Locating Suction Side Air Leak

[butterblum]

I re-plumbed my pool pump area about 3 years ago, and all has been fine - but about a month ago, I started having some air leakage into the skimmer basket throughout the course of the day. I run my variable speed pump at around 1900RPM for 18hrs a day, and then 2400 and 2900 for the rest of the night - running the pump at full speed will successfully purge the pot of air, but it starts to leak back in over time at the lower speeds; there is a decent sized bubble (50% of the lid) by the end of those 18 hours.
I don't have a super complicated suction side setup, and I have already checked the pot lid and drain plug o-rings and applied a fairly heavy amount of lube to help them seat properly. I have also checked the o-ring on the union on the input side of the pump - it is seated in the groove properly and doesn't appear to be an issue. Assuming it isn't the Jandy valve (the screws are all tight), how do I go about finding which PVC joint is leaking air? I am assuming this wouldn't be caused by the sand filter, the heater, or the pump housing itself, but please help me understand if that is an incorrect assumption.
I should also add that I get the problem regardless of whether the skimmer or the main drain are selected.
Any help is welcome.
Thanks

 

[PoolStored]

Air in the pump basket on low speed is usually caused by one of two issues.

1) Pump lid leak. Sometimes pump basket lids will seal better at higher RPM than lower RPM because the pressure in the pump basket is lower at higher RPM.

2) Outgassing. The pressure in the pump basket is lower than atmospheric so dissolved gasses in the water can come out of solution. The lower the pressure, the more outgassing that will occur. However, at higher RPM/flow rates, any dissolved air that comes out of solution will quickly get taken out of the pump basket because of the higher flow rate. If the RPM is low enough so the flow rate is lower than the rate at which air rises in the pump basket, the air will remain in the pump basket.

There is nothing you can do about #2 but for #1, lubing the pump lid seals and drain plug gaskets can sometimes improve the problem. Use a silicone LUBRICANT like Magic Lube II. Don't use a ptfe/teflon lube. Lube the o-ring and the drain plug o-rings. I would also re-build the diverter valve in the picture with new o-rings and lube them too.

Aside from that, you've already found the best solution. Run at high speed for a short period of time, daily, to clear the bubble.

 

[butterblum]

Aside from that, you've already found the best solution. Run at high speed for a short period of time, daily, to clear the bubble.

Any theory as to why this would have just started recently?

 

[Dirk]

Using more lube won't "fix" a leak or solve for a worn or otherwise defective o-ring. You only need enough to allow the ring to slide against the other surfaces, so it doesn't bunch up and cause an uneven seal. O-ring lube is not sealant, and using too much can actually make things worse.

Use this MO to find suction-side leaks above ground, even on the areas you think you've eliminated:

 

[PoolStored]

Any theory as to why this would have just started recently?

My guess is that it is the valve. Rebuild it. First place I'd start. Valves and pump and o-rings get replaced annually in the PoolStored household.

 

[butterblum]

Don't use a ptfe/teflon lube. Lube the o-ring and the drain plug o-rings.

Thanks - I just ordered o-rings for the pump, as well as MagicLube II.
What is the reason for staying away from the PTFE/Teflon lube?

 

[PoolStored]

Order the o-rings for the diverter and rebuild it too...

Pool Lube - Further Reading

www.troublefreepool.com

 

[wireform]

This kit will replace the O rings in the diverter valve.

 

[butterblum]

Order the o-rings for the diverter and rebuild it too...

I just replaced the pump o-rings, which all appeared to be okay - however, the o-ring for the union on the suction side was brittle, despite being the same age as the rest of them.
I'm assuming this seal was the problem....

 

[butterblum]

Order the o-rings for the diverter and rebuild it too...

Pool Lube - Further Reading

www.troublefreepool.com

So far I have replaced the pot lid o-ring, the pump drain plug o-rings, the pump inlet/outlet union o-rings, and the Jandy diverter valve o-rings.
I am still having issues, and it doesn't matter whether the diverter valve is on the main drain or the skimmer return; I also have the filter bypassed to make sure that that isn't part of the issue.
I am currently running on a low speed with Saran wrap around the plumbing joints between the diverter valve and the pump inlet - can update later.

 

[Dirk]

Anything after the pump (like the filter) would not cause a suction-side air leak.

Replacing the o-rings is not the same as testing them for leaks. Did you try the running-hose-leak-detection trick?

