Cavitation or Vortexing?

[LiquidFire]

Hey Guys,

I have a little problem with a waste water lift station. The lift station is comprised of 2 pumps which take turns opperating. These pumps are sometimes cavitating, and I have an idea why, but just wanted to see what some more experienced people had to say about it. The wet well is approximately 5 feet from the well inlet to the centerline of the pumps intake (they pump in parallel and are approximately 7 feet apart). The owners are telling me cavitation is happening because of vortexing. They think this because of the fact that the inlet effluent 'waterfalls' into the wet well. It falls probably about a foot or so into the well. There is a steel grading to help disrupt the flow, but there are still reports of cavitation. I have visited the site a few times and watched the wet well for a while, never once saw any vortexing. So I am suspicious of vortexing. Another reason why I do not think it is vortexing is because 3 weeks ago they had major flooding in the wet well. Water levels doubled. I was not there when it happened, but the pump operator said the pumps still did cavitate a bit (whatever 'a bit' means, I'm not sure, that is just what he told me, 3 times after I questioned it) even with all the increased head. This leads me to think that it may indeed be obstructions at the suction? However I am fairly unfamiliar with how much obstructions can affect the pressure into a suction inlet. Theoretically I can see cavitation being an issue if there is enough obstruction because in order to maintain flow, pressure would drop below vapour pressure in the suction. But I am just unsure if this is actually possible on this scale. Yet it seems to be really the only explanation. Is there something I am missing?

Couple of Specifics:
- Centrifugal chopper pumps 2000GPM at 50ft
- NPSHr is being well achieved (assuming no restrictions)

Thanks Guys!

 

[Artisi]

Do both pumps indicate the same problem when they run?

During the period of "flooding" was the inlet still free falling into the the wet well?

Vortexing is not always a surface phenomonen, it can be from the sump wall or floor or even from the steel grating you have installed.

It could be pre-rotation of the flow, or the pumps running too far out on their curve, have you actually checked flowrates against head?

 

[LiquidFire]

Yes, both pumps indicate it.

The inlet was not free-falling during the flooding. Since the water level rose about 6 feet, the water would have backed right up into the incoming line.

I have done preliminary calculations and found that the head is right around where it should be (50 ft.) and am working on preliminary averages and numbers to estimate flows (both the suction and discharge pressure gages are shot on both pumps).

One other tid-bit to note: I was looking over the drawings for the wet well and the drawings indicate that the suction virtually rests on the ground in the wet well (hard to confirm if this is actually how it was built). This seems to me like a terrible design, or is this normal?

 

[Artisi]

Good practice is a minimum 0.3 of inlet diameter from the sump floor to the inlet of the pump, if more than this it is ok.

Question, how do you know it is cavitating?

It seems unlikely that pumps of this nature would be suffering cavitation unless the inlets are blocked.

I would be checking flow rates and heads against the pump curve.

 

[QualityTime]

Hi;

I just solved a problem with a person on this site who thought he had a caviation problem. See the following post

http://www.eng-tips.com/viewthread.cfm?qid=276105&page=1

If you have free fall of water into the sump you are introducing air bubbles (i.e. air entrainment) into the water. It gets sucked up and causes "cavitation" noise. But it is not cavitation by definition. You can test if it is cavitation or it is air entrainment very easily if you can get access to hear the pump. The approach to fixing the two problems are very different

 

[QualityTime]

Votexing is not the same as air entrainment

 

[Artisi]

Seems the OP has already ruled out air entrainment from a free-falling supply.
" LiquidFire (Mechanical) 2 Aug 10 8:31
---------.The inlet was not free-falling during the flooding. Since the water level rose about 6 feet, the water would have backed right up into the incoming line."

 

[QualityTime]

Hi Artisi;

He also said the operators said the pumps are "sometimes" cavitating. He should correlate "when it cavitates" and relate it to water level in the wetwell. It may not "cavitate" when the well is flooded due to storm conditions

He did say that "...The owners are telling me cavitation is happening because of vortexing. They think this because of the fact that the inlet effluent 'waterfalls' into the wet well. It falls probably about a foot or so into the well. There is a steel grading to help disrupt the flow, but there are still reports of cavitation. I have visited the site a few times and watched the wet well for a while, never once saw any vortexing. So I am suspicious of vortexing..."

 

[QualityTime]

"One other tid-bit to note: I was looking over the drawings for the wet well and the drawings indicate that the suction virtually rests on the ground in the wet well (hard to confirm if this is actually how it was built). This seems to me like a terrible design, or is this normal? "

Having the suction at the bottom of the wet well is proper design. This prevents vortexes forming underneath the suction pipe. I ran across this very situation many years ago in a very very large raw sewage pump station in a very very large municipal wastewater treatment plant. The suction was elevated and hydraulic modelling showed vortexing occuring underneath the raised suction. We raised the floor by filling it with concrete fill

 

[Artisi]

Hi QualityTime,

And what is cavitation, usually a badly misused term to explain why a pump is not operating as expected. Many well qualified engineers don't really understand cavitation so any comments from a field operative is always suspect in my mind plus comments like "cavitation is happening because of vortexing" certainly doesn't engender any confidence in what is actually going on with these pumps. Also, far too many unknows at this point to make any real judgements, the flows and heads haven't even been established yet.

