Suction Lift Pumping 3

[Milkboy]

Hello Group

I have a query regarding suction lift pumping

The pump is a Double Suction Centrifugal with 2x Mech seals.
The Pump is primed with a system which fills the pump casing with water ready for pump start up.

So the pump is full and it beings to lift the water from the reservoir.
Where does the pump / fluid get its NPSH from?
Being a pump novice, I always understood you have to have a head? Does the priming system keep a pressure at the suction?

My next query is regaring the 'vacuum'
The pump maker says the stuffing boxes / seals will see vacuum on start up.
Does this mean all the fluid placed there by the priming system will go move out and let the seal run dry?

My thoughts were if there was a vac the water would try and fill the space?


This pump always seem to have a dry running seal issue.
The air is/should be vented when priming but it seems like
some is always present.
All joints are thought to be tight (after initial problems)
Could air / gas be evolving somehow??



Any advice appreciate

Any URLs or pdfs available on Suction Lift pumping and the like would be appreciated also

Thanks






-
Milkboy

 

[BigInch]

You're right, the pump needs a suction head. Its just that lift pumps don't need as much as other pumps. In fact, they need less than what atmospheric pressure can provide to them, if the pump was placed at water level.

You start counting NPSH from 0 absolute pressure, so if there is not a vacuum over the lake or reservoir, you're starting with atmospheric pressure 14.7 psig, which is equivalent to about 33 ft of head. If you were pumping fresh water with the pump level with the sea, you would start with an NPSH available of 33 ft and subtract your flow losses, vapor pressure and lift above the water's surface from that 33 ft. Pumps arn't perfect, so most of the time they will not lift that whole 33 ft and they require some NPSH too, so most of the time you will find them actually lifting from around 21 to about 28 ft or so.

As you raise the pump higher than water level, you lose more and more of that 33 ft of atmospheric lifting capability and eventually you get to the point where it becomes less than the NPSHr of the pump and you start losing the draw.

The seals will see a vacuum, however after priming, when water finally fills the suction line and enters the pump, they should not be running dry. If the pump were located at an elevation close to its lift capability, it would take longer to prime and run dry longer. If it was located at an elevation higher than its lifting capacity, it would run dry and more likely than not, quickly burn the seals.

If you try to lift water too high via suction, it is possible that you are reaching the water's vapor pressure, in which case low temperature boiling will occur and create vapor and loss of suction. It is also possible that some air may be entering your pump suction, if it is not submerged properly and it is pulling air into the intake.

Any of those sound like something you've seen?


BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[Milkboy]

Thanks BigInch

I will question the height the fluid is being lift first of all.

Basically, we are seeing seals which have symptoms of dry running though to be caused by air trapped in the box.
The pump was primed and vented very carefully and turned by hand to release air trapped in pockets (hopefully) but we still see a 'dry run' seal.


The part I'm unclear on is the stuffing boxes though.

Does a Prime System usually fill the entire pump?

Or does it only fill suction?

Are both methods available as I think I have the former, where the whole pump is filled.


I have heard a figure banded about of 5 minutes for the seals seeing vacuum!

Is this normal?

Does this mean the seals run dry for 5 minutes!?

If the pump is filled totally why would the boxes loose the fluid?


I guess if the prime system only fills the suction side you will have to wait x amount of time for the seals to become wetted. 5 minutes seems awfully long though


Thanks in advance





-
Milkboy

 

[BigInch]

A good priming should fill the pump, such that water could escape from a vent at the top of the pump, if you have a vent located there. Suction lift pumps should have the capability to accomplish that themselves. 5 minutes can feel like an eternity when you know things are heating up. I'd say that sould be the maximum limit with a forced shut-down following, with less time allowed if temperatures start to run high, so you should probably shoot for well under that. I don't think I'd really be too comfortable with anything longer than 30 sec if its a large unit, but that depends on what temperatures you get and how fast things get hot.

Is there a possibility that you may be boiling dry. How hot are things getting around there? And .. just how high is the lift, and what are the flowrate and power ratings of this pump?

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[JJPellin]

If the pump is double suction, then the mechanical seals will only see suction pressure whether the pump is running or not. If the suction pressure is a vacuum, then the seals will have a tendency to introduce air if they leak at all. The seal manufacture may be able to tell you how much their seal design can take in terms of vacuum. But, if you are seeing seal failures, you already know the answer. Obviously, the seals are under too much vacuum if the faces are running dry from air being sucked in by the vacuum in the seal chamber. The best solution is to put a plan 11 flush on the seals from the pump discharge pressure. If the pump has throat bushings, it should be a simple matter to size the seal flush orifices to keep a positive pressure the seal chambers when the pump is running. Also, with the plan 11 flush, if there is any air in the seal chambers at statt-up, the flush flow will quickly push the air out and fill the seal chamber with water.

