Water is gas main due to air from Compressor

[GasDistEng]

Hi-

A recently installed natural gas main is showing evidence of water infiltration. Water is getting into the customers regulator and meter and freezing. The gas main was cleaned and "pigged" after the installation. The system is currently operating at an MAOP of 60 PSIG.

When the main was pressure tested (at 90 PSIG) it is likely that water (in the form of vapor) got into the main from the compressed air (Air compressor).

Using a Psychometric chart I was able to calculate the amount of water at atmospheric conditions (14.7 PSIA). I assumed the Dry Bulb temp was 32 Deg, RH = 99%... etc.. Cameup with 9 lbs.

How would I calculate the water infiltration at 90 PSIG Air? I think I need to use partial pressure?

Any help would be greatly appreciated!

 

[pmover]

think in terms of mass of the source air. use the psy chart for inlet conditions as this chart provides a rough estimate of the amount of moisture in the air at the time of air compression. determine inlet flow conditions and then the amount of water can be determined.

btw, natural gas piping upon completion of pneumatic testing &/or pigging, should be purged of all air using the source of natural gas. once purged of all air, the piping system and source of natural gas should be liquid/moisture free (assuming "pipeline quality" natural gas). if moisture is a continuing problem, then the investigation needs to be directed towards seeking, isolating, and correcting the source of liquid/moisture.

i trust the compressed air system is not connected to the natural gas system - please confirm.

hope this helps!
-pmover

 

[GasDistEng]

Thanks...

We only used the compressed air for pressure testing. We purge out the air with a slug of nitrogen before the natural gas. We are thinking that perhaps the 90 PSIG air was full of water vapor and condensed in the pipe prior to purging.

I was advised in a physics forum that...

"a the ideal gas law, it would just be the ratio of the two air pressures: 90/14.7 * 9lb=55lb"

 

[quark]

Moisture holding capacity of air decreases with increase in its pressure. There is a simple method, suggested by Sarco, to calculate the final capacity of compressed air.

At 32F, 99%RH(??) and atmospheric pressure, moisture content will be 0.004lb/lb(or 0.00032lb/cu.ft) of dry air. If the total moisture content is 9lb as per your calculation, mass quantity of dry air should be 9/0.004 = 2250 lb. Specific volume of air at these conditions is approximately 12.5cu.ft/lb. So volumetric quantity of air should be 2250*12.5 = 28125 cu.ft(?).

If the air pressure is 90psig then new volume of air should be 28125*(14.7/(90+14.7)) = 3948.78cu.ft. Total moisture in this quantity of air will be 3948.78*0.00032 = 1.2636lb.

That means rest of water vapor will be condensed as soon as the compressed air temperature reaches 32F (in aftercooler, receiver etc.)

Further, if your ambient temperature won't drop below 32F then there will be no condensation in the pipeline. Care should be taken during depressurization. If depressurization is instantaneous, the process can be approximated to an adiabatic process and the temperature drop will be significant to cause condensation. If you follow a slow(isothermal) depressurization then you can avoid condensation totally.

The link below gives you a tabulated data of moisture holding capacity of air at various temperature and pressures.

http://www.flourmilling.net/compress.html

There are lot of simplifications in the method I suggested above but this gives you maximum amount of water vapor that can exist in compressed air.

Good Luck,


I wish us all A Happy and Prosperous New Year

 

[GasDistEng]

Quark-

Thanks for the information. I will sit down and digest it.

Before you made your post I sat down and crammed out a calculation. I think I'm on the right track. The amount of water appears low. However, the actual pipe volume is very small. Still I've read that (for example) On a 75F day with 75 percent relative humidity, a 10-hp compressor can introduce 7 gallons of water into a compressed air system.

Instead of trying to type it all out here - I actually took the time to figure out how to use my free web space that is offered as part of my internet account. I posted the calculation there. Please forgive the very very crude looking web page. See link below.

http://home.comcast.net/~timbickford/htdocs/Water_Vapor_In_Pipe-Rev-1.pdf

Thanks..

 

[zdas04]

Guys, the natural gas will be 100% RH at the last temperature and pressure where there was a coherent gas/water interface. Go to GPSA Engineering Data Book (Figure 20-4).

Now if we assume that this 60 psig gas was in a separator with a liquid level upstream of compression at 11 psig and 80F (call atmospheric pressure 14 psia) then Figure 20-4 shows 800 lbm/MMCF. Cool that to freezing at constant pressure and you have 130 lbm/MMCF -- the missing 670 lbm/MMCF is now a solid.

You'll almost never have freezing from a static test after the first month of operation.

David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

The Plural of "anecdote" is not "data"

 

[GasDistEng]

David-

To clarify. I am really only trying to determine the maximum possible amount of water that could have entered the pipe during the pressure test - proir to service. I believe that most, if not all, the water entered the main during the construction phase. It's possible that a poor pigging job could be the culprit.

There are a lot of opinions around the office that the water was from the compressed air. Therefore - I am trying to quantify it.

I've been thinking about my calculation

http://home.comcast.net/~timbickford/htdocs/Water_Vapor_In_Pipe-Rev-1.pdf

I think the mistake is that I used the RH of the air at atmospheric pressure for the heated compressed air exiting the compressor. Could use some help here..

Thanks