Cavitation in Centrifugal Pump

I just hope this pump is nothing to do with the aviation fuel pump in your previous posts!

The pump should be sized correctly to avoid cavitation. You need to know data on the specific application and pump to know if it would work without cavitation. Not knowing this data I cannot give any answer, but I can tell you that all the information you need can be found if you know what to look for.

Cavitation may occur depending on the elevation of the well, S/G and temperature of the pumped fluid, and rate of flow, even if NPSH is within pump specs.

So, you can make even a submerged pump cavitate.

first.. centrifugal pump don't suck.. they lift, although EVERYONE say "Suction" instead of "INTAKE"

second do you really know what cavitation is? if not this is the simple explanation.

Cavitation means that cavities or bubbles are forming in the liquid that we're pumping. These cavities form at the low pressure or INTAKE (suction) side of the pump, causing several things to happen all at once:

The cavities or bubbles will collapse when they pass into the higher regions of pressure, causing noise, vibration, and damage to many of the components. We experience a loss in capacity.

The pump can no longer build the same head (pressure)

The pump's efficiency drops.

The cavities form for five basic reasons and it's common practice to lump all of them into the general classification of cavitation. This is an error because we'll learn that to correct each of these conditions, we must understand why they occur and how to fix them. Here they are in no particular order :

Vaporization

Air ingestion (Not really cavitation, but has similar symptoms)

Internal recirculation

Flow turbulence

The Vane Passing Syndrome

Are you are confusing "Cavitation" with "air ingestion"??

"first.. centrifugal pump don't suck.. they lift, although EVERYONE say "Suction" instead of "INTAKE""

Although I've seen you say this before, I can't agree with you, brich. A centrifugal pump is no different to any other impeller-driven fluid-pumping device (eg, fan, blower, compressor, pump, etc.), it sucks from the inlet and blows from the outlet (some vacuum cleaners and leaf-blowers can be connected to either suck or blow). If the pump does not suck, then how on earth does the fluid get into the pump and reach the impeller? If you are only accustomed to submersible pumps, and positive suction pressures in terms of barg, then I can understand a certain lack of understanding. But anyone who deals with a wide range of centrifs will know that (subject to being correctly primed) they can operate perfectly well when located above the liquid source. Take, for example, contractor type dewatering pumps, which are usually standard centrifs, with a rotary vane pump to evacuate the air from the suction line for priming, but which then operate as a normal single-stage centrif, but located well above the liquid source, which rather imples that they SUCK (subject, of course, to the laws of physics).

You mention yourself that "These cavities form at the low pressure or INTAKE (suction) side of the pump". It is this very 'low-pressure zone' that draws the fluid into the pump. (And, in fact, it is the suction systems' capacity to feed fluid quickly enough to this low-pressure zone that determines whether or not cavitation will occur)

It is also confusing to say that they 'lift'. In fact, they pump just like any other pump. The 'suction' is subject to the very same laws of physics and NPSH restraints. The only difference is that the discharge is expressed in terms of a column of liquid (head, in feet or meters) rather than pressure in psi or bar. This is simply because a centrif gives a certain head, irrespective of fluid SG, whereas a PD pump gives a certain pressure, irrespective of SG. Your terminology does nothing to explain the differences between centrifs and PD pumps.

Feel free, like last time, to OT or ignore my post if you wish. I won't be offended.

To return to the OP's question, having the pump immersed in the liquid makes no difference to the NPSHa vs NPSHr equation, apart from eliminating suction pipe losses from the NPSHa part of the equation. However, in practical terms, and in normal circumstances, cavitation is unlikely to be an issue unless:

a) The suction vessel is too small for the flowrate (suction starvation)

b) The fluid is at a high temperature, and therefore close to Vapour Pressure .

c) The fluid is volatile and therefore close to Vapour Pressure at ambient temp.

GA to Spinco, who clearly knows his stuff, but his final sentence is most telling, NPSHa vs. NPSHr is everything as far as cavitation is concerned. Suction Specific Speed is really just a unitless indication of where a pump/impeller design lies between radial flow and axial flow. This obviously has some effect on the NPSHr, but only in the pump design phase. For a given pump size/design calculating Suction Specific Speed will not change the NPSHr vs. NPSHa scenario, and could confuse the issue.

A great professor of mine once said: "There is no such thing as suction, nothing "sucks"". The term suction is misused, as what is commonly perceived as "suction" is actually simply reduced pressure... Anything with a pressure less than atmospheric (14.7psi, depending on elevation) has the ability to do this so called "sucking". A more accepted term one can use would be "Vacuum", or perhaps "negative pressure". However many industries use the word "suction" to describe something that pulls fluid through reduced pressure.

