Minimum Flow Protection Requirement After Pump Upstaging 3

[ghamsa]

We have total of two (2) pumps that are connected in parallel. The discharge outlets of these pumps are connected in 14” common discharge header. One pump is in operations while the second pump is on stand by.
These pumps share one common minimum flow protection flow control valve (FCV), which is installed on the 14”common discharge header. This FCV is set at 1000 GPM.

We are planning to re-stage these pumps from 12 stages to 16 stages. My question is: Do we need to increase the minimum FCV setting or leave it as is?

 

[Artisi]

Cannot see any requirement to change the setting.

Naresuan University
Phitsanulok
Thailand

 

[ccfowler]

ghamsa,

Keeping the same minumum flow setting will keep the same protection from low flow issues at the suction and first stage, but it may not deal adequately with low flow temperature rise through the additional stages. You should take this into consideration.

Also, by adding more stages, the FCV will be handling a greater pressure drop. You should consider whether the FCV can withstand this more severe duty.

 

[MortenA]

Not unless the shut-in pressure from the restaged pump is higher than the design pressure for the discharge piping?

Best regards

Morten

 

[ghamsa]

Thanks all for your contributions. To make the situation more interesting, please read below:

We checked the piping design pressure and found it less than the pump shut-in pressure. But, the operating philosophy of the pumps will be changed. Both pumps will operational, no more stand-by pump.

Now, do we need to increase the FCV setting or leave it as is, although 2 pumps will be running at the same time?

Thanks

 

[BigInch]

Since you will be running two pumps your flowrate will tend to double which will probably also require a higher system pressure. If you continue to use FCV, that setting will obviously have to increase to accound for the new flowrate.

Your probable increase in system flowrate and pressure requirement suggests either of two options might be possible,

1.) reduce rpm just enough to give you a shut-in discharge pressure less than the pipe allowable and install a PSV to protect for overspeeding your new rpm setting.

2.) put in a length of piping with a pressure allowable > equal to shut-off head from pump up to the FCV with a pressure override that prohibits delivering pressure downstream of the valve higher than that pipe's allowable. You should still have a PSV on the lower pressure piping in case the FCV pressure override fails.

You will have to do a new system analysis to find the new pressure and flowrate to see which option may be the best for your system.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[ghamsa]

Biginch......thanks for your replay

Our pumps are constant speed driven. In my post I meant by FCV is the minimum flow protection control valve (FCV). Please see my post above. This FCV is supposed to open during low flow conditions only when ever the pump disharge flow rate reduces to 1000 GPM.

Since these pumps will be restaged from 12 stages to 16 stages and both pumps will be operational, does the FCV setting needs to increase or remains at 1000 GPM?

Please read my post above to see the orignal pumps system set up.

Personally, I believe that the FCV setting should remain as is (1000 GPM)

Thanks all for your help

 

[Artisi]

What was / is the function of the FCV in the system, is it to limit the flow or to limit increased discharge pressure should the flow be restricted at the deliver point?
From you posting it seems as though it is to limit pressure.

Adding another 4 stages will greatly increase the "shut-head" on the system and as you have noted, to beyond the rating of the system.

You need to do a hydraulic study to find the calculated operating conditions of the system, ie, produce a system curve to give you an idea of the new duty point for the system (2 pumps operating). From here you will see the new flow rate and operating head (pressure) allowing you to check the pipe work integretary at the new duty.
From the system curve you can then also decide at what point - flow / head - you need as a set point for the FCV to safeguard the system.

Also you may also have to consider the response from BigInch
"2.) put in a length of piping with a pressure allowable > equal to shut-off head from pump up to the FCV with a pressure override that prohibits delivering pressure downstream of the valve higher than that pipe's allowable. You should still have a PSV on the lower pressure piping in case the FCV pressure override fails."



Naresuan University
Phitsanulok
Thailand

 

[ghamsa]

The function of the FCV was and will remain to limit increasing the discharge pressure when the flow is restricted at the delivery point.

After adding 4 additional stages, the shut-head will remain BELOW the system rating. Therefore, the FCV setting should not changed even if 2 pumps are operating at the same time. Please advise.

Thanks all for your support on this matter.

 

[Artisi]

Am I missing something or just confused.

You stated that the shut head was higher than the pipe rating (I now assume you are refering to the shut-head pressure of the pump/s without the benefit of the FCV).

