Sump pump cavitating like crazy! Help Please...

[Marrowbone]

Just an observation. I've been messing with my own plumbing, changing my small mag drive return from submerged to in-line. Temporarily, I have the pump sitting on the floor next to my new undrilled AGA sump. It pulls the water up out of the sump through vinyl tubing to a pair of barbed/threaded elbows at the sump rim, then back down to an barbed elbow screwed into the pump... then straight up to the tank (for now).

Anyway, I've noticed that the double elbows are restricting the flow just enough to create the bubbles of water vapor so nicely described by MarkO. I can clearly see them after the elbows and not before, leading me to believe that any significant restriction of water flow before the pump could generate bubbles. Turbulent water passes through the obstuction into a very low pressure area (created by pump suction) and in some very small places, would have low enough pressure for the water to boil.

I think caecosystems has an excellent idea for a fix that applies to larger pumps and their even more complex flow issues.

Ved

 

[michaelross6]

Andrew:

The other posts described cavitation, so I won't go into that. One thing I don't think anyone else has mentioned is that you have a strainer on the intake. Since cavitation happens when the flow of water in the inlet side is insufficient, I'd remove that strainer if possible. The reason that slowing the outlet flow stops the cavitation is that the flow INTO the pump also decreases, thus decreasing negative pressure on inlet side. Increasing the diameter of the inlet plumbing helps, but if the short run you have is the same size as the inlet opening, that should not be an issue.

Another source of bubbles is the valve itself. You may try using some vaseline or silicon lubricant on the O-rings of the valve assembly to see if that helps.

FWIW,

Mike

 

[drewread]

There is no strainer and there never has been.

 

[caecosystems]

Marrowbone,
Thank you for the recognition. Your bands website is awesome. If your plumbing goes up and over the wall of your sump, do you ever have problems, when you turn your pump off, losing prime? If so..a foot valve will help keep prime in the line. 8)

 

[michaelross6]

Andrew:

No strainer? In the close-up of the sump where you mention the water levels, there appears to be some type of black, 2" long "something" with slots in it. I can't see it very clearly, though.

If you truely have cavitation, it is not a funtion of outlet flow, only too low of a pressure on the inlet side. Any change that affects that pressure will change the cavitation. Decreasing resistance will decrease cavitation, whether it is slowing outlet flow, or increasing the diameter or decreasing the length of run of the inlet side.

FWIW,

Mike

 

[drewread]

Nope... no strainer. What you see is actually the bulkhead (It is about 2" long.)

;c)

 

[Anonymous]

So,

What's your plan?

 

[drewread]

Well at this point the plan is to completley drain the sump and put a series of 5 or 6 over under baffles closer to the 'input' end of the sump. I will also plumb the protein skimmer return back into the input section of the sump. All of this will be in order to reduce the number of bubbles that are getting through to the sump 'output'.

As I will have all of the water out of the sump while the silicone is drying I will attempt to dip the output bulkhead and all associated fittings in PVC glue in order to eliminate the possibility of an air leak at that point.

The I will see what will happen.

I will let you know when that is finished.

 

[Anonymous]

Do as you think best, but I don't think your problem is air entering the intake, or you would not be able to control it by reducing the pump output. Just out of curiosity, what are the specifications of the pump. Size, brand, type, HP?

 

[Marrowbone]

[hijack] Thanks caecosystems, flash animation rules! Nope, no problems with the return pump so far, though it is not time-tested. The pump should remain flooded in an outage since it sits well below the water level of the sump, and shouldnt completely drain. [/hijack]

Ditto cwa46, air may not be entering the intake. It could just be water vapor. I still like the idea of making the run of pipe to the pump at least 10 times as long as it's diameter per caeco.