Total head calculation

it is true...

You still have to lift the water.

Atmospheric pressure is acting on inlet and outlet.

look at it like a balance with atmospheric pressure on both sides. You still have to do work to lift one side...it's just that by removing all the pressure from one side of the balance you will never get more than atmospheric acting on the other side.

Alternatively ask yourself, how much force can a vacuum exert? Once you have removed all the molecules from a chamber and have a perfect vacuum can you then 'suck' any harder??? No you can't...it's not the vacuum that is exerting pressure...it's the atmosphere on the outside of the vacuum.

Del

Hi Pompie

As del has written pressure works on both sides.

A pressure gauge anywhere on the delivery side measures against atmospheric pressure. If you don't add it you will be below pressure all the way up.

Separate NPSH calculations are however still required. The pump will cavitate when the suction height and friction exceeds the available.

PS. You need to consider terms like 'suck' carefully...yes a pump can 'suck' and try to create a vacuum, but it can only ever suck to a maximum height of 33feet because that's how far the atmospheric presure will push the column of water.

If you had your pump on the moon it wouldn't lift water very far at all as there is little atmospheric pressure.

Consider how a mercury barometer works... a vertical evacuated tube has it's open lower end in a dish of mercury...atmospheric pressure pushes the mercury up the tube...it doesn't reach the top as it reaches a point where the weight of the column of mercury is equal to the atmospheric pressure.

Del

The trouble is, a pump does not create a vacuum if it is not primed with water. It is very inefficient at moving air (similar to small centrifugal blowers).

It does provide a differatial pressure, but has limitations on the suction side that are based on atmospheric pressure, inlet conditions, suction pipe length and diameter.

a pump gerates differential pressure. In order to know what the pressure at the inlet of the pump, you must do all those calculations, friction, elevation change ect... Then, if the pressure on the fluid being pumped drops b elow its vapour pressure (it s boiling point) the fluid will be 2 phase, and pumps like only 1 phase fluids.

Pompie.. please explain how the piping circuit goes and I will explain gladly

regards

The manufacturers curve will show what the pump is capable of with a flooded suction(inlet) with 0 pressure and 0 lift. Since fluids are solids (they don't compress) the volume at the discharge will equal the volume at the inlet. Anything that restricts that flow to the inlet such as Friction loss or lift, must be deducted prom the discharge.

When the pump starts

• the impeller throws the fluid out and into the volute
• forcing the fluid out of the impeller drops the pressure at the impeller eye
• atmospheric pressure pushes towards the impeller eye
• friction loss in your inlet pipe will reduce the flow towards the impeller eye, so this must be deducted from the performance curve
• the distance the fluid is lifted (suction head) is the difference between the surface of the fluid and the center of the eye. this distance has to be deducted from the curve also

Hallo guys,
I am impressed by all your knowledge - but in this thread it seems to me, you are using your knowledge to drive Pompie into confusion, even if your explanations are correct. Please go back to the question as it was asked!

Pumps can suck liquid or gas media for sure - and Hendrik and Del so far are right, as the pump needs to create a negative pressure in the suction line, to get the medium lifted into the pump. And here is the interesting point of your question:

Depending on the pump design (used pumping principle) their capebility for generating a negative pressure is quite different. For instance: an pump design as typically used in heating water circuits or in your car (for the cooling water) have a very poor ability for suction - they better should be installed in a way they are always kept flooded (below water level of an open basin or in low section of a cloosed circulation loop. A volumetric discharging pump like piston-pumps or gear type pumps (those pumps typically aren't centrifugal pumps) can have a very good suction property, and you can install them above the fluid level.
It was mentioned already by other contributors: what a certain pump is able for, should be advised by the pump maker (media limits, suction height, etc.).

What does it mean:
if we focus again at your information, we have to talk about centrifugal pumps. Most of them have poor suction properties once they were running empty. Therefore these pumps better held flooded. Once they are flooded they are able for sucking the media within small limits, and these limits need to be considered in calculations (if suction height is used).
Another reason for using centrifugal pumps in a flooded start-up situation is, that they typically are used for high flow rates. In piping sections of negative pressure, but also of just small positive pressure, risk for cavitation effects is high.

