wet gas 2

[VikingUK]

Anybody know of a quick and simple way of determining how wet a gas is coming from a well ? I'm thinking of something like a draeger or kitigawa sample tube/hand pump.

 

[JLSeagull]

How simple? Consider moisture analyzer sensors such as GE Sensor Panametric moisture transmitters.

 

[zdas04]

You can always just assume that you had 100% RH at the last place you had a coherent gas/liquid interface. Further assume that the cooling that typically occurs in the wellbore has caused condensation and you're still at 100% RH at the wellhead. Then you can use your wellhead temperature and pressure along with GPSA Field Data Book figure 20-3 to estimate the water content.

I've used Meeco Water Boy meters, Panasonic Panameteric meters, and Draeger tubes and all have confirmed that the 100% RH assumption is very good and it will give you data at least as accurate as the commercial meters.

David

 

[dcasto]

or, are you talking Hydrocarbon dew point? In that case, gases from conventional oil and gas wells are at or near equalibrium. But to know for sure, it takes a gas chromatograph to get the composition. We pay under %20 to ran an analysis.

 

[VikingUK]

Basically I need to know roughly the water content to calculate how much MEG to inject to prevent hydrate formation. Thanks David I will give your suggestion a go.

 

[dcasto]

water content in lbs/MMCF is equal to
= (10^(3.5551-0.94283*LOG(psig+15)+0.01576*(Temp Deg F-31)))

The GPSA databook goes through the Hammerschmidt's equation for inhibitor injection in Chapter 20.

 

[VikingUK]

what does ^ denote ? is it divide ?
I assume * is multiply ?
and should there be ( ) around the entire equation ?

I will look up the equation in GPSA... thanks

 

[JLSeagull]

Exponential - raised to the power of, ...

 

[Latexman]

/ = divide.

I thought math was the universal language?

Good luck,
Latexman

 

[zdas04]

The equation above has some serious problems even if it did get a star. ASTM publication D1142 has the technique that is used in the GPSA 20-3 chart and every other version of this data that I've ever found published. The ASTM data deviates significantly from a semi-log straight line. The empirical equation above yields a semi-log straight line. There is a range of data that the equation matches the line very well, but that range is different for every pressure.

I've attached a chart that shows the impact of this difference for 4 pressure curves. At 0 psig there is an excellent correlation between about 20F and 90F. At 2,000 psia the curves only match at one point.

I would be reluctant to use that equation for anything that involved spending money. I've spent many months of effort trying to develop an empirical equation that matches the ASTM data and have been unsuccessful. The closest I've come has been a set of three equations (each is a ratio of logarithmic polynomials) that each match a range of temperatures. I don't give those equations away.

David

 

[dcasto]

The GPA is working on a universal equation Here is a link to a spreadsheet for more detailed cals. But in normal 50 to 120 degree temp range, I call it close enough.

http://www.bing.com/search?q=McKetta+equation&first=11&FORM=PORE&adlt=strict

Project 074 Effect of Gas Gravity, H2S, and Salinity on the Water Content in Natural Gas

Objective: The goal of project 074 is to extend the work completed under project 032 to mixtures with heavier
hydrocarbons (e.g. gravity > 1.0), mixtures with CO2, H2S, and to verify the salinity correction.

Background: The revision of the McKetta chart and the new thermodynamic model are based on the data measured
under project 032, data measured under other GPA projects (e.g. RR-174, RR-187, ongoing project 987), data
provided by member companies (e.g. TP-28) and literature data. This work has identified the need for follow-up work
beyond the original scope to fill-in gaps and address inconsistencies in the existing data. For example, some recent data are inconsistent with the historical correction for gas relative density.

Value to Industry: Project 074 will provide additional data to validate the new model, particularly for gases with
gravity > 1, sour gases and gases in contact with brines. In particular, the project will address water content in a) propane and butane b) mixtures of hydrocarbons, c) mixtures with CO2, H2S, and d) mixtures in contact with brine in
order to validate model parameters and predictions.

 

[zdas04]

Just like any empirical equation, as long as you understand the limitations, it can do a good job. I thought the spreadsheet was encouraging that they have a different equation for high temp than for low temp. That was the same conclusion I reached. I also like their disclaimer that it doesn't work above 70 bar.

The GPA project is going to be interesting to follow. When you compare GPSA Field Data Book Figure 20-3 (Water in Natural Gas, nearly linear on semi-log and gentle curve on log-log vs. pressure) to Figure 20-4 (Water in CO2, severe hockey stick on log-log) and Figure 20-5 (H2S, gentle hockey stick on log log) it is easy to see that a generalized equation for mixtures is going to be a tough problem. My guess is that the end result will require a serious program to integrate all the parameters and that there won't be a closed-form solution for all the areas they're studying. I could easily be wrong, but this one looks really difficult.

David