Lean Amine Pump Placement

[TYLam]

In most column designs, we put the bottom product pump directly downstream of the column. The column height is adjusted (during design stage) to provide sufficient NPSH for the pump. The only exception that I know of is in the regenerator section of a amine plant. The "accepted practice" is to place the lean/rich heat exchanger downstream of the column bottom product and the pump downstream of the heat exchangers. According to a very experienced designer I talked to, the cooled amine is less likely to cause pump cavitation. This puzzles me. What in the amine application makes pumps behave differently. I checked the performance of the lean amine pump in such a configuration (pump downstream of the heat exchangers). According to the operator, the pump cavitates as well. I checked th NPSH. NPSHA is 48 ft. NPSHR is 32 ft. So, the pump should NOT cavitate.

Anybody has a technical explanation of the phenomenom?

 

[Montemayor]

Ty:

Your post is hard to understand. You haven't stated outright that you're experiencing cavitation in your Lean Amine pump, but I'll assume you are.

You also state that it is "accepted practice" to place the Lean Amine pump downstream of the Amine exchangers. I've designed and operated many Amine Stripper/Absorber process applications (since back in 1962) and I can categorically tell you that it is NOT "accepted practice". It is the engineering, logical smart thing to do. You also fail to state that there is an Amine cooler involved, immediately downstream of the Amine exchanger. This is essential for the Amine process in order to ensure that the Amine routed to the absorber is as cool as it can get for maximum CO2 or H2S absorption. I have always endeavored to design my Amine exchanger and cooler for minimum pressure drop loss so that I can install my Lean pump immediately below and after the cooler. That allows my Lean pump to operate as cool as it can. You also fail to state the obvious and normal fact that most (or all) Amine reboilers are TEMA kettle type with the Amine exchanger(s) and cooler directly installed UNDER the reboiler - which indicates that the Amine reboiler MUST be installed in an elevated position vis-a-vis the exchangers, cooler, and Lean pump, giving you a design control on the NPSHa of the system. Another fact that is in your favor is that the Stripper (not "column") operates at approximately 5 - 10 psig pressure, which helps your NPSHa out (since the Lean Amine getting into the Lean pump is NOT saturated, but subcooled).

You don't state that you have an air-cooled Lean Amine cooler, so I have to assume you are cooling with Cooling Water. I have operated many Lean Amine pumps - most of them Goulds Model 3916 - basic cast iron construction - and never had any corrosion or cavitation problems. The cavitation problem is a possibility if:

1. You are not cooling your Lean Amine solution sufficiently to reduce the solution's vapor pressure effect and the effect of the residual CO2 content - remember that both Rich AND Lean solutions carry CO2 loadings);
2. An NPSHr of 32 ft is too high for me. Without knowing any more basic data, it seems like a wrong choice of pump;
3. As said previously, I can control the elevation of the equipment and therefore, lend a positive hydraulic static head to the pump. If your exchangers have too much pressure drop loss, then this is another wrong design.

Your application defects seem to lie both in equipment design and pump selection - based on the scarce basic data you've supplied. If you give us all the specific and detailed basic data, perhaps a better or more accurate analysis can be made of your system.

I hope the above helps you out.

Art Montemayor
Spring, TX

 

[TYLam]

Art,

Thank you for your response. I apologize for not stating the problem clear enough. I involved in a number of plant designs although this is my first time involved in an amine unit design team.

Starting from the amine stripper, the proposed layout of the new unit is:

Amine -> Lean/Rich -> Lean -> Filters -> Air -> Absorber
Stripper Exchanger Pump Cooler

The amine stripper has a kettle type reboiler. We have three other amine units on-site (a refinery) that has the same configuraion. The amine (DEA) takes out both CO2 and H2S.

In most distillation column design, one would place the bottoms product pump directly downstream of the column. Of course, the elevation of the column has to satisfy the NPSHR of the pump. This arrangement allows optimized design of the heat exchangers downstream. My question is: what properties of the lean amine solution makes this configuration undesirable?

