Cavitation in gas carrying media 1

[MartinLe]

I'm not at all sure that I really understand the hows and whys of cavitation in gas carrying media, here's some thoughts, if I'm way off please point out.

In water, the mecahnism as far as I understand is that the pressure in the pump is locally lower than the vapor pressure, small vapor bubbles form, these collapse, this can damage the pump and the whole process looses kinetic energy.

I'm interested in biogas slurries, the relevant gases are CH4 and CO2.
CH4 is practically insoluble, so a pressure change would mean that the tiny bubbles entrained will grow and shrink, correct?

CO2 will be mostly dissolved, lower pressure will mean that small bubbles form. Bubble formation will be limited in rate because the CO2 is dilute. With a pressure rise I'd expect the CO2 to dissolve again.

Either case, while bubble forming from the entrained/dissolved gasses will in all likelyhood occur at lower temp/high pressure than cavitation in pure water. In my mental image of gas bubbles collapsing there's somehow no shockwaves (like in vapor bubbles): CH4 bubbles don't vanish, they just shrink, CO2 dissolvesbut the kinetics is different and I somehow would expect these bubbles to collapse slower. I'd expect no damage to the impeller purely from this. But this is very much intuition and could be very wrong.

The observations I hear from our field guys is that when a pump cavitates, you get noise and performance losses, no one mentioned damage to the impeller (But it's entirely possible ). So I don't think my mental model is totally wrong. Or is it?

 

[bimr]

Entrained air results in loss of capacity, head, efficiency, pump damage, and possibly pump prime.

 

[georgeverghese]

Check if the pump nozzles are arranged to enable self venting of the pump casing - casing top exit so that any entrained gas bubbles in the feed wont get trapped in the casing.

 

[Artisi]

1."The observations I hear from our field guys is that when a pump cavitates, you get noise and performance losses, no one mentioned damage to the impeller (But it's e
Do you field guys really know / understand what cavitation is?

2. Entrained gas in the feed will expand in the eye of the impeller (low pressure area) but go back into solution if it can pass the impeller eye, unlikely to cause any noise but may show up as a reduction in flow, excessive gas will expand and could put the pump off prime - you may get some noise as the pump is off performance and flow (if any) isn't entering the impeller eye correctly.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[LittleInch]

I don't really know what you mean by "gas carrying media". From your example, if this slurry is at atmospheric pressure, then any "gas" contained in the liquid which exists as bubble will rise to the surface and enter the gas phase above the liquid. In this instance the effective "vapour pressure" of the "liquid" is the same as your local atmospheric pressure.

If you lower the pressure, but in a contained environment like a pipe, this gas will become trapped in the liquid, and you are looking at some sort of two phase flow. If you try and pump this you will run into a number of problems as most centrifugal pumps do not like two phase flow and there could be some noise as these gas bubble appear and then become compressed.

Is it the same as cavitation on e.g. water. Depends to me on whether the gas is already present in the fluid before it gets to the inlet of the pump. If it is then you probably won't get too many bad effects on the impellor, but the pump will not work effectively and if there is too much gas it might not work at all, especially if you don't have flooded suction.

It would be instructive for you to draw a sketch / profile of what you're thinking about so we can see where you're coming from because at the moment I think you have it really quite wrong.

In these instances ( effectively "boiling" liquids) all you have for NPSH is the static head minus the friction losses as the pressure on top of the liquid cancels out the vapour pressure" exactly, whatever that pressure is. Hence your pump is often in a deep pit or some considerable height under the liquid level.

Highly volatile petrochemicals held at high pressures are similar.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[MartinLe]

Do you field guys really know / understand what cavitation is?

Should have writeen "I heard from the field guys" since the company decided we somehow don't need service technicians and they were all fired some time ago. I recall our service manager (long gone too) quoting his guys talking about "gravitation" in our pumps so ... probybly not, but I can't ask.

The complaints from customers and observation from service guys was low flow and vibration/noise. low flow means compared to data sheet, in some cases the customers changed the operation of their plants and pumped far thicker slurry, leadingto the problem. I assume the causal chain is: higher viscosity in fluid -> mor pressure loss in piping to pump -> more problem with cavitation/entrained gas -> vibration/performance loss. Does this make sense?

Artisi, I understand you correctly that entrained bubbles can completely blockthe eye of the impeller? (with the result total loss of flow and very likely damage to the bearings) This is a problem with heating pumps and other small pumps?
I don't see this happening with our pumps since they can pass solids a sized a few cm, I somehow don't see gas bubbles this large entrained in the flow. Correct me if I'm wrong.

georgeverghese, we installed our pumps in a way that allowed self venting.

 

[Artisi]

The lowest pressure in an impeller is in the eye, depending on operating conditions this can be quite low and gas / air entering thus area will expand into large volumes, enough to fill the eye completely and stop flow. It has nothing to do with solids handling capabilities.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[LittleInch]

IMO, 2nd para - correct

3rd para, This affects pumps which are above the liquid level and where you're using atmospheric pressure to push the liquid vertically up into the inlet of the pump which usually needs some sort of priming system. Too much gas means that the pumps "loose prime" and are essentially trying to ump gas instead of liquid and they don't do this very well. Flow normally stops or gets very low in these instances.

When you have flooded suction (i.e. inlet of pump lower than the liquid level, this still occurs, but is self fixing, i.e. the liquid will fill the pump again when flow slows or stops until it happens again, sometimes at a high frequency leading to intense vibration.