The leak remaining regardless of the position of the suction-side manifold valve does point to the leak being between the valve and the pump (including each of those components), but doesn't rule out the possibility of more than one underground leak (one somewhere in the skimmer plumbing, and a second one in the drain plumbing). That, of course, would be less likely, but still possible.

Or if the diverter valve is not sealing properly, than a leak, say, underground in the skimmer plumbing would get past the valve even if the valve was in "drain mode."

There are several sets of o-rings in a Jandy valve, did you change all of them? Did you put the running hose to the valve after you rebuilt the valve? Also note that the running-hose trick wouldn't work as well, or at all, if you have more than one leak in the suction-side plumbing. Like if the valve had a leak and a pump o-ring was still leaking. Again, not likely, but possible. Leaks can be frustratingly hard to find, and I'm just giving you some possibilities that can throw a wrench into your sleuthing, that you have to start considering if all the easy tests and fixes aren't solving it.

 

[butterblum]

Anything after the pump (like the filter) would not cause a suction-side air leak.

Replacing the o-rings is not the same as testing them for leaks. Did you try the running-hose-leak-detection trick?

The leak remaining regardless of the position of the suction-side manifold valve does point to the leak being between the valve and the pump (including each of those components), but doesn't rule out the possibility of more than one underground leak (one somewhere in the skimmer plumbing, and a second one in the drain plumbing). That, of course, would be less likely, but still possible.

Or if the diverter valve is not sealing properly, than a leak, say, underground in the skimmer plumbing would get past the valve even if the valve was in "drain mode."

There are several sets of o-rings in a Jandy valve, did you change all of them? Did you put the running hose to the valve after you rebuilt the valve? Also note that the running-hose trick wouldn't work as well, or at all, if you have more than one leak in the suction-side plumbing. Like if the valve had a leak and a pump o-ring was still leaking. Again, not likely, but possible. Leaks can be frustratingly hard to find, and I'm just giving you some possibilities that can throw a wrench into your sleuthing, that you have to start considering if all the easy tests and fixes aren't solving it.

I have tried the hose trick to no avail, and yes - I did replace all 3 of the o-rings in the Jandy valve.
I also thought about whether it might be a skimmer pipe leak and if it is still skirting by the seal, even when the diverter is set to main drain only.
Is there a way to plug up a leak in an underground pipe without digging the whole thing up?

 

[Dirk]

I have tried the hose trick to no avail, and yes - I did replace all 3 of the o-rings in the Jandy valve.
I also thought about whether it might be a skimmer pipe leak and if it is still skirting by the seal, even when the diverter is set to main drain only.
Is there a way to plug up a leak in an underground pipe without digging the whole thing up?

Well, a somewhat drastic, but foolproof way, would be to fully isolate each side of your suction plumbing (one at a time, of course). You would cut the pipe leading to the skimmer, and cap the ends. Then run your leak tests, then restore that plumbing with couplers and then do the other side. Like I said, drastic.

Unfortunately, that wouldn't rule out the possibility of both sides having a leak! I know, frustrating. The only sure fire way to rule them both out is to isolate both sides at the same time. After cutting out both sides, you'd need to construct a temporary path to the water: you'd buy some pipe and glue that in, running it over ground and then with an elbow run it down into your pool. If that solved the leak, then you'd know it's not in your valve or pump or above ground plumbing, and definitely underground. Then you'd restore one side or the other and determine which underground branch had the leak.

There might be a better way to do all that. Professional leak detection might make more sense. They surely have tricks or equipment beyond my knowledge. I'm just describing the bull-headed DIY approach. Hopefully others here have something less drastic to suggest. @ajw22?

Edit: Oops, I just realized I didn't actually answer your question. I misread "plug up a leak in an underground pipe..." as "find a leak..." There are systems, both DIY and pro, that claim they can stop up an underground leak. They're crazy expensive, and IMO priced not by what they actually do, but rather by what it costs to do it by digging up your deck! Here:

PipeFuze

PipeFuze is a polymer based pipe injection system designed to repair plumbing lines without the hassle of cutting through concrete or digging up the line.

www.pipefuze.com

This is the one I know about. I don't have any direct experience with them. Perhaps there are competitors that have a better price.

 

[butterblum]

Well, a somewhat drastic, but foolproof way, would be to fully isolate each side of your suction plumbing (one at a time, of course). You would cut the pipe leading to the skimmer, and cap the ends. Then run your leak tests, then restore that plumbing with couplers and then do the other side. Like I said, drastic.