Then you say that it's below the system rating (again I now assume you mean the FCV will limit the system pressure down-stream of the FCV)

If the FCV is working and set correctly - yes the pressure down-stream of the valve can probably be maintained below the pipe rating - provided it can handle the increased flow from 2 pumps which, depending on the static head friction losses etc, etc, could be twice the flow rate, ie 2000 gpm.

Re-interating my earlier comment- you need to study the whole system and not just focus on the FCV.



Naresuan University
Phitsanulok
Thailand

 

[MortenA]

Ghamsa

If your restagde pump has a higher shut in pressure than the design pressure the i will suggest that

A: Re-evaluate SP with regards to the new pump curve.
B: You need a safety valve with discharge return to preferrable your source (if its a vessel or similar) or to the suction side of the pump. The PSV SP shall be the design pressure and the capacity equal to the capacity of both pumps running at the same time at the SP+10%. If the shut-in pressure is less than SP+10% then i would assume that a "D" size valve is OK.

The (mechanical) safety valve is for system protection whereas the minimum flow valve is to protect the pump against overload.

Best regards

Morten

 

[BigInch]

I believe this is an FCV that controls FLOW into the discharge line, but this still has not been explained fully.

<img src="

http://img156.imagevenue.com/loc424/62202_pump_config_122_424lo.JPG"

alt="image hosted by ImageVenue.com">

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[BigInch]

Got it..

Ghamsa, you still have not defined the configuration clearly, please indicate if it is more like Fig.1 or Fig.2

Figure 1 represents a common "FCV" directly controlling flow to the downstream pipeline. Flow through pumps and downstream pipeline is reduced.

Figure 2 represents a common Recycle line installation that diverts a portion of flow back to the pump suction. Flow through pumps is maintained, flow going to downstream pipeline is reduced by the amount diverted to the pump suction.

We will wait for your answer.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[ghamsa]

MANY THANKS: Biginch

Our pumping system is represented by FIGURE 2. I will briefly explain the system again in light of FIGURE 2.

BEFORE:
These 2 pumps (A and B) are constant-speed driven and equipped with 12 stages each. Pump A is in operations while pump B is on stand by. The control valve in FIGURE 2 represents the minimum flow recycle valve, which is set at 1000 GPM. The function of the FCV was and will remain to limit increasing the pump discharge pressure when the flow is restricted at the delivery point.

AFTER:
Both pumps (A & B) are upstaged from 12 to 16 stages. Both pumps will operate at the SAME time. I will assume that the PUMP shut-head pressure AFTER restaging to 16 is below the system piping design pressure.

Now the question:
Does the minimum FCV setting needs to be increased or remains as is (1000 GPM), considering the facts that both pumps A & B operate together and are equipped with more stages?


Once again thanks for you all

 

[BigInch]

With two pumps running, you will have to maintain minimum flow in both, therefore you will have to increase the minFCV capacity to 2000 gpm.

The pressure at the pump discharge nozzle will not be "limited" by the minFCV. The discharge pressure at the pump will always be the equivalent pressure for the head developed by a pump corresponding to the flowrate going through that pump at any given time. Note! If the flows through the pumps go low and approach 0, you will tend to develop full shut-off head and the pressure equivalent of that in the pump discharge manifold. As flowrate through the pump increases, the manifold pressure will tend to reduce as head reduces with the increased flow, which it will do according to the values shown on your pump curve.

Therefore, the maximum pressure that could be developed in the pump manifold is the pressure equivalent of the shut-off head whenever flow approaches zero through the pumps, as would occur in the case of a downstream pipeline flow restriction. Your minFCV should then open to try to keep your pumps running at less than the max operating temperature, but the pumps will tend to produce maximum pressure at that time.

When the flow leaves the pump, it is split, with one portion being diverted back to the pump suction line and the other portion going down the pipeline. The flows into the downstream pipeline and into the recycle line will be proportioned corresponding to the relative resistance provided by the downstream pipeline verses the resistance provided by the recycle control valve (and recycle pipe diameter), but the flow through the pump(s) will tend to remain at the same rate, provided that your recycle line has sufficient capacity to pass the additional diverted flow and perhaps allow passage of up to the full flowrate of both pump flowrates added together. If your pipeline is operating at pump BEP flow, and the recycle line takes the additional diverted flow, the pump discharge head will not increase. If your recycle line will not take the additional diverted flow, your pumped flowrate will decrease and your discharge head & pressure will tend to immediately approach shut-off levels.

