McKetta-Wehe" chart given as Fig. 20-4 in the GPSA Engineering databook, which is an industry standard for determining the water content of sweet natural gases. The error up to the pressure limit of 500 psia has been found to be less than 10% of the chart values.

Regards,
Ankur.

[Propacket]

In that case, it is an excellent one.

[bustanul aifin]

thank you Mr Ankur for your sharing....it's nice input for me

[Padmakar Katre]

Dear Ankur,
There is one extension available with Hysys called 'Saturate' You can contact Aspen Tech support to know the equations used or the references for this extension. I am sure they will not mind to part/share the information with the user. May it help you.

[MrShorty]

Ankur:

Interesting. As P.Engr noted, it is a very idealized calculation (basically Raoult's law P(H₂O) = P0(H₂O) * x(H₂O) where x(H₂O) is assumed to be 1). P(H₂O) is the partial pressure of H₂O, P0(H₂O) is the saturation pressure of pure water, and x(H₂O) is the liquid mole fraction of H₂O.

I wonder if the estimate would improve if you used the Henry's constant for LNG to get a more accurate measure of x(H₂O)?

I also notice that the estimate is limited to 0 C and higher because you are using the vapor pressure equation of liquid water. If you substitute the sublimation pressure of ice, can you extend the estimate below 0 C?

[ankur2061]

Dear All,

There has been a very healthy debate on the subject matter and I have spent considerable time and effort researching the subject. Today on the auspicious day of Diwali I am presenting my latest efforts on the subject matter by providing a spreadsheet which is very accurate in not only predicting the water content of sweet natural gas but also for sour natural gas. My Diwali gift to all the esteemed readers and friends of this fantastic forum. Comments are welcome. In the earlier posts on the same topic I have mentioned about this research of mine and wish to thank the authors of the article from which the spreadsheet is inspired. The title of the paper and the link for the article is also mentioned in the spreadsheet.

Comments are welcome.

Regards,
Ankur.

Attached Files

•  New_Correlation_for_Water_Content_of_Sour_NG.xls   52.5KB   1358 downloads

[csp_process]

Nice work Mr. Ankur,
I'm late into the discussion. It was one nice and healthy discussion.
Water content of Natural gas is widely researched topic with thousand of publications.

Following are the books which have clearly indicated merits & demerits of each method- equation w.r.t Composition, Pressure, temp., even salt content sometime. User has to decide correct equation/method applicable to his case.

1. Standard Handbook of Petroleum & Natural gas engineering- by Williams C. Lyons, Gary J. Plisga- Subject matter sought by you can be read starting at Page 6-274 in second edition.

2. Gas Production Engineering by Sanjay Kumar

These books are being taught to students and referred by Industry for decades.

[anais90]

Dear Ankur,
There’s no way to use the simulation software for moisture calculations in NG? I’ve been trying with ProSim Plus 3 but it didn’t show good results. What can I do with the unit operations that add water to the Gas stream?

Thank you for the spread sheet,

Regards

Cristóbal Gordillo

[ankur2061]

Cristobal,

You can add water to dry Natural Gas by using the "mixer" unit operation in any simulator such as HYSYS or Prosim. This involves addition of water to the NG stream at the given pressure and temperature and using a flash unit operation in the simulator. The vapor outlet from the flasher at the NG operating pressure and temperature will give you the saturation water quantity of the NG.

Hope this helps.

Regards,
Ankur.

[marchem]

you may visit this thread,

http://www.cheresour...-in-properties/

Ankur deserves our thanks for the spreadsheet however if you need accurate values of water / natural gas equilibria a more accurate model as ISO 18453 or equivalent is preferable, these models are mentioned in the above thread

[Dandy Komboris]

Dear Mr. Ankur

From your Excel sheet (New Correleation for Water COntent of Sour NG) I input some data.
I have measured natural gas data from the field.
Pressure is 8 Kg/cm2 and Temperature is 72 F.
I convert the date so i get 585.1125 kPa (absolute) and 22.22 Celcius
From your excel sheet i have 215.7 Lb/MMSCF

And from the Water COntent of Hydrocarbon (GPSA Engineering Data Book Fig 20.4) I also get the result is 210 Lb/MMSCF.

But from the field we have different result Sir.
We use a device to measure Water Dew Point in Our Natural Gas FLow.
From the device we have dew point temperature is -20 Degree Celcius so that we have 48.57 Lb H2O/MMSCF.

I still confuse about this.
Whether my device is wrong or maybe i am wrong for using The Chart and your formula (perhaps my natural gas spesification doesn't meet the spesification for using Fig 20.4 from GPSA Eng Data Book 12th Edition or for using your Formula in Excel Sheet).

Please Enlight me Mr. Ankur


Best Regard


Dandy Komboris