Unfortunately, that wouldn't rule out the possibility of both sides having a leak! I know, frustrating. The only sure fire way to rule them both out is to isolate both sides at the same time. After cutting out both sides, you'd need to construct a temporary path to the water: you'd buy some pipe and glue that in, running it over ground and then with an elbow run it down into your pool. If that solved the leak, then you'd know it's not in your valve or pump or above ground plumbing, and definitely underground. Then you'd restore one side or the other and determine which underground branch had the leak.

There might be a better way to do all that. Professional leak detection might make more sense. They surely have tricks or equipment beyond my knowledge. I'm just describing the bull-headed DIY approach. Hopefully others here have something less drastic to suggest. @ajw22?

Edit: Oops, I just realized I didn't actually answer your question. I misread "plug up a leak in an underground pipe..." as "find a leak..." There are systems, both DIY and pro, that claim they can stop up an underground leak. They're crazy expensive, and IMO priced not by what they actually do, but rather by what it costs to do it by digging up your deck! Here:

PipeFuze

PipeFuze is a polymer based pipe injection system designed to repair plumbing lines without the hassle of cutting through concrete or digging up the line.

www.pipefuze.com

This is the one I know about. I don't have any direct experience with them. Perhaps there are competitors that have a better price.

Any chance the leak could be coming from the pump itself?

 

[Dirk]

Any chance the leak could be coming from the pump itself?

Yes. The body could have a crack somewhere. Cavitation could be at play, too, I think. The impeller(?) can generate bubbles under certain circumstances (not exactly sure about the actual source of the cavitation bubbles). @mas985 knows more about that. But if you see any Russian submarines tracking your pump, it's definitely cavitation!

 

[Dirk]

Have you done a bucket test yet, to determine if your pool is actually losing water? Some leaks only occur while the pump is running. A suction-side leak in that case would not lose water. A more severe suction-side leak might not lose water while the pump is running, but would when the pump is off. A bucket test would be another data point in your troubleshooting.

 

[mas985]

Yes. The body could have a crack somewhere. Cavitation could be at play, too, I think. The impeller(?) can generate bubbles under certain circumstances (not exactly sure about the actual source of the cavitation bubbles).

Cavitation occurs when the water pressure drops below the vapor pressure of water (<1 PSIa) and the water basically boils. This can occur near the impeller inlet but as the vapor bubbles travel through the impeller, pressure rises rapidly and the bubbles then collapse and disappear. So you will never see any cavitation bubbles in the pump basket nor much past the pump outlet. Plus you will hear the distinct sound of cavitation much like pebbles traveling through the pump.


Are the swept 90s DWV and not schedule 40? If so, those are quite easy to crack so given what you have done to date, I would inspect those closely.

 

[wireform]

The socket depth of all those glue joints are shallower then the regular sch40 fittings so the chances of a air leak from them are much greater.

 

[butterblum]

Cavitation occurs when the water pressure drops below the vapor pressure of water (<1 PSIa) and the water basically boils. This can occur near the impeller inlet but as the vapor bubbles travel through the impeller, pressure rises rapidly and the bubbles then collapse and disappear. So you will never see any cavitation bubbles in the pump basket nor much past the pump outlet. Plus you will hear the distinct sound of cavitation much like pebbles traveling through the pump.


Are the swept 90s DWV and not schedule 40? If so, those are quite easy to crack so given what you have done to date, I would inspect those closely.

View attachment 606166

The socket depth of all those glue joints are shallower then the regular sch40 fittings so the chances of a air leak from them are much greater.

So I am in danger of cavitation through the skimmer line when I am vacuuming - I know that I can't go above about 3000RPM on my pump without inducing it (the sound in the pump is a dead giveaway).
And I did just use off-the-shelf big box store fittings, so I wonder if I pulled too much vacuum (at high RPMs) and caused a leak at one of the joints.
Could excessive cavitation in the pump housing lead to these symptoms?
I guess I will re-pipe the suction side of the above-ground piping next year.

 

[mas985]

So I am in danger of cavitation through the skimmer line when I am vacuuming - I know that I can't go above about 3000RPM on my pump without inducing it (the sound in the pump is a dead giveaway).
And I did just use off-the-shelf big box store fittings, so I wonder if I pulled too much vacuum (at high RPMs) and caused a leak at one of the joints.
Could excessive cavitation in the pump housing lead to these symptoms?
I guess I will re-pipe the suction side of the above-ground piping next year.

Cavitation in a residential pool pump is fairly rare. In most cases, a seal/o-ring/gasket will leak allowing air into the system well before the pump cavitates so the pump ends up losing prime before it can start to cavitate.

Are you sure the pump was actually cavitating?