Sticking with centrifugal pumps, I beg to differ somewhat with about pumps sucking gas or air as a media.

Centifugal pumps do poor job of sucking air. Their air suction capability is probably on par with an inexpensive bath exhaust fan, a lot of noise and some air movement provided there is no restriction on the suction or discharge. I doubt a standard centrifugal pump could draw more than 1 inch of water in suction. (Positive displacement pumps are a different beast, but not a topic of this discussion)

The imprtant issue is to draw a pressure datum through the pump suction inlet, and understand how relative pressure is affected at different points around the datum line. Once that light bulb went on (AHA!) the understanding became simple.

If the suction liquid was above the line, life was usually good and the pump usually had an easy startup and operation.

If the suction liquid was below the pump suction inlet, then one had to carefully consider how water got to the pump inlet. and how the elevation, friction factor, temperature, and velocity interact and compare to the pump's stated NPSH.

Keep reviewing the suction conditions with respect to the pressure datum, for me it was key to understanding how well a pump may or may not work.

A pump does not "suck" anymore than you "suck" liquids through a straw. Atmospheric pressure (at sea level) (14.7lbs/square inch) pushes the liquid through the straw in absence of an opposing positive pressure in your mouth. The "sucking" is a result of a decreased pressure on the discharge side of the pump via an input of energy.

Pompie,

You pose a good question, for which there is no quick and easy answer. However, let me enlighten you about the Total Dynamic Head of a centrifugal pump and how the suction pipe influences the pump discharge.

First of all, to determine what the particular centrifugal pump you are considering will do in a specific installation, it is necessary to produce a "system head curve". This is done untilizing the Hazen & Williams formula for friction loss in a certain size pipe. Typically, you can get tables from most pump manufacturers listing the friction loss per 100 feet of pipe based on the size and flow through the pipe. For instance, if you are using a 4" SDR21 PVC pipe, the table would tell you that the friction loss in that pipe per 100 feet of pipe is .589' of head (pressure) per 100 feet of pipe. If the pipe is 600 feet long, the friction loss through that pipe would be 3.534'. That is the "friction" loss.

To that figure, you must add the "static head", the elevation difference from the water surface to the discharge point. If that number is 25', then you have a Total Dynamic Head of 28.534' in the "system". The suction comes into play here in two ways: First, the static head is from the water surface. If the pump sits above the water and must "suck" it up, for want of a better term, then that is included in the static head. Further, the suction line may not be below the pump level at all, but, rather it may be 100 feet away from the water supply. The pump must suck the water that distance. Since the water is moving through that suction line, it is exerting friction losses and must be considered in the same way the discharge line is considered.

To develope the system head curve, you must use the tables I mentioned earlier. Pick a series of flows from the tables and list them on a sheet of paper top to bottom. Then, from the tables, select a number of flows from the tables starting lower and continuing higher. Let's take the example above. If you anticipate you need a flow of about 100 GPM, you might select from the table the numbers starting at 20 GPM and continuing in 20 gallon intervals, ie, 20, 40 60 80 100, 120, 140 160 180 and 200. Form the tabkle, select the corresponding friction loss for your size pipe. List that to the right of the flow. Then, in the next cloumn, add the friction head (this will be the same throughout the demonstration). This will give you a value for each flow you have listed.

Now, select a centrifugal pump from the manufacturer's catalog. Lay the values from you chart on the manufacturer's pump curve. This will give you a locus of points that will start at 0 GPM for flow and at the static head for pressure. This curve will rise higher as it proceeds to the right on the manufacturer's curve. Where ever your system head curve intersects the manufacturer's characteristic curve, that is the delivery point you can expect from that pump in that system of piping.

This is the cheap and dirty version of an extremely important theory of operation. But I hope it gives you some insight into the quesiton you have asked. As for the pump being able to suck, that has been adequately address earlier by Del et all.

Pompie, you asked:

They tell me a pump does not suck the water but it is delivered into the pump by atmospheric pressure.