The design information I quoted in the original post: NPSHR = 32 ft and NPSHA = 48 ft is from a operating unit with the lean pump downstream of the lean/rich exchanger. NPSHR is from the pump curve and NPSHA is calculated from the vapor pressure of the amine solution. Regardless whether the pump is a good choice or not (it is already there), the pump should not cavitate but it does! A corollary of the question above is: Is there anything else we have to take into account (and how) in calculating the NPSHR of an amine pump other than the vapor pressure of the mine solution?

You did mention "the solution's vapor pressure effect and the effect of the residual CO2 content". I know how to calculate the NPSHR of the pump based on the amine solution's vapor pressure. I would appreciate if you can give me some guidance as how to include the effect of the residual CO2 (and H2S?).

TY Lam

 

[Montemayor]

Ty:

You or any other user of your Lean Amine Pump do not calculate the pump's NPSHr (Required Net Positive Suction Head). The pump's manufacturer is the ONLY one able to do that. I suspect (hopefully) that you really meant NPSHa.

There is no known way to me for taking into account the negative effect of the gas loading in a hot Amine solution. What a designer does for your system is:

1. Elevate the Amine Kettle reboiler as high as feasibly and economically possible in order to obtain the highest static head;

2. Bring down the temperature of the Lean Amine as much as possible as it enters the Lean pump (size the DEA exchanger appropriately to maximize the heat exchanged);

3. Design and obtain the minimum Lean Amine solution pressure drop across the DEA exchanger; this insures the maximum Lean Amine solution positive pressure at the Lean pump suction; this techique calls for real design expertise because there is a fine line and relationship between the heat exchanged and the pressure drop in the shell & tube exchanger(s). Sometimes, if the pump's NPSHr is very high, the designer has to resort to using two(2) amine exchangers instead of one.

I don't give any importance to your unit's design data; the only data that counts is the actual, field data taken under actual design operating conditions. If your pump was working before and it is now cavitating, it is common sense that something has changed - and it probably is in the factors that affect your NPSHa. One common factor could be exchanger fouling or loss of exchanger heat transfer efficiency. Here, I'm assuming everything else is kept constant - such as the Stripper operating pressure and the Lean pumps's suction temperature.

If you can't change the pump out with one that requires less NPSHa, and the situation continues to jeopardize your continuous operation, you have little option but to either increase the Stripper pressure (which I wouldn't do because it increases the corrosion rate and amine degradation) or to install a solution cooler between the exchanger and the pump - which presents another problem of decreasing the NPSHa due to the pressure drop introduced with the cooler. It is always cheaper to change out the pump for a better, suited model.

I wouldn't waste my time trying to compare generic distillation columns with Amine Stripper units. They are two different animals and it's a waste of time trying to generalize between them. The more time and experience you acquire with Amine units, the better you will appreciate the common-sensical design that has gone into them. Amine process units have been in operation since 1935- 70 yrs! - and there isn't anything that has not been tried, experimented and changed in them. The way the unit is configured is according to time and success-tested empirical know-how. I don't believe you or anyone else is going to improve on the existing configuration anymore than I did over the years - and I've operated and modified many units. Trust in the basic design and configuration and stick to the fundamental engineering basics. The system works - and works well and consistantly as it originally was configured.

I hope this experience helps.


Art Montemayor
Spring, TX

 

[TYLam]

Art,

Thank you for taking the time to share your experience. To clairfy, the NPSHr value I quoted is from the pump curve published by the pump manufacurer. It dos surprise me that the mysterious effect from the residual dissolved gases which increases the NPSH actually required was not quantified.

TY Lam

 

[121202]

TYlam,

I suspect that the problem is occurring due to the evolution of dissolved gases when the pressure is dropped upstream of the pump as opposed to the Amine solution boiling. I also do not believe that further cooling of the Amine will be terribly beneficial in preventing gases from evolving. If dissolved gases were not considered in the NPSHa calculations then they will have overestimated the NPSHa.

Rgds