 

[QualityTime]

You have to eliminate the obvious and be there when "cavitation" occurs. First of all I assume people believe cavitation is occuring because of the the "classic cavitation noise":

1. You should be there to witness the "cavitation" and record the water level. Have the station staff record water levels when cavitation occurs
2. During the "cavitation period" check to see if the pump suction is plugged with rags.
3. Throttle the pump discharge. If you still get the "cavitation" noise your problem is not cavitation. It is air entrained in the water.

Cavitation and air in the water produce the same cavitation noise but the solution to fixing either problem is very different from each other
caviat

 

[ash9144]

You may want to check the suction pipe for leaks. Even pinholes would allow air to be pulled into the line. Also check the pump pressure/flow against the curve.

 

[QualityTime]

I am going to ask a stupid question but is the pump completely flooded during startup under dry weather operation?

 

[Artisi]

You can never ask a stupid question when it comes to site trouble shooting, some of the things I have come across over the years are just unbelievable - for instance - one of the best;

"the pump doesn't go round when we started it"
"I see, tell me, did you check the pump motor alignment after installation" knowing what the problem was already and baiting the site-engineer - guessing the answer would be yes.
"yes"
"do you know that we shipped the pump / motor spacer coupling a few days after the pump /motor unit was dispatched"

Long silence, followed by "I'll get back to you".

I heard later from an other source, the coupling was retrieved from the store, fitted and aligned and the pump ran Ok.

 

[ccfowler]

LiquidFire,

If the suction inlet(s) is (are) too close to the bottom of the well, the abrupt acceleration of the flow can induce sub-surface vortex formation that can persist after either pump stops. Since you seem uncertain about the dimensions for the clearance between the end of the suction lines and the floor of the well, you can get a fairly accurate measurement by fabricating a long "L" shaped rod that can be dropped into the well adjacent to the pump. You can use the distance from a fixed reference point to the bottom of the well and then raise the rod to touch the bottom of the suction pipe with the top of the protruding portion of the rod. This should permit clarifying the height of the suction inlet from the bottom of the well with an accuracy within less than 1/4" and possibly much better. You know the configuration of your site, so you should be able to design such a measuring tool fairly readily and relatively inexpensively. (You may need to cut one or more reasonably small access holes in the floor or grating adjacent to the pump.) If the protruding rod end is long enough, you can use it as a probe to "feel" obstructions in the vicinity of the suction inlet. I would want to use this tool to probe the entire vicinity of the suction inlets to find any potentially unpleasant surprises.

I would pay careful attention to any obstructions, whether intentional, inadvertent, or just loose trash, that could cause disruptions of the flow anywhere in the vicinity of the suction inlets.

I agree with previous comments that vortex formation can be erroneously blamed for performance or cavitation problems, but vortex formation can introduce nasty, erratic problems. Once formed, a subsurface vortex can persist for a very long time (hours) without being obvious at the surface. Careful placement of baffles can help to break up "vortex laden" flows, but baffles must be used very carefully since they can induce the very vortex problems that they are intended to prevent or cure.

Vortex issues are quite thoroughly discussed in a number of handbooks and texts. (Sorry, I can't put my hands of any of these at this moment to cite specific references for you, but I have studied many very helpful documents in years past.) You will find much emphasis on avoiding abrupt changes in the velocity of the flow to avoid vortex formation. The water falling into your sump is almost certain to result in subsurface vortex formation that could (but also may not) move with the flow into the vicinity of the pump suction inlets.

Since this is a waste water system, the accumulation of debris is likely to be troublesome matter for reliable operation. If this problem is erratic in nature, it is likely that too-large objects may be blocking portions of the suction inlets, and these can play a role in both vortex formation and unfavorable flow rotation in the suction pipe.

If you have (or can arrange) a sufficient minimum water surface height in the well, it may be wise to shorten the suction pipe by some modest amount to assure that all debris will pass freely into the pump as intended.

Valuable advice from a professor many years ago: First, design for graceful failure. Everything we build will eventually fail, so we must strive to avoid injuries or secondary damage when that failure occurs. Only then can practicality and economics be properly considered.

 

[QualityTime]

Curious question....it sounds like the you retrofitted an existing pump station. Was the station "cavitating" before you made the modifications? And what exactly where the modifications?......pumps....working water levels?.....

 

[dabluffrat]

You mentioned, The lift station is comprised of 2 pumps which take turns opperating & the pumps pump in parallel. Do you know if the station & pumps were designed to run together in parallel (2 pumps both running)? Where on the curve are you currently operating the pumps, out past BEQ or back near MCSF?

Did you know that 76.4% of all statistics are made up...

 

[QualityTime]

So is there any resolution to this? We are all curious

 

[stanier]

There could of course be another explanation which is not related to hydraulics. Is there perhaps a problem with the electrical installation?