Locating Booster Pumps

[BenjieL]

We have a pump that moves oil products through a pipeline. To speed up the transfer operations, it was determined that a booster pump will have to be added. What are the parameters involved in determining where a booster pump will be optimally located?

Thank you for reading.

BenjieL

 

[parsec]

A booster pumps discharge pressure adds directly to the main pump discharge pressure (just as the pump TDH is in addition to the pump suction pressure) . There are typically two main considerations when locating a booster pump in an existing piping system. Both consideartions are based on the relative location in the piping system accounting for the size and amount of piping and fittings and due consideration for elevation changes.

1. The pressure drop between the main pump and the booster pump at the most limiting flowing condition (probably highest flow rate) can not be so large that you have insufficient Net Positive Suction Head (NPSH) for the booster pump to operate correctly. This should consider elevation change between the two pumps and is especially important when the flowing liquid is volatile.

2. The booster pump discharge pressure should be below the design pressure for the piping system at the worst case flowing condition (most likely the lowest flow or ideally, pump shutoff). If it exceeds the design pressure, relief protection may be desirable and/or required by code. If it is impossible (or highly unlikley) to isolate the line, flowing conditions may be considered rather than pump shutoff.

Also, the booster pump location should follow the same rules as any pump such as suction line clear of fittings for approximately 10 to 20 pipe diameters and further (approximately 50 diameters) from the main pump.

 

[BobPE]

Benjiel:

Are you planning on locating the booster pump in the same line that is moving the oil or is it a side stream flow from a larger pipe line?

Bob

 

[BenjieL]

Bob,

Yes ... the plan is to install the booster pump along the
same pipeline.

Thank you for your time.

Benjie L.

 

[BobPE]

Benjiel:

Thanks for getting us more info. You worry me when you say speed up delivery in your original statement. Like parsec outlined in his response, the booster will only add to pressure, not flow. Is it you intention or has it been determined that increases in flow are required? If so, were they hydraulics considered in this evaluation? Depending on the size of your system flows, this is an extreemely complex engineering problem. If flow increases are desired then then location of the booster station will depend on this although great increases in flow should not be expected.

What you are going to have to do is to force the original pump to the right of its curve assuming it has room left to operate to the right of the curve) by using the suction of the new pumps to decrease the head at which the original pumps are pumping against. This movement of the original pump to the right will allow for some type of increase in flow. This flow rate will be the rate of your new pumps. This all makes it complex since usually series pumps are identical in characteristics making the calculations easy.

In doing this you will be operating in an extremely tight window, increasing the chances of failure of both pump systems.

I would look at modifying the original pumps to meet your needs before considering a booster pump.

I hope I helped....

 

[mgp]

I would suggest that you:

1. Add together the pump curves for the main pump and the booster pump, in order to get a third pump curve (as if it was one pump).

2. Establish the pressure loss curve for the pipeline

3. Find flow from the intersection of the two curves

4. Find from main pump curve the pressure available at this flow

5. From this pressure subtract the required NPSH for booster pump

6. With pressure result (delta P) and flow find length of pipe required (e.g. by making a curve with pressure loss per pipe length)- this is where your booster pump should be.

BobPE (or anyone else) please correct me if I am wrong - I did something like this many years ago and I did not quite follow your worries.

Regards
Mogens

 

[BenjieL]

Parsec/BobPE/Mogens,

Thank you all for your inputs. We have a meeting on this tomorrow and I feel that I am prepared to discuss the technical aspects of the issue.

Technically, we do not really expect an increase in the flow per se. The set up is like so:

*** At the shipping end, the pump delivers approximately
4000 barrels/hour

*** At the receiving end (due to friction losses and
what-have-yous), the rate drops down to around
2800 barrels/hour. The pipeline is about 8 miles
long.

What we want to do is simply place a booster pump (somewhere along the pipeline) so that the rate at the receiving end will still be 4000 bbls/hour or approximately anywhere close to the output of the main pump at the shipping end.

Besides installing a booster pump (there are lots of idle pumps in the refinery), is there any other solution to this problem?

Thanks again and regards to all.

Benjie L.

 

[BobPE]

Benjiel:

I was afraid you were encountering problems like that, that is why i was so long winded in my response. The problem you have is that its a law of physics that flow in a system is conserved. What that means is that if you have 4000 bbh going into the system then you must have 4000 bbh coming out the other end. There is now way for that to change unless you have product leaving somewhere else other than the intended point of delivery. A booster pump will not help increase flow in your situation because in theory you would have to make new flow up in the system. Booster pumps usually only increase delivery pressures which are independent of flow.

Good luck....

Bob

 

[parsec]

I agree that clearly the law of conservation of mass must be conserved. However, the premise that you have 4,000 bbh pumpped into the suction side of the exisitng system assumes that the main pump is capable of overcoming the pressure drop on the discharge side. If it is not, the pump simply backs up on its curve until the pump curve intersects the system curve. If the pressure differential is too high, the flowrate decreases below the required 4000 bbh. This is what I presume you are trying to prevent or correct by the addition of the booster pump. Lets not make this too complicated.

 

[BenjieL]

Parsec,

Yes indeed ... this is what we are trying to accomplish by adding a booster pump somewhere down the line.

I really assumed that this was a very simple problem but I guess life is full of surprises.

Again, I would like to thank all of you (BobPE and Mogens as well) for all your inputs in analyzing the situation.

More powere to all of you.

Benjie L

 

[BobPE]

parsec:

I agree with you that the pump is the control. I assumed when Benjiel sadi he had 4000 bbh entering the system, it was read throug a meter and the pump was already on its operating point. So if it went into the system it would have to come out.

Bob...