What you describe looks like classic low NPSH / cavitation / two phase flow issues for centrifugal pumps. cures are use another type of pump, lower flow, bigger inlet pipe, pump lower than current operation.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[zdas04]

Guys, entrained gas has NOTHING to do with cavitation. Entrained gases can effect pump performance, but they CANNOT lead to cavitation. "Cavitation" can be defined as "the formation and subsequent collapse of condensible vapors in the flow". Methane, CO2, O2, etc do not experience phase change at the conditions we see in most pumps, and therefore do not contribute to cavitation. Cavitation requires a phase change, so it is only an issue when pumping liquids near their vaporization temperature somewhere in the process. For a centrifugal pump, the pressure at the eye of the impeller is the lowest pressure in the system, so if the temperature is near the boiling point of the liquid being pumped (at the pressure in the eye of the impeller) you can get vapors that will then condense in the impeller and/or in the volute.

The pump performance issues you are talking about don't sound like cavitation to me, but there are a lot of other things that can go wrong with a fluid that is mostly liquid with significant entrained gas.

David Simpson, PE
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[LittleInch]

zdas04,

I agree that if the gas is present in the liquid before it gets to the pump eye then it isn't cavitation - one way to stop cavitation is actually to introduce a small gas supply into the liquid u/s of the pump. Doesn't have much good for the performance of the pump though...

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Artisi]

Depends on how much is introduced, and a small reduction in capacity is certainly better than cavitation damage.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[Artisi]

Back to the original question of entrained gas in the pumped product. As pointed out by zdas04, it doesn't result in cavitation - hope this answers your question or do you have an actual product handling problem resulting from the gas entrainment?

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[MartinLe]

LitleInch, I didn't see your first comment when writing mine. I don't really know what you mean by "gas carrying media". From your example, if this slurry is at atmospheric pressure, then any "gas" contained in the liquid which exists as bubble will rise to the surface and enter the gas phase above the liquid.

We are dealing with thick, viscous slurries. small entrained gas bubbles don't rise easily.

The strategies we used as prevention where short, fat inlet pipes, positionng the pump low and playing with flowrate (via VFD). The latter was a fix after the problems I mentioned in my last comment showed up.

I'll let your input sink a bit.

 

[MartinLe]

(I wrote my last comment before reading zdsa)

zdsa04, would entrained gasses lead to impeller damage?
As I understand it (and the discussion so far) the symptoms of two phase flow and 'proper' cavitation would both include performance loss and vibration, both more pronounced the lower the NPSH. Two phase flow problem may start at higher NPSH than cavitation. Remedies are the same for cavitation and two phase flow. So far we are all on the same page, I think. I think the mix-up is one of terminology.

Artisi, we think that part of the problems some of our customers had had to do with entrained gas, I think I described above why I/we think so, and what we did so far which mostly seeems to work. I'm not trying to solve a concrete problem right now. I want to test my understanding of the problem because this is somehow a recurring discussion here.

 

[zdas04]

The most frequent locations of cavitation damage in a centrifugal pump are: (1) late in the volute; and (2) outer edges of the impeller. Liquids that are very near NPSH-r can have cavitation shortly after the eye, but that is reasonably rare. I've never seen caviation in the suction piping or in the eye of the impeller--the effect required an increasing pressure to collapse the bubbles.

David Simpson, PE
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[Artisi]

MartinLe: Seems you are discussing more than one installation "some of our customers" same problem in all cases? This is unusual unless all installations are identical - that means identical in every aspect. Is this the case are the pumps mismatched to the application?


It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[MartinLe]

The customers are farmers operating biogas plants, all using different feedstocks but usually some mix of manure (liquid and solid) and energy crops or grass. all similiar at first glance, but actually the differences may be severe. Many operate their plants outside of the envelope they were designed for, by using more solid feedstocks Some of those encounter the problems I described. So yes, the pump is a bad match for the application but mostly because the application was changed at some point.

 

[Artisi]

From what you say, the problem may not be entirely "gas entrainment" in all cases. What pump are you using? a bit more technical detail might throw a bit of light on the subject. I have over 10 years in the paper industry pumping paper stock - not unlike what you are describing, fibre sludge, air entrained etc.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[MartinLe]

This:

http://www.uts-biogas.com/fileadmin...er_ENG/091013_UTS_Pumptechnik_4000_UK_low.pdf

... in the configuration labeled ZPS. look at the top right picture to get an idea. distance to tank is 2-4m, seldom longer, feeding pipes are DN300.

 

[bimr]

Don't think you have a problem with air; it is just a problem with pumping sludges. At best, this is a tough application.

It looks like you are using an auger type of pump which is a good application. Entrained air will not be a big problem with this type of pump.

Due to the unusual and complex fluid mechanics associated with the wide varieties of sludges, each sludge pumping application is a unique design problem, and one must develop site-specific design criteria based on detailed evaluation of the specific sludge characteristics.

The key to pumping sludge is to use a pump properly sized to develop sufficient head, and smooth pipe sized to produce the proper velocity (neither too high nor too low) without constriction or projections and with as few bends as possible. The key to maintaining such a system is to have large easily opened cleanouts on the pump, at any elbow on the suction side of the pump, and (where possible at all elbows on the discharge side of the pump. Quick-disconnect air and/or water hose connection on both the suction and discharge sides of the pump are desirable where possible. Sludge lines should rarely be smaller than 150 mm (6 in.) and should preferably be larger.

It is impossible to develop any conclusions from word of mouth. First hand observation is necessary.