Pump Cavitation

[AmherstReef]

After installing a more powerful external return pump I now know what cavitation means. I had massive micro bubbles and could hear the air inside the pump, now I know why.

The pump inlet and outlet are 1.25" and of course my bulkhead and return plumbing is all 1". I checked with the manufacturer and they said this was fine but obviously its not helping.

I've eliminated the cavitation for now by closing the ball valve a bit reducing the outlet flow. Without re-drilling the bulkhead I will need to try some other changes. Between the bulkhead and pump inlet there is a union, a 90 ,a ball valve and a few couplers. All of these could be causing the cavitation so I'm going to redo this and eliminate what I can to see if that corrects it.

 

[mr.maroonsalty]

Since it sounds like you want to save your existing plumbing layout, I'd start by stepping up the pipe from the bulk head as large as I could fit maybe 2" or even 3" then make the turn to the pump and reduce it down to the 1 1/2" inlet. This may not be enough to fix the cavitation but it's a huge step to reduce friction loss before the pump.

A 1" 90 el is like adding 5' of 1" pipe. Depending on the flow rate that could turn into a good deal of resistance, eapecially after adding on the equivalent lengths of the adaptors. Are you absolutely sure the water in the sump is high enough that a small vortex isn't to blame? There is a rule of thumb the proclaimed bean masters have for how much straight pipe you might need after a radius on the pump inlet. Straight runs from the sump to the pump is ideal.
I'm brainstorming here but what would happen if you added an over the top 3/4 run tying into the inlet with a tee? Would there be a significant enough suction/Venturi like effect to add enough flow, especially if it stepped straight to 1 1/2 or 2".

 

[Srixon]

Cavitation is a restriction on the inlet causing "cavities" that implode on the impeller and pump valute causing damage. Your problem seems to me that your pump is sucking air in ....do you see any vortexing at the in take? The valve on the discharge is a step but it would seem to me that a better altertnative would be in increase sump volume in the chamber where the intake is or making a sheild to break the vortex if that is happening.

 

[AmherstReef]

There is definitely no vortex, the bulkhead is at the bottom of a 40G breeder.

I need to get a new bulkhead and then start removing items between it and the inlet. The union needs to go and I may swap the 90 for a 45.

 

[SkiFletch]

Thought this was the case from the beginning, too much inlet restriction. Minimum manufacturers reccomendations often really push the limits, I always upsize mine, it just makes the pump happier. If you want to try and save yourself some work and keep the bulkhead, you can try bumping up to 2" pipe and have a longer/straighter run into the pump. If you can get a long straight run in, you'll at least make the water flow more laminar and help smooth out the inlet flow. That coupled with the larger diameter of the pipe will change where/how the pressure drop occurs and MIGHT be enough. Sounds to me though like you'll have to drain the sump anyway you slice it the way it's setup, so it might be worth it to just go straight for a larger bulkhead size... Restricting the outlet because you're cavitating from inlet restriction over-works the pump and is NOT a recipe for long pump life. That baby is going to get HOT and will drastically shorten shaft and bearing life. Get a small desk fan or something going to blow on the motor and keep it cool until you're ready to do the repair.

 

[AmherstReef]

Interesting thoughts Ski although my understanding of centrifugal pumps is that restricting the output REDUCES the amount of work it does and that in turn REDUCES the power consumption and also the "strain" on the pump.

Although by no means am I an expert

 

[SkiFletch]

The amount of "work" a pump does is directly proportional to the pressure drop it is forced to create. By restricting the outlet, you are increasing the pressure drop it needs to create. In the process you are slowing the impeller which is what is eliminating the cavitation, but the motor has to draw more current to deal wih the pressure drop. More current in the motor equals more heat in the motor internals and heat is our enemy.

 

[captain7359]

Reducing the flow through a pump reduces the wattage drawn and produces less work by the pump. Pumps increase pressure from the inlet to discharge, not drop it.

with 99% certainty I can say your problem is not cavitation, but an air leak somewhere on the suction side of the pump. Nearly all of the hobby pumps we deal with are not powerful enough to create true cavitation and only some of the monster reeflo pumps might stand a chance in some situation, but it would involve serious restrictions on the pump intake. Use some liquid pipe dope or a few more wraps of teflon on your threaded joints, check for cracked fittings, incomplete fusion joints on any pvc joints, or loose screws on the pump housing. With more flow you may be drawing more water with bubbles entrained and those small bubbles create tons of micro bubbles when shredded in the pump. Reducing the flow may have slowed the flow in the sump enough to allow the bubbles to float up and not get sucked in to the pump. What make/model pump do you have and what is your sump layout like? Do you have a filter sock on the feed into the sump or skimmer outlet to prevent bubbles making it into the water within the sump?