If this is true[,] why...?

Here is the fundamental answer you are looking for that will ease your dilemma—

You have been innocently "sucked" into a quandary pertaining not to technical concepts but to (commonly confused) word meaning; and what they told you is incorrect—because the pump (or at least it's intake) does suck.

If they had realized, they would have explained that (liquid) sucking (too) is the displacement of the liquid under differential pressure as between the pump's (the sucking machine's) intake and its output—between less pressure in an evacuated/evacuating space within the pump apparatus, and the atmospheric pressure impinging upon the water being pumped/sucked. So sucking and pumping are, in essence, one and the same thing...the misunderstanding arises because we ordinarily talk about sucking in the biological (the commonplace as opposed to technical) sense, as if it has no relationship to pumping, in the mechanical sense.

But, if we look at various forms of "sucking" by people and other creatures (feeding babies, drinking with straw, breathing with lungs, the feeding of some fish, gill breathing by some fish...to name a few...even down to the cellular level), we see that the same mechanical principle (as with pumping) applies.

When, for example, people sip with a straw or feed at a nipple, muscle contractions (analogously to your impeller) of oral, lingual, and facial muscles provides the motive force to expand volume (to partially evacuate the volume), hence reduce pressure, in the oral cavity, throat and tracheal apparatus; this by which atmospheric pressure is "able" to displace the liquid inward (to neutralize the reduced intra-oral pressure) where it can be swallowed. The same applies, say, with the inhalation phase of lung breathing...with contraction of diaphragm muscles below the lungs serving as the motive, chest-cavity-expanding/lung-evacuating force; this would be more in line with a piston pump, or bellows, but the pump displacement principle is the same.

So with your rotary pump, when the (liquid- or gaseous-fluid-primed) impeller forces liquid-state water out through the pump outlet and reduces pressure within the impeller chambers, it is differentially higher atmospheric pressure (the weight of the atmosphere) on water at the intake that force more "outside" water through the intake to replace it, allowing more water to be impelled (displaced) out the outlet, and more water still to be (yes, that's right) sucked into the inlet. If it were not for the atmospheric pressure upon water at the inlet, the pump would only work momentarily because no "new" water would replace the water that was impelled out of the pump. (If that was the true reality, there would be no such things as pumps...or, say, straw sipping or air-breathing creatures, for that matter.

Now go explain this to your teachers; they will think your are a very bright student; or they put a dunce cap on you.

Pompie,

The thing to always remember about fluid flow is that it goes from an area of high pressure to an area of low pressure. Seems simple, but people seem to forget it a lot. The higher the high pressure the better the flow. The exception to this tenet is the pump. A pump is any device that adds pressure to a fluid flow.

You have asked why inlet conditions of a centrifugal pump affect outlet conditions and therefor pump performance. This goes back to the first thing to remember about fluid flow. The pump will raise pressure of the working fluid, but it starts that process at the pressure going in. The more pressure available going in, the more there will be going out; thus, improved working head and pump performance. More pressure more flow. It is really that simple.

You might want to learn more about Net Positive Suction Head (NPSH) (or inlet conditions) and what determines them at

http://www.engineeringtoolbox.com/npsh-net-positive-suction-head-d_634.html

You will find there a lot of other useful information about pumps as well.

Best Regards.

Ahem! It seems he was asking why his teachers said a pump does not really provide suction (when it really does). Post 12 seems to have covered it and explained why his teacher were "innocently" mistaken. It's a language conceptualization problem, not a technical principle problem that Pompie ran up against.

Sometimes ya gotta read carefully between the lines to see what's really bothering a questioner. This was one of those times.

Thanks everyone my suspicion was confirmed.

I am however going to work in an agriculture / irrigation environment where all the customers, salesmen and technical persons call it "suck" and convincing them otherwise won't improve anything. The customer / farmer is only interested in delivering the water where it is needed.

Someone mentioned NPSH and made me curious to learn more. I know how to calculate it now

In old books and documents the pumps were sucking and the price list contain suction hose.

When did pumps loose that ability? (no need for answers - it is only words)