6 replies to this topic

[hansgupta]

Dear All,
This is regarding a new reciprocating compressor installation which is already piped up and in place as per the attached pdf file. I am trying to evaluate if the current piping arrangement, control valves and other control instruments are sufficient for the safe and stable start-up and operation of this compressor.
This is 2.7 MMSCFD recip compressor on an FPSO with suction pressure of 8.5 bara and discharge pressure of 20 bara. The produced gas from the gravity separator is almost always more than 2.7 MMSCFD and the excess gas is flared. In case of lower production of gases, the capacity control line (or the recycle line) is to be used to control the suction pressure.
The problem that has been realized in the current design is that the flaring of the gas is possible only from the suction KOD and therefore suction KOD always need to be in service even if the compressor is not operating. i.e., if we want to operate the gravity separator to continue oil production without running the compressor, we need to keep the suction KOD in service, to be able to flare the produced gases.
The compressor vendor requires that there should be a new line from upsteam of SDV-2 to the flare to isolate the suction KOD from gravity separator. The current line from suction KOD to flare will not be required in that case. As the system is already built, I am trying to find out if it is possible to start/up and operate the compressor that way it is right now, or by making minimum possible changes.
I am looking for suggestions for a startup procedure and control philosophy to achieve this objective.
Your recommendations are highly appreciated.
Thanks

Attached Files

•  Recip_Compressor_Controls.pdf   18.08KB   452 downloads

[Art Montemayor]

hansgupta:

Allow me to make some comments first:

1. Please state WHY “the compressor vendor requires that there should be a new line from upsteam of SDV-2 to the flare to isolate the suction KOD from gravity separator”. We can’t discuss startup, operating, or shutdown procedure without knowing that.
2. Also, state EXACTLY where (or better yet, show it on your sketch) the proposed line to flare is going to be located and how it is to be actuated.
3. Your sketch shows a centrifugal compressor. Are you sure you have a reciprocating compressor? Please confirm. This is an important point and the correct symbology should always be used to ensure correctness of information.

It is possible to start/up and operate the compressor that way it is right now – however, some changes or additions must be made and these should comply with the existing startup and operational philosophy. If, indeed, you have a reciprocating compressor then you can startup on total recycle, employing the existing capacity control valve – which is nothing more than a recycle valve. This assumes you have valve capacity for the compressor. In order to initiate this type of startup, you must first purge any air or inerts from the compressor’s system by bleeding in natural gas from the gravity separator. This can be done manually or instrumented. You must control the amount of pressure in the suction KOD such that there is a positive pressure there and you don’t suck a vacuum. Once the compressor is running on total recycle (100% capacity control – with nothing going downstream of the discharge scrubber), then you can start to feed more gas from the gravity separator at will or on instrument demand. The compressor will come up to capacity (in accordance with what is available from the gravity separator) and will automatically start to feed the pressurized gas downstream as it becomes available.

[Dev 009]

hansgupta:

Allow me to make some comments first:

1. Please state WHY “the compressor vendor requires that there should be a new line from upsteam of SDV-2 to the flare to isolate the suction KOD from gravity separator”. We can’t discuss startup, operating, or shutdown procedure without knowing that.
2. Also, state EXACTLY where (or better yet, show it on your sketch) the proposed line to flare is going to be located and how it is to be actuated.
3. Your sketch shows a centrifugal compressor. Are you sure you have a reciprocating compressor? Please confirm. This is an important point and the correct symbology should always be used to ensure correctness of information.

It is possible to start/up and operate the compressor that way it is right now – however, some changes or additions must be made and these should comply with the existing startup and operational philosophy. If, indeed, you have a reciprocating compressor then you can startup on total recycle, employing the existing capacity control valve – which is nothing more than a recycle valve. This assumes you have valve capacity for the compressor. In order to initiate this type of startup, you must first purge any air or inerts from the compressor’s system by bleeding in natural gas from the gravity separator. This can be done manually or instrumented. You must control the amount of pressure in the suction KOD such that there is a positive pressure there and you don’t suck a vacuum. Once the compressor is running on total recycle (100% capacity control – with nothing going downstream of the discharge scrubber), then you can start to feed more gas from the gravity separator at will or on instrument demand. The compressor will come up to capacity (in accordance with what is available from the gravity separator) and will automatically start to feed the pressurized gas downstream as it becomes available.

Do agree with Art expert opinion regarding start up of compressor, but what I understand from your post is that you want to isolate the suction KOD during oil production without using compressor.
Please go through the scheme which depicts startup of compressor without using KOD too. & Yes vendor is right in sense that any surge occured during such conditions should be controlled by providing a flare line from gravity separator. I tried to show it as per my little knowlegde .
Also the suction to compressor may be taken from downstream of gas cooler (which indeed will be better) marked wrongly my self.

Thanks
dev

Attached Files

•  modified_scheme.pdf   81.55KB   215 downloads

[hansgupta]

Art:

Thanks for your comments..Please see the revised attachment showing the vendor recommended changes.

1) This is a reciprocating compressor. I, by mistake, have drawn it like centrifugal machine.

2)The current control philosophy from the vendor requires that:

a) Initially SDV-2 and SDV-5 are shut and the produced gases are flared through the "new" line from the gravity separator. (the line which we don't have right now).
b) purge the equipments between SDV-2 and SDV-5
c) Build a pressure of 3.5 bar-a by opening SDV-2, opening recycle valve and by using PV-5610.
d) Shut SDV-2 again and bring the system to " Ready to start motor".
e) Press "START Motor" button. Compressor is now in ready to "LOAD" condition. SDV-2, SDV-5 PV-5610, PV-5606 are closed right now. PV-5505A is open in auto mode.
f) Pressing "LOAD" button will open the suction and discharge SDVs,i.e., SDV-2 and SDV-5. and bring the PV-5610, PV-5606 and PV-5505A to "AUTO" mode.

So, the main objective is to start-up and operate this compressor the way the system is right now (without making the changes recommended by vendor as it is very expensive now to add an extra line).

Please suggest the required changes you were referring to achieve this. The objective again is to make minimum (hardware) changes.

Thanks
Hans

Attached Files

•  Recip_Compressor_Controls_Rev1.pdf   253.11KB   244 downloads

[sepower]

Hi My name is Patrick and I am currently working on a reciprocating compressor project and I would like to know whether you would have a control philosophy that I could look through so I can get an idea on what I am going to need on the electrical side of things. I have never designed a gas compressor before but I have extenive diesel and standard gas engine experience, But I am learning quick that gas compressors are completly different. Your assistance would be greatly appreciated. my email address is patrick@sepower.com.au thank you

hansgupta,

I have one concern related to the location of SDV-8 at the compressor suction, is it confirmed by the vendor that it will not have any impact on the compressor i.e. during shutdown by pulling vacuum.

Thanks

[Technocrat]

Dear Hansgupta,

You can keep the suction KOD in line even if the compressor is not working. By closing relevant valves you have to ensure that the gas is not flowing downstream the suction KOD. You can control the pressure of gravity seprator and suction KOD through PV-5505B.

Regards.