When describing a pump, the term "inlet" should be used in place of this "suction" word, as suction does not exist, except for when used as a slang word describing something non-desirable.

^{(Even I use the no no word suction on occasion when describing the inlet side of the pump that pulls fluid)}

GA from me.. in fact I was going to reply to the last post from Holzfeller, but been rather busy putting some very large pumps in some very deep wells, that lift and not suck.. but as you've said in fewer words than me, what I would have said... I'll not bother now!

RVZ717 has kindly answered for you on the question of 'suction', but what about the question of 'lift'? Your deep well pumps clearly 'lift' fluid to the surface, (as would a diaphragm pump, progressing cavity pump, gear pump, etc, if they were being used.). What 'lifting' is done, for example, by a recirculation pump in a closed-loop circuit, or a transfer pump in a tank farm transferring fluid to the same (or lower) elevation?

Just have to give a direct answer to this question to clear up any misunderstandings:

Q) If the pump does not suck, then how on earth does the fluid get into the pump and reach the impeller?

A) The atmosphere "pushes" the fluid into the pump.

Any time you have one area with less pressure than another, the higher pressure area will push the fluid into the lower pressure area. The low pressure area does NOT pull fluid into itself. The air that makes up our earths atmosphere has mass and weight. This weight pushes in on everything on earth, and can vary slightly depending on elevation. At sea level 14.7psi is always pushing on everything. Any time you reduce some pressure by removing a fluid or gas from an area, this 14.7psi rushes in to replace the lost volume. This inrush of pressure is attempting to bring everything back into pressure balance on both sides. This effect is commonly confused by many, and those people typically refer to this as "suction".

For the rest of the world the word suction is just fine, suck away.

In the Engineering world, it does not exist, not properly anyway.

How about NPSH? Right or wrong, the term suction is used formally in many engineering texts. I had a physics professor that preached there is no suction. OK, fine, but it doesn't change the fact that it's the term the engineering world uses.

It's like preaching that Newton's equations aren't perfect because they don't account for relativistic effects. OK, great, fine, but for all practical purposes it works, people understand it, and trying to gain some additional level of perfection would only complicate the communications (for the everyday engineer).

NPSH is actually describing positive pressure, not suction, although I will give it to you that the word is in the acronym.

Ahhh... This all sucks anyway

I think you will find that suction does exist in the engineering world. It is a term used and understood by thousands or millions of engineers in many fields, including pump engineering. If you google the terms 'suction' and 'engineering' together you will get millions of results, many involving esteemed engineering institutions, academic institutions and engineering companies, all using and understanding the term 'suction'.

What has happened is that someone in the past has misinterpreted, misunderstood or mis-defined the term, assuming that suction means 'pulling'. In fact, none of the definitions I have seen make this inference, they all refer to the movement of a fluid (or solid) into a low-pressure zone (see previous definitions). The following is an extract from Wikipedia:

"Suction is the flow of a fluid into a partial vacuum, or region of low pressure. The pressure gradient between this region and the ambient pressure will propel matter toward the low pressure area. Suction is popularly thought of as an attractive effect, which is incorrect since vacuums do not innately attract matter."

So you can see that suction is only nonexistent if the definition (or understanding) of the term is wrong.

just because it's in Wikipedia does not mean it is correct!!

What if it's also in Introduction to Fluid Mechanics, (Fox, McDonald, Pritchard), 6th edition?

"Introduction" implies that a basic understanding has not yet been achieved, in which case you'll need to use as many layman's terms to describe something to give the readers a good understanding of what you are saying. I would wager that an advanced physics based fluids book will not have this term, but maybe I'm wrong.

Good explanation!

I quite agree, but I thought it summed things up quite nicely. Do you disagree with it?

hang on while I get up offa the floor.. we agree on something!! Progress

but in my terms, used in artificial lift circles, Reda/Suhlumberger, Baker hughes/Centrilift, Borets, Weatherford and a few other makers of centrifugal pumps for the oil industry, we never refer to the hole at the bottom of the pump as the "suction", and I think after reading all the comments, this is where you and me have been at logger heads, as to the correct terminology for the "place" were the fluid enters the pump.

In AL, we say intake, as the intake may be a bolt-on part that allows the fluid into the pump, remembering the pump "intake" is sometimes 2000ft plus below the dynamic fluid level and it can be depending on the GOR of the well fluid, replaced with an Advanced Gas Handler or a Gas Separator, all of which have an intake part to them.

So while we are talking about the same thing, we are using different terms to describe the same thing.

Now the pumps I use do lift as they are vertically set in a deep well and each stage (of two parts) give a certain amount of lift dependent on several factors of which we know about and have had many a post on.

