PDA

View Full Version : Liquid pressure amplification (LPA)



US Iceman
07-04-2007, 01:04 AM
I realize this is not a new technology, but I would like to get the opinions and comments of others who may have worked on these systems or seen them.

What kind of problems have you seen or heard of?

Is the UK firm the only one selling these now? I did find a website from the UK for these but it has not been updated in almost 4-5 years! That made me wonder if these were still being sold or if the website is waiting to expire by neglect.

My understanding is the original firm was sold to a branch of an electric utility, which was then discontinued until a license was sold to a firm in the UK (the one mentioned in the website as Hy|Save UK.)

No detail is too small to mention.

Thank you for your comments and thoughts.

Dan
07-04-2007, 03:22 AM
Hy-save or liquid pressure amplification is something that makes good sense at the simplest level. Inverters, ECM fan packs, XDX valves, surge piping systems and many other good ideas do not always play out in the practical world.

Liquid pumping requires minimum requirements, for example such as net suction head... or whatever.. in order to avoid cavitation. I disabled such a system when I was told that under my contract and also in order to make this new innovation work, that I had to add $3,000.00 worth of refrigerant into the receiver beyond what was required for it to work beforehand.

It's the marketeers who kill these potentially good ideas. They sell, sell, and promise things to decision makers who are not in touch with the maintenance workers.

A liquid pump is a wonderful concept for guaranteeing good liquid quality at the TEV. But....

You have one more thing to go wrong. You have introduced pumping technology into applications that have no history nor technicians who understand all the N-abbreviations necessary to keep a pump intact.

You might have applied this to a parallel system that uses liquid depression for gas defrost differential... how do you make THAT work?

Like so many improvements that have valid potential, liquid pumps become misapplied, incorrectly installed, or unmaintained, and are taken out as nuisances because the energy promises pale in comparison to the performance and maintenance issues. And when I am using the word "issues" I am really saying "dollars."

US Iceman
07-04-2007, 05:30 AM
It's the marketeers who kill these potentially good ideas. They sell, sell, and promise things to decision makers who are not in touch with the maintenance workers.


Absolutely right. Most of these good ideas are built on a "house of cards". A none too stable foundation based based on some fact but held together by hope.

You mentioned the additional refrigerant charge as one issue. While this adds to the problem of over-selling the features of the "device" it also works to create a skepticism that it will never work properly.

Unfortunately, most of the these good ideas go bad because someone forgets the basic fundamentals of applying various components.



...depression for gas defrost differential...


Now that's a new phrase for me. Can you expound on this more please?

Dan
07-04-2007, 03:27 PM
Now that's a new phrase for me. Can you expound on this more please?
Aha! Finally the tables turn. I get to expound to you:)

Gas defrost is sometimes referred to as reverse cycle defrost. In other words, you send a high pressure gas backwards to normal flow from the compressor unit through the suction line and still against normal flow through the liquid line all the way back to the compressor unit.

This is the "reverse flow" concept. And all flow is a result of providing a higher pressure toward a lower pressure. So we have to provide a higher pressure or lower pressure at opposite ends in order to reverse what we consider the normal flow of high pressure liquid to low pressure suction.

There are two ways of accomplishing this in a typical commercial refrigeration application.

1. Raise the discharge pressure above the liquid pressure. Typically this requires a differential valve in the discharge line that only opens after the discharge pressure going to the defrost header is 30 (Just grabbing a number) psig above the liquid (or receiver) pressure. Thus you get the reverse flow.

2. Liquid depression. You use a differential valve in the the liquid line before the liquid header that will not permit flow from the receiver unless the liquid header pressure drops is 30 psig below the receiver (or condensing) pressure. Thus you get the reverse flow.

I wish we had parallel terms. One of my bosses in the old days used the term "liquid depression" and it stuck with me. Would "discharge elevation" be a parallel descriptive for the first method?

US Iceman
07-04-2007, 03:38 PM
Thanks for the explanation Dan. What you describe now makes sense.

I'm glad to see the commercial refrigeration guys invent their own terms and descriptions too. For awhile I thought only us industrial guys did that!:D

Gary
07-04-2007, 03:46 PM
The achilles heel of the liquid pump is cavitation. Cavitation happens when the liquid pressure is dropped below saturation (at the eye of the impellor) forming a vapor bubble, then as the bubble moves outward in the impellor it implodes violently. This can very quickly destroy a centrifugal pump.

