4 replies to this topic

[manchester]

I am designing a De-Oxygen Tower for water injection facility. This tower using fuel gas to strip oxygen from sea water.This tower is designed with 9 m height, 0.5 m dia. and 1 m liquid hold up at the bottom. At tower downstream, I planned to install a reciprocating plunger pump,with discharge pressure of 3500 psig.

Senior people in my workplace comment that it is a mandatory to install a centrifugal vertical pump prior to reciprocating plunger pump, to provide adequate NPSHA and to maintain liquid level at De-Oxy Tower at stable level, due to pulsation results if I install plunger pump only at downstream of the tower.

However, I've calculated that NPSHA is enough already without installing this centrifugal pump prior to reciprocating pump. The remaining problem is liquid level disturbance due to pulsation of reciprocating pump. However, I think this problem can be solved by installing a pulsation dampener at De-Oxy tower downstream prior to reciprocating pump.

Am I right? Or wrong? Needs advice from this forum expert due to my junior experience in this matter.

Edited by manchester, 21 October 2010 - 07:13 PM.

[Art Montemayor]

Manchester:


I have gone through the process evaluation of a similar application in the last year, so the details of what I found and applied are still fresh in my mind. I’ll comment on your post directly in RED:

Senior people in my workplace comment that it is a mandatory to install a centrifugal vertical pump prior to reciprocating plunger pump, to provide adequate NPSHA and to maintain liquid level at De-Oxy Tower at stable level, due to pulsation results if I install plunger pump only at downstream of the tower.
The centrifugal pump does not have to be vertical. A simple, conventional horizontal model will do just as well. I selected a hermetic, sealed model. This type of pump ensures that the NPSHa to the PD injection pump will be adequate. There is no concern for PD pump suction pulsations. However, the centrifugal booster pump has to pump an excess of fluid constantly in order to ensure sufficient fluid to the PD. This results in having to have a recycle flow back to the source 100% of the time. The system is inherently energy inefficient.

However, I've calculated that NPSHA is enough already without installing this centrifugal pump prior to reciprocating pump. The remaining problem is liquid level disturbance due to pulsation of reciprocating pump. However, I think this problem can be solved by installing a pulsation dampener at De-Oxy tower downstream prior to reciprocating pump.
If you have sufficient NPSHa to satisfy the PD injection pump requirements, then there should be no problem using your proposal and I would not bother with a dampner on the PD pump suction. HOWEVER, make sure that your calculations and suction conditions are correct for the PD pump. You cannot afford to make a mistake here with respect to the PD pump’s suction requirements. I would normally not be concerned with maintaining a constant liquid level in your tower; HOWEVER, be aware and take into careful consideration that you are employing an “inordinate” type of tower (L/D = 18) and you should allow a generous dead band for your tower sump liquid level operation.
You fail to mention WHAT TYPE OF PUMP CAPACITY CONTROL you are proposing. This is a vital and important design point when you employ a PD type of pump directly. This is a key handicap in using a PD pump. You are normally limited in the type of pump capacity control you can employ – and this affects the liquid level in your source (the tower sump). With this type of pump you are limited to using: (1) speed control; (2) stroke control; (3) stop/start control.
Economics and feasibilities usually do not permit the first two options; therefore, you are left with only stop/start control of the pump’s motor. This aggravates the NPSHa and the height of the liquid level variance in the tower sump. That is why I am cautioning you to be very careful in checking out your calculations and your tower sizing specifications. Also, bear in mind that you may have to consider unloading the discharge pressure of the PD pump when stopping and starting in order to not tax the amperage rush in your motor windings. Don’t forget the basics of operating a positive displacement machine!
Good Luck.

[manchester]

Hallo Master Art. Thanks for your kind reply. This is very good for me to learn from a senior engineer like you.
1.Control system
I attached the control scheme of the system. Hopefully you can comment the design. The current design is using speed control (I know you already told me that this is not economic choice, however, the project management already decided to use this scheme).
2.NPSHa
This is a Floatation Production Unit. As you may see in the attached file, the tower is planned to be built at 2nd floor, and pump will be located at 1st floor. I think this will ensure adequate NPSHa for PD pump besides providing adequate liquid level in tower.
3.The people in place are seriously considering the effect of pulsation of PD Pump to –what you called- “inordinate” tower. This is why I am asking how to overcome this issue. I would consider review my tower calculation. I would consider to increasing tower diameter to provide bigger liquid hold up.

Attached Files

•  Control scheme.xlsx   67.42KB   71 downloads

[Art Montemayor]

Manchester:

Something basically is confusing me. You state you are designing the system discussed and that there are "Senior people in my workplace". How can you be a student (you have posted this in the Student Forum) and responsible for an industrial system? Please explain this for my peace of mind.

Please study the remarks and markups that I have made to your Excel Workbook. You will note that I bring up some very practical and commonplace items. As I stated, I have been there and done this before. The suction side pulsations are a minor and unimportant item, in my experience. I worry more about stopping and starting this pump against such a high head of discharge. The potential damage to a pump's crank mechanism and the danger of the results are of major concern.

You can always make the sump of your tower a bigger diameter than the tower itself. This will add surge and level capacity to the system and allow the level control to function appropriately. In my opinion the use of a flowmeter device to control the speed of a PD pump is short of inpractical. This type of control is really a manual operation that requires 100% human supervision. I can't imagine that you could guarantee that the liquid flow into the pump will always be constant. That simply is not a real life situation. The flow, in my opinion, will always be varying - more or less - all the time.

Attached Files

•  Disposal Pump Control Schemes.xlsx   125.78KB   97 downloads

[manchester]

Dear Art,

I've gone through site visit to the construction location, and I found that there will be no sufficient space between De-Oxy tower and PD Pump.

Hence, I think it is a mandatory to provide centrifugal pump to provide sufficient NPSHA to PD Pump (you may see the sketch attached). I need your help expertise to review my control scheme (attached). Is the system already sufficient?

Attached Files

•  scheme control-1.xlsx   14.24KB   44 downloads

Edited by manchester, 03 November 2010 - 12:41 AM.