So now hopefully we have found common ground.. you say suction, I say lift & intake.. and we understand that we BOTH mean the same thing!

That's interesting! You're in the UK (gas industry) and the typical term is intake. I was in the power industry here in the US for about 13 years, we always said suction. I'm now in water/wastewater engineering and we use both. Typically we use "intake" to describe a couple things (such as a raw water pump intake (the area where the suction piping terminates in the lake), or the building (intake house)). But the piping, valves, etc are all labeled suction/discharge.

Also, now that I think of it, my physics professor was from the UK.

So I wonder if suction/inlet/intake is consistent between geographic areas as well as between industries... seems reasonable enough.

NM... shoot that theory, I just looked.. HOLZ is from the UK... RVZ is US... hmmm... maybe just between industries??

Actually I'm neither in the UK or in the gas industry. I work worldwide and in oil, with gas a by-product.

I reside in the UK (sometimes) and when they are not bombing the crap out of each other in Libya, or Syria or Bahrain... off I go to run, pull, troubleshoot and educate/train the local engineers in the aforementioned countries on how to run, pull, troubleshoot their own ESP's.

As for different geographical/industrial use of certain words, phrases, etc.. who knows, I can speak fluent American and Canadian and much of our language in AL is the same wherever we are from, however I do note that there is ONE word common to all.. "Discharge" and in the ESP industry we do measure & record discharge pressure.

Like I said.. we've found common ground... all it took was a step back to look at what we couldn't see!

Now what was the OP's original question??

I've lived mostly in the UK, but I'm canadian by origin, and have worked for UK, US, anglo-american and european companies. When I started in pumps, on chemical process centrifs, it was exclusively 'suction' and 'discharge' everywhere, including all the european languages I can speak or decipher. It is only since I defected to the dark side (PD pumps) that I've seen more frequent use of 'inlet', 'outlet', etc. I don't know if it's down to different industries, different pump types, or the efforts of UK physics professors. I have never paid too much attention to this, but probably will now, just out of interest.

You could be right, and I can understand the reasons for your terminology. In fact, with 2000ft of static head above the "intake", to call it "suction" doesn't quite seem appropriate.

No not static head above, I said "Dynamic fluid level", which is something completely different!

HAHA... I see I've ruffled some feathers.

I know, you know, your neighbor knows, your postman knows, everyone knows what the term "suction" is referring to, which makes a great word to describe something so everyone knows what you are talking about. My only argument is that although well known, and widely used even in industry, the term is very commonly misused. Any physics professor will tell you that suction does not exist, and technically speaking, they would be right. Realistically speaking on the other hand, suction is a widely used term to describe a difference in pressures.

As I said before I use the word all the time when talking to customers, service techs, other engineers etc because it's a universally understood word. However, from a strictly technical standpoint, it's not quite right.

No that we have split hairs...

Do you prefer "negative pressure", "vacuum", or "suction"?

Are there other words you use to describe this differential pressure?

No, don't worry, feathers are unruffled, and we are still cool under the collar.

My point is that the term is only misused if you have the wrong definition or understanding of what suction is. If you say that suction is a force that pulls something in, then suction obviously cannot exist as even a perfect vacuum has no attractive force on its' own. But this definition, or understanding, is incorrect. Most definitions state quite correctly that suction is the movement of a fluid or solid from a zone at a certain pressure to a zone at a lower pressure. If you accept this definition you will see that it describes exactly what occurs on the suction side of a centrifugal pump. The fact that many people do not understand this definition does not make it false.

What irritates me is the non-sequitor, or inconsistency, in the logic of the argument that suction does not exist. Its' basic and first premise is an incorrect definition of what suction is. It goes:

• "Suction is an attractive force that pulls stuff in"
• "The laws of physics say that stuff can't be pulled in"
• "Therefore suction does not exist"
• "QED" (clever me!)

When you correct the first line to the true definition of suction, the argument disappears and we can all heave a big sigh of relief, as we don't have to change all our pump literature afterall.

Rather than claiming that suction does not exist, our esteemed physics professors should rather be ensuring that everyone understands the true definition of suction.

Sounds good, I'll take that.

For those bantering back and forth about whether pumps suck or not... here's a video of a deckhand's helmet ripped off at a turbine intake...

http://www.youtube.com/watch?v=lRgS9N5Ixq4

If I were him, I definitely say it sucked...

He might say the event sucked... I would say the atmosphere just got a little pushy.

To put it simply, yes NPSH will be affected , it will decrease with flow increase. In your case the pump is immersed, see if the immersion is sufficient. It happens in cooling towers, cavitation due to insufficient immersion.

T.Q.Choudhury

NPSHa will indeed decrease with increased flow. But NPSHr (which is what the OP mentions) will in fact increase.