IMO, cavitation could be minimized if not resolved with an inlet inducer, but to my knowledge this hasn't been tried. The usual "fix" is to have a very tall head of liquid entering the pump to increase the pressure at the eye of the impellor.

Gas defrost differential: In a rack system, one or more evaporators go into defrost. Gas (hot or cool) is injected into the suction line(s). This condenses into liquid in the evaporator. The liquid flows backwards through the liquid line into the liquid header. In order to have this reverse flow, the gas pressure must be raised or the liquid pressure lowered. IOW, there must be a pressure differential for reverse flow gas defrost.

Dan
07-04-2007, 03:58 PM
The liquid flows backwards through the Liquid Line into the liquid header. In order to have this reverse flow, the gas pressure must be raised or the liquid pressure lowered. IOW, there must be a pressure differential for reverse flow gas defrost.
Reply With Quote

As always, Gary, your economy of words explains it best. Other than the IOW... which I assume means "in other words."?

US Iceman
07-04-2007, 04:06 PM
The usual "fix" is to have a very tall head of liquid entering the pump to increase the pressure at the eye of the impeller.


The fix as you have labeled it is a mandatory requirement. Any pump that does not have sufficient suction head on it will cavitate. Some of the early picture I saw of the Hy|Save systems had the pump right on the floor next to the receiver.

There is NO way those would have worked for any suitable length of time. The other problems revolve around constant suction pressure to the pumps (which in this case is the system discharge pressure).

If the suction pressure to the pump would change, the liquid could flash off and in turn create cavitation.

Thanks for the comments so far guys. I'm trying to find out why this method seemed to have died out in the US, and is only being marketed through the UK (if it still is?).

Gary
07-04-2007, 04:12 PM
I thought you explained it very well, Dan. The reason I also explained it is that we were typing at the same time. You, being younger and faster, hit the send button first.

IOW = In Other Words

Gary
07-04-2007, 04:26 PM
I'm wondering if a liquid pump wouldn't work very well at the outlet of a subcooler.

It seems to me that the subcooling would lower the bubble point and thus prevent the cavitation.

Gary
07-04-2007, 04:44 PM
For those who may be wondering, an inducer is a screw type device mounted at the inlet of the impellor; A pre-pump, if you will, which gradually raises the pressure of the liquid before it goes into the eye of the impellor.

US Iceman
07-04-2007, 05:06 PM
I'm wondering if a liquid pump wouldn't work very well at the outlet of a subcooler.

It seems to me that the subcooling would lower the bubble point and thus prevent the cavitation.


The subcooling would be one way of raising the threshold for cavitation. This is essentially what you create with the column of liquid stacked up on the pump inlet. The static head is a form of subcooling.

Another way is as you suggest Gary about pre-cooling the liquid below it's saturation temperature.

Of course another way to reduce the potential for cavitation is to use larger pipes (less friction loss), ball valves, long radius elbows, and... drum roll please, insulate the liquid lines to prevent heat gain.;)

The way I look at inducers is they reduce the losses encountered at the impeller eye. If the liquid is pre-rotated into the impeller eye in the right direction, the pressure losses are decreased. In that manner, the suction pressure to the pump is increased from what it would have been without the inducer.

The NPSH required is based on a specific flow condition to achieve stable flow through the impeller eye at a specific volume flow rate and pump speed.

Small high speed pumps traditionally have much higher NPSH requirements than larger, slower speed pumps (for the same volume flow rate).

Dan
07-04-2007, 06:26 PM
Iceman, could you enter the NPSH and other pump abbreviations into the glossary? Is that Net Pressure Suction Head?

Pumping technology will become a very large factor in commercial applications now that refrigerant emissions are topmost on the political agenda.

Gary makes a good point regarding subcooling to eliminate cavitation in liquid pump systems. But I see an irony there. We are subcooling in order to do further subcooling? An energy analysis would be in order, I suppose. Energy usage and maintenance costs are the issues that need to be on the table together.

Gary
07-04-2007, 08:12 PM
Pumping technology will become a very large factor in commercial applications now that refrigerant emissions are topmost on the political agenda.

