Sizing a pump with a elevation in the discharge line. 2

[Ndimo]

Good day to all,
In the process of trouble shouting a mud disposal system, I found that although the discharge point has an elevation below the suction point; thus giving a negative static head, there is a hill in between wich is 100 ft above suction elevation. How could I take in account that hill in the system determination ?
Thanks

 

[Artisi]

The pump needs to be able to develop 100ft head so as to pump to the elevation of the highest point, ie round 100ft. Once the pump is primed,running and the discharge line is full to the highest point and flow is established over the high point a syphon effect may then come into play, this will then change the total head on the pump and introduce a new set of operatinmg conditions.

This has been discussed here before and maybe someone wiill be able to give you a lead into the right area so you can review it.

Naresuan University
Phitsanulok
Thailand

 

[Ndimo]

Would that means I will have to size the pump and motor for starup conditions? ie 100 ft static head more.
If so, this pump would run too far from the BEP once the primed !

 

[Ashereng]

Or two pumps?

 

[vpl]

boubacar

It sounds like you need to draw out your piping system and figure out the total dynamic head for the system.

You need to figure out the head loss for the length of pipe, the total elevation difference, and any pressure changes (remember Bernoulli?)

You might also want to put a vent valve at the high point to make sure that when you first fill the system you get rid of any air bubbles that could disrupt operation at a later time.

I wouldn't worry about the pump not operating at its BEP when first starting and filling the system; just make sure it's designed for the head loss it normally sees. And, as Asherang commented, you can always use a booster or startup pump to first fill the system.

Patricia Lougheed

Please see FAQ731-376 for tips on how to make the best use of the Eng-Tips Forums.

 

[Assumptions]

One should draw a diagram for the system. In the best case, once the line is filled the lowest pressure that could be achieved at the top of the hill is zero PSIA. That would be equivalent to 33.9 ft of water at 4C at sea level. Since it appears that you have a heavier fluid the head in feet of fluid would be proportionately lower.
If the circumstances permit one could prime the system with water and then start with the heavier fluid.

Since your fluid will have a vapour pressure the lowest pressure at the top will be the vapour pressure of your fluid.

 

[Artisi]

The pump will need to be sized to run at its operating point- (to be established) and capable of pumping to approx 100ft to fill the discharge line to its high point. The pump can run left of BEP during the 'filling stage".

Naresuan University
Phitsanulok
Thailand

 

[freefallingbody]

Hi,

As I understand:

The key requirements are:

1. You need to start pumping water/liquid initially upto the 100 ft high hill.
2. Operate at BEP once you got the flow maintained.

Solution:
Select a pump which has "shut off head" slightly more than 100 ft and has BEP near your actual operating pressure. Consult a pump manufactuer or use their software.

Actual operating pressure = pressure drop in the line ( calculate this using equations )+ height of highest point above suction - depth of discharge point from suction.
Ofcourse, you will use the same units of pressure for static and dynamic heads.


Dinesh

 

[Artisi]

freefallingbody

what you have summarised is pretty accurate, the only unknown is any syphon effect which will also effect the actual operating head.



Naresuan University
Phitsanulok
Thailand

 

[freefallingbody]

Artisi,

You are Right.
If the highest point to which water is lifted is say 100 ft and the water is discharged at 20 ft below the suction line ( i..e return line length of 120 ft), I think the static head would be -20 ( minus twenty). Now, if we calculate dynamic head and add to this static head for different flows, we can construct a system curve.

Due to the minus sign of static head, the system curve will start below zero axis and curve up. Obviously this will result in much less head seen by the pump.

My question is now:

1. If the liquid is not water, how the syphoning effect influences the scenario? I have seen some paper mills where high density pulp pumping is done at 8 to 10 % consistency. Some pump manufacturers say that due to high viscosity of pulp, syphoning effect is very much less.

Do you agree?

Regards

Dinesh

 

[Ndimo]

I guess that, the syphoning effect influence would be the same in term of startup head, but the power demand would be directly proportional to the slurry density.
Regards
B Barry

 

[bulkhandling]

There are possibly more tricks here than we thought.

Assume a pump is "correctly" sized to cover all the range from priming to normal operation:
-- to start the pump for priming, high flow low head --> medium flow, medium head --> low flow, high head (fluid reached highest point). After this point, you expect more flow with reduced head, but, head stays the same and flow still low -- syphon not formed. Why? because low flow with low velocity CANNOT push air out. Slack flow after the climax. --- You may develop a start-up procedure to overcome this problem.

Sludge may be non-Newtonian flow. You may consider pump derating in head and efficiency, and at low flow far from BEP, the derating can be considerably greater then normal flow. You may even not able to get the head you checked from your pump curve.

Assume piping is primed and syphon is formed:
-- You need to check the piping system after the climax. At the part of steep slope down, static drop can be considerably greater than piping friction loss -- possibly drop down to vapor pressure and slack flow formed. --Check cavitation and consider orifice station.

For the syphon conditions in downward piping, katmar's posts in my thread "Small pipe discharge: open channel or siphon" may give you some good hint.

Just some thoughts for your consideration. Good luck.

 

[Artisi]

Are you pumping paper stock ?

Bulkhandling is right in that there might not be any syphon effect, you need to do a lot more research ie, flows, pipe sizes etc .

Naresuan University
Phitsanulok
Thailand

 

[Montemayor]

Every thing Bulkhandling has said is correct and proper engineering judgment and procedure. However, one very basic and importnat point has been left out of this very interesting thread: The piping system must be 100% full of liquid. This very basic and fundamental point is often overlooked and all the presumed equations fall apart.

In order to maintain and to ensure that liquid fills 100% of the piping system, a resistance element - usually a control valve is installed at the pipe's outlet on the other side of the hill. And in controlling a liquid inventory in the system, one has inherently introduced an additional head element into the TDH requirement of the pump. Additionally, do not forget to debit the system with the irreversible friction loss; this fact eradicates the ability to obtain 100% "recovery" of the head used to get over the hill. In fact, in real pumping systems there is never 100% recovery of head.

As most have noted -especially Artisi - this is anything but a "simple" problem. What appears to be simple is actually a taxing fluid transport problem.

 

[stanier]

There are a number of ways of lifting the HGL above the pipeline grading. A pressure sustaining valvve at the end of the line will do this however with a slurry the choice of valve is a nightmare. A Red pinch valve, Valvetech ceramic ball valve or eccentric plug valve may be suitable. In any event the device will need on going maintenance.

At Bougainville Copper they installed Warman slurry pumps at the end of the tailings line, connected in reverse. The idea was to run them backwards and generate electricity. Unfortunately the revolution there stopped the project.

Putting a break tank at the top of the hill would work with water but not necessarily with a slurry. Free surface flow could result in high velocities and wear of the pipeline.

 

[bulkhandling]

Without knowing the details of the mud properties and the system configuration, it's a bit hard to tell if this is a hard project or not.
If the down hill pipe is relatively long and flat (especially the last part), hopefully the chance of slack flow is not high. With a pinch valve, throttling or shut close is not a problem. Hopefully the mud does not contain too much large particles so that wearing is not a big deal.

A startup proceure should includes pre-fill of all the piping (air pockets vent out), the pump will run close to design flow at the beginning of the operation and syphoning will be there most probably.

If THERE IS chance for a slack flow due to steep down slope, pinch valve(s) or orifice station should be considered. Be very careful about cavitation. Maintenance for "MUD" flow shall be OK.