 

[mr.maroonsalty]

Another thought if you don't want to redrill is to try and find a 1" bh that is 1 1/4 mpt on the nut side. A 1 1/4" Bh nut will thread on a fpt, but the threads on most 1" bulk heads are too shallow to receive a mpt. I don't see any reason not to use a bulk head this way if 1 1/4 threads screwed properly on it; this would make a 1" bh into 1 1/4" and with that fitting take it straight away to a larger diameter pipe.

 

[mr.maroonsalty]

Interesting capt... Plug the bulkhead and fill the feed pipe with water and wait checking for leaks.

 

[AmherstReef]

Capt there are absolutely no bubbles in my sump and yes I been using a filter sock for years. I have no leaks between the bulk head and pump inlet so hard to believe its sucking in air. The pump is a BLowHole 2700

 

[captain7359]

spray down all the connections with water to see if you can stop the bubbles and that will be the leaking joint. Can you put a 90 on the inside of the sump on the bulkhead to pull water from closer to the bottom just for kicks to see if it reduces the bubbles?

 

[AmherstReef]

I'll definitely check again for leaks but there is no way to put a 90 in, the bulkhead is to far to the bottom, besides, there is no vortex

 

[SkiFletch]

Interesting point Capt. It could be a negative leak on one of those fittings. I've seen moderate sized magdrives cavitate with enough inlet restriction, so I still wouldn't discount cavitation; but your suggestion is surely something to be tried first cause it's a lot easier. Negative leaks suck (literally) and are a real pain in the tail to find. At least you can try the water trick on it.

 

[der_wille_zur_macht]

Captain and amherstreef are correct, restricting the outlet reduces the work done by the pump. Unless you're in a fringe case (i.e. right near either extreme of the pump curve).

Manufacturers that publish real pump curves provide data that backs this up. Reeflo themselves, for instance - go look at their pump curves. As head (pressure) increases and the pump is moving less water, the wattage decreases. Since voltage is constant the decrease in watts implies current is lower.

Can we see a photo of the plumbing configuration? I am having a hard time visualizing this from your description. Diameter is only part of the puzzle. Putting a restriction (90, compact ball valve, etc) too close to the inlet can cause problems regardless.

 

[AmherstReef]

Just an update for everyone. ReeFlo told me this morning that restricting output on their pumps will not cause any damage. They also gave me this info:

Bubbles/CavitationCommon causes

- an imbalance of the CAPACITY between input and discharge, the intake line must be greater than or equal to the discharge line (including any tees within say 3-4 feet), a quick check for this is to partially close down the discharge side and see if the bubbles subside
- an elbow right at the input that creates turbulence
- an elbow or screen on the intake in the sump
- tiny pinhole openings anywhere on the intake line-to be sure apply sealant to the outside of all joints on the intake line-this especially true where any flex joins with rigid

 

[der_wille_zur_macht]

Well I guess the good news is, their advice more or less agrees with ours.

Bad news is, you still have a problem. :lol:

Good luck - if you post a photo maybe we can critique from a design perspective, otherwise - pinhole somewhere on the intake.

 

[AmherstReef]

Here is the set up, only difference is the smaller pump

 

[der_wille_zur_macht]

I'm assuming the black stuff is 1" flex pvc? That doesn't look horrible at first glance though it does look like that is a compact ball valve (which are usually not full diameter). But you seem to have plenty of head on the output so I'd be surprised if it was cavitation.

I suppose you could test for bubbles, and if you don't find any, step up to 1.5" right after the bulkhead, and switch the compact BV out for a full size (Lowes of all places has really nice single union 1.5" ball valves, and they're really cheap - brand is American Valve I think). Whether you have enough restriction to cause cavitation or you're sucking air through a pinhole, if you replace all the plumbing on the intake side you should fix the problem either way.

I wouldn't worry about stepping up the bulkhead size if you do go up with the plumbing.

 

[AmherstReef]

I may just redo and upsize like you suggest , what do you mean by compact vs full size , it's a 1" ball valve , the 1.25 or 1.5 are identical just larger. Got it at Lowes