Gary makes a good point regarding subcooling to eliminate cavitation in liquid pump systems. But I see an irony there. We are subcooling in order to do further subcooling? An energy analysis would be in order, I suppose. Energy usage and maintenance costs are the issues that need to be on the table together.

Are you volunteering to do the research?

US Iceman
07-04-2007, 10:48 PM
But I see an irony there. We are subcooling in order to do further subcooling?


Not irony, mandatory requirements for the pump to operate properly. We have plowed over this ground before.

The static head is needed to increase the total pressure at the pump suction. A little safety factor is in order to cover other possible operating conditions also.

I guess you can say we are subcooling to get more subcooling, but I disagree. We need a specific amount of subcooling (total pressure) to prevent the liquid refrigerant from flashing off due to heat gain or friction losses.

Without that, the pump will not literally pump as it gas binds with vapor.

Constant issues of cavitation are developed by constantly operating the pump at different conditions such as; insufficient NPSH (available), pump discharge pressure too low causing the flow rate and NHPSH (required) to increase increase, or possibly the liquid pressure being reduced suddenly.

All this system is is a poor man's liquid overfeed system (with minor differences).

From the sound of the discussion so far, the pumps of the LPA system were quite a problem, heh?

lana
08-04-2007, 08:18 AM
Hi everybody,
I was reading your intuitive discussion and enjoying it.
I thought to share some reading materials with you. You may have read them but I give it to you anyway.


http://www.eca.gov.uk/NR/rdonlyres/0A58C133A39046F09FE0F8BDFA791D16/0/RefrigerationLiquidPressureAmplification.pdf


http://www.indiacore.com/bulletin/papers-wi4/chadvenkat- (http://www.indiacore.com/bulletin/papers-wi4/chadvenkat-liquid-pressure-amplification-technology.pdf)liquid-pressure-amplification-technology.pdf (http://www.indiacore.com/bulletin/papers-wi4/chadvenkat-liquid-pressure-amplification-technology.pdf)


https://navyenergy.nfesc.navy.mil/publications/techdatasheets/tds2005.pdf

http://www.freepatentsonline.com/5291744.html

Best regards,
LANA

US Iceman
10-04-2007, 09:58 PM
Thank you for sharing those links lana.

It seems everything I read on these systems is extremely positive (all of the energy savings concepts and marketing;) ), but I do not find very many of these systems being installed.

Which leads me to think this was a good idea, gone bad!

If the major mechanical problems were due to the pumps, then those can be solved. On the other hand Dan brings up a very good point below.



You have introduced pumping technology into applications that have no history nor technicians who understand all the N-abbreviations necessary to keep a pump intact.


Could the lack of use of these systems be because the technicians do not understand them?

Walshyd
12-04-2007, 04:51 PM
Where do I start! LPA has been fitted to all Chillers at heathrow and Gatwick Airport about a year and a half ago as an energy saving device. unfortunatly it has proven not to work on an EXV system as the EXV will fight against the liquied pump. The installation is also very poor as they fit a liquied vessel before the pump inlet to stop the pump from running dry. On two occasions these ruptured whilst carrying out Hp tests on a general maintenance and now turns out that it is covered by the pressure vessel regs. It would be far better to fit 60htz fans and an inverter to any aircooled system as thi will have the same effect and not screw up your Exv but you still need to integrate it to the systems controls correctly. Mcquay wouldn't even warrenty circuits if they changed the compressor.

US Iceman
12-04-2007, 05:13 PM
Where do I start! LPA has been fitted to all Chillers at heathrow and Gatwick Airport about a year and a half ago as an energy saving device. unfortunately it has proven not to work on an EXV system as the EXV will fight against the liquid pump. The installation is also very poor as they fit a liquid vessel before the pump inlet to stop the pump from running dry. On two occasions these ruptured whilst carrying out Hp tests on a general maintenance and now turns out that it is covered by the pressure vessel regs. It would be far better to fit 60htz fans and an inverter to any air-cooled system as thi will have the same effect and not screw up your Exv but you still need to integrate it to the systems controls correctly. Mcquay wouldn't even warranty circuits if they changed the compressor.


WOW. Where to start indeed!:eek:

I'm going to break this down for individual questions.



unfortunately it has proven not to work on an EXV system as the EXV will fight against the liquid pump.


I'm not quite sure what you mean by fight against the liquid pump. Where there capacity control issues with the EXV's or superheat control problems?