Johnny Pellin

 

[Scipio]

Bear in mind unless you have a tapered bore in your seal chambers, without a high point vent in your seal chambers you're going to trap air at the high point, between the seal face and the throat bushing. May or may not be an issue, depending on what, if any, seal flush you have installed.

Try checking the seal drawing, most that I've seen on double-suction pumps actually have a note stating specifically to vent the seal chamber during priming.

Assuming that's all done and priming isn't the problem, then depending on your seal arrangement, I'd agree with JJPellin, you're likely pulling air into the process side of the seal chamber from atmosphere. In either case, his suggestion of a Plan 11 should go a long way to solving your problems, provided the seal chamber has a throat bushing.

 

[Milkboy]

It turns out the seal is a single cartridge seal design with a double balance line face pair.
Like a double cartridge seal inboard faces but without the outboard seal!

I recall a Flexibox RR seal from a while back which was a traditionally balanced design (70/30 I'd guess) but was rated vacuum to 25 Barg.

Could this seemingly special double balance line seal be having adverse effects (letting air in say) under vacuum
as the dynamic o ring is in a wider cavity than a standard o ring groove and can, of course, shuttle axially depeding on the direction of pressure.

Should we just have a standard hydraulically balanced seal?


Regarding the height - an unconfirmed 18ft


-
Milkboy

 

[Scipio]

Not really sure what you mean by a 'double line face pair' seal, but if it's just a single mechanical seal, suction pressure on one side, and atmospheric pressure on the other side, it'll be pulling in air. At 18 feet suction lift you'll be looking at aroudn 1/2 atmosphere in the seal chamber (assuming no pressurized primary flush). Seals always pass fluid from inboard to outboard or vice-versa to lubricate the seal faces, if the net pressure is outboard to inboard, you're pulling in air. Even if it's not accumulating in the seal chamber, if those are wet-running seal faces, you'll burn 'em out that way.

 

[Milkboy]

Scipio

A normal balance single seal is, lets say, 70/30 or '70% balanced'.

With pressure on there is opening and closing force but a slight advantage to the closing as its 70% (as opposed to 50% being thoretical perfect balance)

If it was pressurised from the other side i.e. 30/70, the seal would blow open as the opening force is larger.

The double balance line as a dynamic oring and groove design which gives 70/30 in the normal direction and allows 70/30 to also be achieve in the reverse direction.

How I explained that ok

I was wondering if having this design may be 'assisting' the air to be drawn in as I have seen traditional 70/30 seals o vacuum.

However, from the excellent advice I have been given, I am more leaning towards air being pulled in but not being moved out of the box by an adequate piping plan

(unless of course someone know if the double balance seal is causing this problem)


One final point which I am still unclear in my mind
When you say there will be 1/2 atmos in the chamber, is that liquid in the chamber trying to pull air in.
Its just that vacuum means to me, nothing is present???

So when people say the seals are under vacuum, as long as no air is oulled thru the seal faces, this is fine. The seals are wetted?

Thanks





-
Milkboy

 

[Scipio]

Milkboy,

I'll admit, I've seen a lot of seals over the years, never one iwth a 'double balance line', so I'm not 100% if it's going to see the same problems I'm suspecting.

As for the the vacuum, yep, the liquid in the suction side of the pump will be at about 1/2 atmosphere, around 8 psi absolute. This is assuming atmoshperic pressure of 14.7 psi on the surface of the reservoir, 18 feet below the pump centreline. In this case, vacuum just means a volume of fluid at sub-atmospheric pressure. Unless the pressure is low enough for the fluid to reach it's bubble point, and start boiling off into a vapour, it will stay liquid.

As to whether or not the seals are still 'wetted', I'd be skeptical. The seal chamber will still be full of fluid (assuming the chamber pressure is above the bubble point, or vapour pressure, of the liquid being pumped), but if there's not enough pressure to drive the pumpage across the seal faces to atmosphere for lubrication, you'll have either an inadequate flow, or complete reversal and air being drawn into the seal chamber from atmosphere. In either case, the seal rings won't be getting proper lubrication or heat removal, and are going to burn out.

