INTERIM REPORT ON RESULTS FROM SITE MONITORING – OAKWOOD, OK. ENOGEX

 

Report Date 6/11/04

 

1.      INTRODUCTION

Four visits were conducted during March and April to the Oakwood gas compressor station operated by ENOGEX. Each visit was scheduled to monitor emissions during the “pigging” operation. Variables monitored were emission flow, tank pressure, separator temperature and pressure, and weather conditions. Gas and liquid samples were taken for later analysis in the lab.

 

2.      METHODOLOGY

 

2.1.   Onsite

2.1.1.      Emission flow was measured using a Magnetrol TA2 thermal dispersion flow meter connected to the tank emissions vent line. Cumulative readings were obtained from the display on the equipment and emission flow rates were registered using a circular chart recorder connected to the meter.

2.1.2.      Tank pressure was monitored using a pressure transducer located at the top of the condensate tank and connected to a circular chart recorder.

2.1.3.      Separator pressure and temperature were read from the gauge and thermometer located on the equipment.

2.1.4.      Weather conditions were monitored using a Davids Advantage PRO weather station. It registered ambient temperature, barometric pressure, wind speed and direction, and UV radiation

2.1.5.      Gas samples were collected using previously vacuumed TO-14 canisters connected to the emissions vent line by a hose and a quick connection fitting. Two types of samples were collected: 1- 6-hour cumulative samples using a Veriflo flow controller and grab samples by connecting the canister directly to the emissions vent line and allowed to fill quickly.

2.1.6.      Liquid samples were collected using Boston bottles submerged into the tank to the middle of its liquid level via a brass collection cage.

 

2.2.   Laboratory Analysis

2.2.1.      Emission gas samples were analyzed using a Perkins/Elmer 5890 gas chromatograph using a modified of GPA 2261 method.  The analysis allows the determination of the composition of air and C1- C5 hydrocarbons in each sample.

2.2.2.      Liquid sample characterization was carried out using the boiling point curve obtained through the ASTM D86 method and the API gravity using a hydrometer.

 

 

 

 

 

 

 

 

 

 

 

 

 

3.      RESULTS

3.1.   The following table presents the results obtained for the analysis of the two liquid samples collected during field monitoring.

 

Visit

03/30/04

04/21/04

Vol %

IBP

24.5

32

10

38.0

35.2

15

43.7

39.4

20

48.5

53.2

30

55.4

61.5

40

61.9

69.4

50

68.6

76.2

60

75.8

82.6

70

82.4

88.1

80

90.0

95.5

85

94.2

97.5

90

97.8

99.7

95

108.4

 

API gravity

77

77

 

 

3.2.   GC analysis results for the samples collected during site visits are presented in the following table. Type refers to the collection method (grab or cumulative).

 

 

Visit

Sample

Type

Air

C1

C2

C3

i-C4

n-C4

i-C5

n-C5

C6+

Remarks

03/30

1

Grab

46.3

7.6

3.5

16.3

5.0

10.3

4.5

3.8

2.9

At arrival on site

2@

Cum

32.2

45.4

5.6

7.6

2.0

4.0

1.6

1.2

0.5

Prior pigging

04/06

1

Grab

68.7

8.3

5.8

8.0

1.8

3.9

1.3

1.2

1.1

 

04/21*

1

Cum

91.3

0.8

1.0

2.7

0.8

1.8

0.7

0.6

0.3

Prior pigging

2

Cum

78.8

5.6

3.2

5.2

1.4

3.1

1.2

1.0

0.5

Pig arrived at the end

3

Cum

86.2

4.3

2.9

3.0

0.7

1.4

0.5

0.5

0.5

After Pig

 

 

@  Gas sample 2  of 03/30/04 was contaminated with methane because in was necessary to purge the tank with dry methane gas to determine if the meter/recorder was working properly prior to the arrival of the pig.

 

*  Gas samples collected on 4/21/04 were contaminated with air due to a faulty seal on the collection hose assembly.

4.      CONCLUSIONS

4.1.    Based on the ASTM D86 and the API gravity, there is very little change in the condensate characteristics.

4.2.   The gas analysis represented by 3/30/04 sample 1 and 4/06/04 sample one are valid representative data for the vent gas composition prior to pigging.  The contamination event on 4/21/04 rendered the “after pigging” samples invalid. We do not have valid “after pigging” vent gas analysis on the Oakwood site, but this can be collected at a later date after a pigging operation without re-piping the tank vent system.

4.3.   U of A model results are promising although calculations were only carried out for methane solubility. A better compositional analysis of the vapor emissions is required to draw any conclusion with regard to the U of A solubility model.

4.4.   Maximum pressure increase above atmospheric pressure in the tanks during pigging on 4/12/04 was measured at 24 inches of water (13.8 oz) and the temperature dropped from 45 0F to 39 0F during pigging. 

4.5.   It was observed that wind velocity and direction had a strong impact on boil-off vent flows.  It should be noted that the vent system was altered in order to install the flow meters and that actual vent flow under normal operations year round are probably slightly less that what was measured during these test periods.  Obviously, the frequency and timing of the unloading of condensate from the tanks effects boil-off also.

 

5.      Recommendations and Action Plan

5.1 Additional data will be collected on four pigging days (two events per day usually) from the Grandview site and then determine if we need to collect additional gas sample data from Oakwood.