On two occasions these ruptured whilst carrying out Hp tests on a general maintenance and now turns out that it is covered by the pressure vessel regs.


That's a different problem unrelated to the concept of liquid pumping. If the pressure vessel failed due to pressure testing that is a manufacturing/design problem. Although I could see how that would sour anyone on continuing to use the system.:o

Walshyd
13-04-2007, 03:16 PM
As a carrier Tec I had the abiltity on the flow 1 to use software to view the basic Temp sensors etc and the software caculates superheats and sub-cooling. An Exv only requires 15 psi differential pressure to open where a txv requires 100 psi+ to open. If you run the machine under normal operating conditions you find that you get a slug of liquied entering the compressor this is becouse the exv has its own parameters that control the device and a liquied pump will increse liquied flow to the orifice causing the valve to pulse in a unstable way as it opens it will send through a slugg of liquied the suction temp will drop and the valve will close right down the senses a rise in tempreture and a low suction pressure and open fast causing a new slugg of liquied and so. Carrier have a form of LPA but all it does is move the liquied through a fixed point on the exv and gives free cooling in a thermosyphon way. As for the failures with the liquied receiver fracturing you are correct that this was just a design error which has been resolved as they now have a steel vessel instead of the 41/8th pipe they used before but as there is only one company installing this equipment it is still a concern not only to the engineers but also to the enviroment as alot of the units fitted are operating on R22 still. Hysave also do not pressure test the whole chiller only the bits they work on but by fitting a pressure vessel and distrubing already tested and certified pipework they should re-pressure test to MAX operating pressures not the 100 psi they did at the airport. And then we could always start on ecomesh!!!!!!!!!

momo
17-04-2007, 03:19 AM
NPSH plumbing term Net Positive Suction Height given in water pump details: arises from the need to estimate the weight (simplified to height) of a column of liquid (water...) which a pump must suck (it takes into account flow resitance in the pipe etc) the liquid's temperature (and therefore vapour pressure) etc. The calculated height is deducted from the NPSH and if the result is negative: problems... cavitation etc and a more suitable pump should be found.
[Bits taken from memory... D_nfoss and other reputable liquid pump manufacturers should help with these calculations and formulas]

Dan
17-04-2007, 03:48 AM
Momo, I have added your definition to the glossary, with some editing. Moderators, especially US Iceman, should feel free to add and modify these definitions. Thanks Momo, for a good definition.:)

HYSAVEUSA
27-06-2008, 01:25 AM
I realize this is not a new technology, but I would like to get the opinions and comments of others who may have worked on these systems or seen them.

What kind of problems have you seen or heard of?
Most problems have occurred due to poor installation inexperienced installers, lack of knowledge of product to ensure proper design to achieve maximum efficiency.

Is the UK firm the only one selling these now? YES, Hy-Save UK has refrained from re-newing existing distributor licenses. This has eliminated the issues of inadequate installations. We remain on top of quality control at all times and have experienced little issues or maintenance problems. I did find a website from the UK for these but it has not been updated in almost 4-5 years! That made me wonder if these were still being sold or if the website is waiting to expire by neglect. Hy-Save UK has updated it website to reflect current news etc. Currently pumps are available through Hy|Save located in Tampa FL, same website hysave.com for further contact information.

My understanding is the original firm was sold to a branch of an electric utility, which was then discontinued until a license was sold to a firm in the UK (the one mentioned in the website as Hy|Save UK.) Hy-Save UK owns all patents and rights to the Hy-Save Technology.

No detail is too small to mention.

Thank you for your comments and thoughts.
All is well at the Hy-Save, hardly expired or neglected.

US Iceman
27-06-2008, 02:23 AM
I would have preferred it if you would have not included all of the additional comments as mine. I don't recall making all of those and should not be given credit for doing so.:off topic:

Magoo
27-06-2008, 02:19 PM
Hi all.
I have one exposure to a Hy-save system and beleive the concept has merit, if applied by competent designers. this particular installation had a history of eating 60HP DWM 8 cylinder compressors on a regular basis. I was asked to give an indepenant opinion. me always up for challenge said I would look and report. Well the Hy-Save system worked a treat, really well. The problem was the design / application company did not figure the improved performance of compressor and system in general. Did not apply the correct TXV's and suction line sizes, fudged the whole ****-up by setting TXV superheats high and forgot that suction pressure drop and to complicate issue added a suction strainer at compressor suction pipe dimension and one filter only, suction line long and complex. Result liquid slugging, low system performance. Recommended that the TXV be re-sized and suction line be upsized and pi** off the suction, [single core ] filter.
Client reaction was, tooooo much money, two weeks later another compressor failed, smashed pistons the usual flood back senario.
There is a saying in that you can lead horse to water, but try to make the sucker drink.