 

[Milkboy]

Thanks Scipio

-
Milkboy

 

[pumpking]

There may well be other areas to look at here as far as installation/operation of the pump - we used to make lots of self priming pumps (LaBour apprentice !!) and 99% of 'warranty claims' were down to poor installation/operation.

Im reading coorrect that it is an 18 ft lift ? That is quite a high lift for such design pump, but not un achievable. What is the pump suction size and suction pipework diamtere ? What are you pumping, SG, viscosity ? Is there a foot valve in the suction ? On the pump discharge, do you have sufficient vertical pipework after the pump (this type of pump rely on priming liquid to be recycled, and fall back into the pump chamber). There's not an NRV on the discharge ?

 

[Milkboy]

@pumpking

10" suction, 8" discharge, ansi 300lb

SG = 1, Viscosity = 1, i.e. water

Not sure of the suction pipework diameter only that it appears larger and there is a reducer at pump suction

The reducer, it is safe so say is not 5x diameters in length.

12" (est) elbow goes straight into a reducer and then to pump suction.


Foot valve? I must check.

Vert Height. Yes there is a good height of pipework after the pump..

NRV? again I must check but the photo I have I would say there is not an nrv.


TIA

-
Milkboy

 

[Milkboy]

@Pumpking

There is an NRV on the Discharge

Does this only apply to Double Suctions or is this for all Suction Lift Pumps?

Is there an instance where an NRV on the discharge of a Suction Lift Pump (Two Stage) such as when 4 pumps are manifolded together.

**(I thought it was Double Suction but its a Two Satger)


TIA

-
Milkboy

-
Milkboy

 

[Milkboy]

hey group
im trying to get follow up on this so im tagging it again today to show additions to the thread
hope this is ok



-
Milkboy

 

[BigInch]

I thought PumpKing (as was I) waiting for you to answer all the outstanding questions.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[Milkboy]

Sorry. You're right BigInch.
Im making a mess of this thread.

Let me tell you all I now know after eventually getting to go and see it in person


Two Stage Suction Lift Horiz. Split Case

(Not Double Suction as I thought at first)

200m head. 500 m3/hr.

1480 rpm

10" suction, 8" discharge, ansi 300lb

SG = 1, Viscosity = 1, i.e. water

15°C to 35°

The reducer and straight run to suction after an elbow, it is safe so say is not 5x diameters in length.

12" (est) elbow goes straight into a reducer and then to pump suction.

Lift - 1 to 2m. Water level to Pump Centreline.
(18ft is probably the total length of suction pipe hence
confustion with ops)

Foot valve - No

Suction Pipe - Well submerged.

Vert Height at Disch. - Est. 3 metres

Primed by vacuum of -0.9 which has already been changed to -0.5 as it was deemed to much and may have been pulling air in.

As its primed by vac, the stuffing box vents are not used / cannot be used.



NRV on the discharge. Yes!




Does the NRV on the discharge pose a threat to the function of a Two Stage Suction Lift Pump ?

It would seem the NRV is there due to several pumps being connected to one manifold.


Mech Seals, carbon v sic are dry running on the hi pressure side.


TIA






-
Milkboy

 

[Milkboy]

oh shoot.

......and Plan 11 to Box



-
Milkboy

 

[longeron]

I think when you typed double balance line you meant balance diameter. see-

http://www.mcnallyinstitute.com/08-html/8-01.html

There might be a balance line in your multistage pump between the two seal chambers to reduce the pressure seen by the seal on the discharge side of the pump.

Any way you work this problem, somehow you've got a seal chamber full of air due to inadequate inventorying of the pump or a vacuum.

I suggest tying the flush connections on both seal chambers to the top of the casing so that the entire pump can be inventoried. A plan 11 from the top of the casing betwen the first and second stage should provide adequate flush and perform this task well- if the seal chambers to have a balance line between them. If not you'll have to use two plan 11s, one from the discharge to the high pressure side and another from either the discharge with a second set of orifices to further break down the pressure or a plan 11 from the discharge of the first stage.

This may be a case when you need an eductor to initially prime your pump. I've heard of this in some cooling tower systems where horizontal pumps are mounted at or above the water line in the cooling tower basin so that the basin cannot provide enough head to prime the pump. An eductor can be temporarily installed to pull water into the casing from a connection at the top of the pump until the pump is fully inventoried and started up.

I hope this makes sense...