6 replies to this topic

[Santosh Singh]

I am working for waste water project in Russia and we have to treat waste water coming from gas processing plant wells. The water is sucked by vacuum trucks and sent to our plant. This water contains H2S, RSH, Higher HC etc.
Please find attached details scheme for the same.
To reduce VOC emissions it is planned to provide nitrogen blanketing in all the tanks. I am trying to connect all the tanks with common header, so that pressure of all the tanks can be equalized and provide the Nitrogen and vent connection at one tank only. Moreover I will put oxygen analyser in the common header.

I will then put one small blower and send all the gases to activated carbon bed.
Please suggest if this is the right approch.

Regards
Santosh Singh

Attached Files

•  Visio-Scheme for VOC Emissions.pdf   37.75KB   100 downloads

[proinwv]

First, I am pressed for time and have not had the opportunity to study your submittal. But I can tell you that using a common header will work BUT very careful consideration must be given to analyzing the pressure drops in individual pipes as tank vary in pressure and volume. The pressure drop can be of an order of magnitude similar to the controlled pressure.

Also, carefully think about the fact that you will be controlling pressure in only one place. That is where the sensing line is placed. Other pressures can vary greatly as the system attempts to balance pressure.

Without this full analysis you can expect pressure/vacuum swings, as well as discharges to atmosphere, or even ingress of atmospheric air.

There is a paper on this subject that is available on my website.

[Santosh Singh]

Dear Paul,

Thanks for your valuable suggestions.

• We had already purchased the package as whole, and as suggested in the paper (Controlling Vapor space pressure on multiple tanks during blanketing) is it beneficial to connect all the tanks with Nitrogen connection (with one common Control valve) for our system ?.
• I am putting the sensing line at the maximum volume tank, so that fluctuations will be less.
• I am trying to reduce the pressure drop to the farthest tank by proving higher dia pipe size compared to the nearest tank.
• Is it possible to control the common header pressure (common header connected to all the tanks) rather than controlling one tank pressure. ?
• Please suggest whether in such type of system oxygen monitoring (oxygen analyzer) is essential.

Regards
Santosh

[fallah]

Dear Paul, Thanks for your valuable suggestions.

• We had already purchased the package as whole, and as suggested in the paper (Controlling Vapor space pressure on multiple tanks during blanketing) is it beneficial to connect all the tanks with Nitrogen connection (with one common Control valve) for our system ?.
• I am putting the sensing line at the maximum volume tank, so that fluctuations will be less.
• I am trying to reduce the pressure drop to the farthest tank by proving higher dia pipe size compared to the nearest tank.
• Is it possible to control the common header pressure (common header connected to all the tanks) rather than controlling one tank pressure. ?
• Please suggest whether in such type of system oxygen monitoring (oxygen analyzer) is essential.

Regards Santosh

Santosh,

As paul said full analysis of the system should be performed to prevent pressure/vacuum swing, but IMO if you want to use one blanketing system for whole plant, control a common header pressure which connected to all the tanks is better idea. In this way, emptying/filling rates of all tanks should be considered and all simultaneous scenarios as per operating manual should be specified. The sizes of the common header and interconnecting lines should be selected such that based on above mentioned analysis you wouldn't have any pressure/vacuum swing in any tank.
It might be based on high distances between tanks the sizes of the header/interconnecting lines to be so large would press you to use more than one blanketing system in several points to meet the requirements. Also sensing line should be near the tank with highest changing rate of the top space due to emptying/filling of its inventory not near the tank with largest absolute volume. If the configuration would be such that you would concern about the air ingress, oxygen monitoring is essential.

One more think, are the tanks constructed from concrete?

Fallah

Edited by fallah, 20 December 2011 - 04:17 AM.

[Santosh Singh]

Dear Paul, Thanks for your valuable suggestions.

• We had already purchased the package as whole, and as suggested in the paper (Controlling Vapor space pressure on multiple tanks during blanketing) is it beneficial to connect all the tanks with Nitrogen connection (with one common Control valve) for our system ?.
• I am putting the sensing line at the maximum volume tank, so that fluctuations will be less.
• I am trying to reduce the pressure drop to the farthest tank by proving higher dia pipe size compared to the nearest tank.
• Is it possible to control the common header pressure (common header connected to all the tanks) rather than controlling one tank pressure. ?
• Please suggest whether in such type of system oxygen monitoring (oxygen analyzer) is essential.

Regards Santosh

Santosh,

As paul said full analysis of the system should be performed to prevent pressure/vacuum swing, but IMO if you want to use one blanketing system for whole plant, control a common header pressure which connected to all the tanks is better idea. In this way, emptying/filling rates of all tanks should be considered and all simultaneous scenarios as per operating manual should be specified. The sizes of the common header and interconnecting lines should be selected such that based on above mentioned analysis you wouldn't have any pressure/vacuum swing in any tank.
It might be based on high distances between tanks the sizes of the header/interconnecting lines to be so large would press you to use more than one blanketing system in several points to meet the requirements. Also sensing line should be near the tank with highest changing rate of the top space due to emptying/filling of its inventory not near the tank with largest absolute volume. If the configuration would be such that you would concern about the air ingress, oxygen monitoring is essential.

One more think, are the tanks constructed from concrete?

Fallah

Thanks for the reply.

As per the scheme all the tanks are RCC except Slope Oil tanks which is made of Carbon Steel.

Regards
Santosh

[eman3001]

As suggested by the others, a full analysis with modelling is required. This is to ensure that the concentrations in the headspace are capable of being mantained at a constant level. The objective is to minimize contaminant levels by feeding nitrogen and monitoring the oxygen concentration. A dynamic oxygen model is required which can be set up in a spread sheet. I need to know the pumping characteristics of the exhaust pump and an estimate of the contaminants concentration( max levels would do).

Regards

[proinwv]

Santosh, you have been given a lot of good information on this thread.

We wish you the best and it would be appropriate for you to come back later and tell us what you did and how it is working.