magoo

Optimum
30-04-2009, 02:41 PM
This is a quick-fix to sort out re-expansion in a high-rising liquid line for poorly designed systems - nothing else. If a system was properly designed and installed in the first place, it would not even be considered.

In the UK this is being marketed for packaged liquid chillers with no payback, and no thermodynamic benefit over inverter / FSC on condenser fans and electronic expansion valves for DX.

There is no use for flooded applications.

All of the manufacturers have looked at it and discarded it for the above reasons.

Additional controls, modified pipe work (increased in diameter to maintain full liquid at the pump inlet) increased refrigerant charge the norm.

if you see it on a packaged plant, cut it out.

I hope that this helps?

Magoo
18-05-2009, 04:37 AM
OPtimum. some where along the way you have got the wrong end of the stick.
The principal as I understand it is that the compressor flow rate is increased by reducing the compression ratio, high flow rate etc., compensated by introducing a pump that supplies the correct supply flow/ rate pressure to the TEV. IE sub-cooling liquid to TEV. Therefore the increased TEV performance has to be included, the increased flow rate in suction has to be included in all calculations. Every two degrees below nominal system / traditional design increases TEV by one percent.
magoo

US Iceman
19-05-2009, 01:47 AM
Providing liquid under pressure (by pumping) will create a form of subcooling that allows the liquid refrigerant to arrive at the TXV in 100% liquid form. The only way this improves performance is if the liquid was flashing in the liquid line before and after the addition of a liquid pump the flashing disappears. You are back to design operational spec, not an improvement.

Performance gains from subcooling only appear if the liquid temperature is reduced therefy providing a lower enthalpy. That is the only way to increase the net refrigeration effect per unit mass (kg or lb), that applies to liquid.

Look it up on a PH diagram...;)

You have to be careful of what this system takes credit for. Some is valid, some is fictional IMO.

HallsEngineer
19-05-2009, 01:57 AM
Hello all,

Well, where to start. The compressor is a pump and the system would have to be pretty big to warrant running a liquid pump over making the head pressure higher and adding an economiser subcooler. The other problem is if the pump goes down you would have to install a bypass just in case. The increased install costs would kill this dead as customers want cheapest thing they can get. The extra duty gained from the expn valves would help but you would have to have a backup incase of failure. the best compromise is variable vr with inverter and electronic expn valves to cope with wide system conditions and economiser subcooler. I understand about the refrigerant charge but the backup has to be there. A properly designed system should not need this adaption.

Entropie
07-06-2009, 12:55 PM
The company which i work for has modified a site with three identical air cooled liquid chillers. The basic unit was a twin circuit with TXV and discharge pressure control by fan cycling on/off. The operation of the chiller is 24/7/365. The purpose of the test was to evaluate the potential of LPA against EXV and ordinary system design. The outcome of the test was that the EXV does give you the same benefit as LPA does compared to ordinary system design.

SimEnergyLtd
04-02-2011, 04:44 PM
Where do I start! LPA has been fitted to all Chillers at heathrow and Gatwick Airport about a year and a half ago as an energy saving device. unfortunatly it has proven not to work on an EXV system as the EXV will fight against the liquied pump. The installation is also very poor as they fit a liquied vessel before the pump inlet to stop the pump from running dry. On two occasions these ruptured whilst carrying out Hp tests on a general maintenance and now turns out that it is covered by the pressure vessel regs. It would be far better to fit 60htz fans and an inverter to any aircooled system as thi will have the same effect and not screw up your Exv but you still need to integrate it to the systems controls correctly. Mcquay wouldn't even warrenty circuits if they changed the compressor.

I was the development manager for this installation at Heathrow. We didn't fit the LPA on chillers with EXV's on them on the recommendation of the supplier. The rupturing of the liquid vessels was, shall we say, a little suspect and it was believed that someone had carried out a certain amount of vandalism on the chillers so that they would fail.