With two pumps running, you can potentially double the flow, if the downstream pipeline provides little resistance. What you will typically find is that the resistance of the downstream pipeline increases significantly when you try to double the flow, hence the flow will not double and you will have a final flowrate in the downstream pipeline somewhere between the flowrate of 1 pump and the flowrate of both pumps added together.

NOTE! For your strategy of increasing the flowrate in the downstream pipeline to work, you MUST NOT be operating at maximum downstream pipeline pressure NOW. That would of course indicate that the downstream pipeline is at maximum flow capacity right now, hence trying to double the flow in that pipeline would require additional inlet pressure (pressure at the pump manifold), which is something that 2 pumps in parallel will not provide.

Hope that helps. Let me know if you'd like to know something else. Summary: Double the capacity of the FCV and the recycle piping (you may need to increase it's diameter too).

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[MortenA]

IMHO you will also need a PSV if the shut-in pressure is higher than MAWP assuming of corse that there is a valve than can close on the pump outlet (not nessesarely an actuated one.

API 521 sec. 3.10.2 states that "No credit should be taken for any favorable instrumentat response. Normal valve position is the expected position fof teh valve prio to the upset..." API 14C (applicable for offshore installations) recommnds a PSv on all pumps also centrifugal if they are capable of exceeding the MAWP.

Best regards

Morten

 

[katmar]

While I agree with BigInch in that I believe you need to increase the set point on the FCV, I think it may be worse than what he has indicated.

You are making two changes to your system, i.e. two pumps running instead on one, and additional stages in each pump.

If you double the set point on the FCV to 2000 gpm then each pump will deliver 1000 gpm, which is the rate that a single pump would have delivered with a closed main take off under your old system conditions. The additional stages will modify the pump curves. The head corresponding to 1000 gpm for the pump with the additional stages may be different from what it was before. If you can tolerate this additional head that is fine, but presumably the 1000 gpm level was selected to give a particular head. If you want to limit the take off piping to the same head, you will have to check on your new pump curve to see what flow rate corresponds to that head. And then double it because you have two pumps running.


Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[BigInch]

Harvy, I totally agree. I really only discussed the "minimum worse case". If the head goes much higher than it is now, it could signal a very severe service for the minFCV, depending on the product and if the recycle is to be operated for any length of time.

In fact, it could be worth doing a bit of a flow study to sort out the order of ramping up in the best sequence and/or best settings. I can see where recycle 1 / bring up 1 pump running/ put 2 on recycle / ramp up 2 won't work well.

It might be better if 1 is running full, to cut back to 1/2 then start 2, or shut down alltogether and bring up both etc. There are some setting combinations that might work better than others, and some that might not work well at all.

I can also see where an FCV on the common discharge could be beneficial for getting some better control on the flow delivered into the pipeline, rather than trying to work with a combination of flow into the recycle lines and coordinate that with what the pipeline wants to do too. The recycle and pipeline flowrate desired will otherwise have to be very carefully coordinated because you can't adjust one without affecting the other.

If you do a simulation now, you could get a handle on those things, rather than build it first and be totally frustrated and/or disappointed with the as-built performance.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[Artisi]

An idea of the pump/s flow and head static head, pipeline length and system configuration would be nice - a pump curve would also be handy. We are all working in the dark at the moment assuming much and maybe on the completely wrong track.

Naresuan University
Phitsanulok
Thailand

 

[ccfowler]

ghamsa,

If the system adjacent to the pumps is represented by "Figure 2" above (courtesy of BigInch), then you are running the risk of creating a very unfavorable NPSHa vs. NPSHr situation for the pumps. The increased flow velocity in the suction piping is potentially bad due to greater pressure drop and greater turbulence. This could adversely affect risk of cavitation damage and reduced pump capacity due to unfavorable flow velocity profiles. Increasing the recirculation flow from the FCV to the pump suction piping probably will make the situation worse unless the suction piping is suitably rearranged (and probably expanded).

Also, since we do not know the actual pump performance characteristics, it may be wise to consider whether the two pumps can actually be made to run in parallel. If one pump is running, and the second pump is to be started, can the second pump develop sufficient head to initiate flow, or will the second pump simply warm the fluid through churning. (I don't believe that we know what fluid is being pumped.) If there isn't much margin avaliable at shut-off and flow can't be established very quickly, then churning in the pump can raise the fluid temperature quickly decreasing the fluid density, and then, the discharge pressure never can get high enough to overcome the existing system pressure.