In the end, the LPA systems worked very well and produced some amazing savings on very large pieces and numbers of chillers. The majority of issues we had with the chillers was poor maintenance.

SimEnergyLtd
04-02-2011, 04:45 PM
I was the development manager for this installation at Heathrow. We didn't fit the LPA on chillers with EXV's on them on the recommendation of the supplier. The rupturing of the liquid vessels was, shall we say, a little suspect and it was believed that someone had carried out a certain amount of vandalism on the chillers so that they would fail.

In the end, the LPA systems worked very well and produced some amazing savings on very large pieces and numbers of chillers. The majority of issues we had with the chillers was poor maintenance.

Carrying on with this, we combined this project with VSD installations on the condense fans which both saved energy and reduced maintenance on these units.

SimEnergyLtd
04-02-2011, 04:48 PM
As a carrier Tec I had the abiltity on the flow 1 to use software to view the basic Temp sensors etc and the software caculates superheats and sub-cooling. An Exv only requires 15 psi differential pressure to open where a txv requires 100 psi+ to open. If you run the machine under normal operating conditions you find that you get a slug of liquied entering the compressor this is becouse the exv has its own parameters that control the device and a liquied pump will increse liquied flow to the orifice causing the valve to pulse in a unstable way as it opens it will send through a slugg of liquied the suction temp will drop and the valve will close right down the senses a rise in tempreture and a low suction pressure and open fast causing a new slugg of liquied and so. Carrier have a form of LPA but all it does is move the liquied through a fixed point on the exv and gives free cooling in a thermosyphon way. As for the failures with the liquied receiver fracturing you are correct that this was just a design error which has been resolved as they now have a steel vessel instead of the 41/8th pipe they used before but as there is only one company installing this equipment it is still a concern not only to the engineers but also to the enviroment as alot of the units fitted are operating on R22 still. Hysave also do not pressure test the whole chiller only the bits they work on but by fitting a pressure vessel and distrubing already tested and certified pipework they should re-pressure test to MAX operating pressures not the 100 psi they did at the airport. And then we could always start on ecomesh!!!!!!!!!

I am genuinely interested to know your thoughts on the Ecomesh concept and installation.

sterl
10-02-2011, 08:04 PM
NPSHr is Net Positive Suction Head Required (what the pump wants....)
NPSHa is Net Positive Suction Head Available (what the vessel and piping provides)
TDH is Total Dynamic Head
To get to why inducers work but sometimes don't solve issues with existing pumps, comes down to how a centrifugal pump works....they accelerate the liquid with vanes that shear against it; the higher the speed of the vanes and the higher the flow rate through the pumps the greater the deflection of the fluid stream. But additionally: the sharper the (fluid's) turn at the leading edge of the impellor the greater the shear and the greater the local pressure difference between the pressure side of the vane and the cavity side of the vane.

Shear of fluid, head developed and torque of pump are ying and yang and torque is not hard to calculate from basic curves and pump laws. What torque won't tell you is the profile performance of the impellor's leading edge in fluid dynamic sense, that is does it leave a big cavity/wake on the backside or not? That is a consequence of the shape of that edge and the local speed of the fluid flow. An inducer starts the fluid spinning before it reaches the eye of the impellor so the momentum change at the impellor is reduced....Thus less torque applied by the conventional portion of the pump. So if you have steady-state cavitation, adding the inducer will make a half-to-one psi difference at the eye of the impellor but your pump discharge pressure is going to go down by considerably more than that.
To anticipate that a size and pitch of inducer will fix all problems in an existing arrangement presumes a lot: even the pump manufacturers measure onset of cavitation with an open top tank of hot water, not a saturated fluid....And some Chemical Pump people offer 4-different styles of inducer for the same size suction pipe.

That said: for a dry expansion type arrangement on halocarbons on the high pressure side with compact machinery, its considerably simpler than a rarefied fluid with a flash volume rate of 1000:1 (as with steam condensate returned by a vaccum condenser) lifting to a high elevation. Though as indicated above: the benefits on the discharge side get a little fuzzy if low required pressure difference expansion devices are employed for the rest of the circuit.

Gary
10-02-2011, 11:01 PM
Wouldn't mechanical subcooling near the pump inlet be a viable means of preventing cavitation?