Normal loss vs a leak...

[davelinde]

OK... so I imagine this a complete newbie question, so hopefully easy.

How much water loss is "normal" and how much indicates a possible leak? We just added water to a new build this week and I've been trying to remember where the waterline is and I swear it moves... although today I'm not sure again. All the plumbing held pressure for a month or more except for three lines... The bubbler niche had no way to cap it and the conduit to the bubbler light and the pool light were never pressure tested. If I'm losing water it is is in increments of a 1/16" per day... though it could all be my imagination.

For now I figure I'll just watch it and if eventually it's to the bottom of the tile I'll know the rate of loss. So... how much loss is "normal"?

 

[Richard320]

Loss varies with temperature and especially for me, wind.

Try the bucket test. Fill a plastic bucket with water and set it on the step. Mark the water level. Check again in a couple days. The pool and the bucket should both have dropped the same amount. If the pool has lost more than the bucket, you have a leak.

 

[JasonLion]

Losing 1/16" a day is well within the normal range.

 

[davelinde]

bucket test... that is just so simple it is clever! Thanks!

 

[davelinde]

So after 9 days the loss is seeming like more than 1/16" per day - maybe quite a bit more
Setting up a bucket took a little doing since none of my buckets were as deep as my first step. I was finally able to get a bucket stacked on a bucket to work using a few bricks to keep the whole contraption from floating away.

Fingers crossed the levels track now... if not I'll be back here begging for the steps to isolate the source of a leak.

 

[mas985]

When doing a bucket test, you need to be careful to not have too much bucket out of the water. The smaller the better. The rim of the bucket can block the wind and reduce the evaporation within the bucket while outside the bucket is still the same. This is why the test tends to be more accurate without wind.

 

[davelinde]

When doing a bucket test, you need to be careful to not have too much bucket out of the water. The smaller the better.

The plumber did say the levels might deviate a little over time. In this case I've unintentionally nailed it - when I had the contraption all set the rim of the bucket was exactly at the level of the pool, quite by accident. We'll see how it looks after 24 hours.

 

[davelinde]

Still in limbo... after 48 hrs the level in the bucket and outside the bucket seem to be tracking. Odd how the level seems to be plummeting when I use my memory and the waterline tile as a gauge but using the bucket the level seems to hardly change over 2 days (GOOD!!).

I'm also trying to remember back to those steam tables from thermodynamics... I'm thinking that when the water was "warm" and the air temps dropped a lot along with the wind picking up... that would explain a bunch of water leaving the pool last week? I continue to pray it's not a leak... and so far, so good.

 

[mas985]

Evaporation is primarily dependent on the dew temperature, water temperature and air speed. There is a relationship between air temperature and dew temperature (i.e. humidity) so in some cases when air temp drops, dew temp can drop too but not always.

 

[davelinde]

OK... so initially the rim of the bucket was so close to the water level that water would ripple in and out of the bucket and the levels seemed to track. The pool level is finally low enough for the test to be valid and in the last 24 hours the levels diverged about 3/4"... so I'm guessing that = leak? It should become apparent in another 24 hours I suppose.

What is the standard approach to troubleshooting this? The majority of my plumbing held 30 psi for a couple months although the main drains were initially losing a few psi just before gunite. The PB said it was the plugs weeping a little and when the mains got covered in gunite, in fact, that drop stopped and held at 30 psi along with the rest of the plumbing (the drains were initially at 50 psi with the slow leak).

There are 3 lines which were not pressure tested at all... the feed to the bubbler was left open because the bubbler/light niche combo had no clear way to cap the feed line. Also the conduit feeding he flood light and bubbler light was never under pressure - I suppose electrical conduit could leak but it's not normal to pressure test it?? I'm open to expert ideas here but I think the place to start would be the untested lines? Is there an easy way to isolate them? The bubbler is on a shelf, so in theory I could "just" pump out 10" of water and eliminate that (though it would eliminate the returns and skimmer too...). I still have the original plug for the skimmer, so I could test it easily. Would I need to pull the drain covers and plug the main drains again to test that? I'm not sure I can dive well enough to do all that work under water so maybe that's when I start paying for help. Is there something I could pack around the wires at the conduit to isolate or at least greatly slow any leak there? In fact, if that works would that be a long term fix?

On the feed side I am fairly sure it's not the waterfall since the water drops over night and we never run it at night? But for the returns I suppose they can leak back? We have not used the vacuum yet... but I suppose the safety cover is not actually water tight. Hmm... do I unscrew the fittings and cap all those lines somehow to test?

For now I just want to find this. I am dreading what fixing it will mean since (if I am correct and I'm leaking...) the culprit is surely buried somewhere under new (expensive) landscaping or pavers. I'm still holding out a slim hope there is no leak, but my hope of that is fading.

 

[davelinde]

The guys installing the solar today said there can also be grout leaks in the waterline tile and skimmer leaks in how it's set... They suggested letting the pool leak down and see if it stops at any point. Logical I guess. I'm guessing I'd want the 30 days of acid startup on the plaster to be done before I let the waterline go below tile though.

Next datapoint... the solar install came with the cover which is on now. I expect the bucket (under the cover now - to lose nothing and pool evaporation to drop to very little? Let's see how the water level does in 24...

 

[mas985]

I was just about to post that about the skimmer. That is a fairly common leak location and probably more likely than the plumbing. Plus it is much easier to fix.

 

[davelinde]

skimmer... common leak location and probably more likely than the plumbing. Plus it is much easier to fix.

oh I really, really hope so (assuming I cannot get my first wish of no leak at all).

Help me understand - the skimmer itself might leak? So just putting the drain plug back in the bottom would not change that?
Now that I think of it there was a bit of a question there.... with the paver height we were told to get a collar extension for the skimmer - so we did. But then when the paver guy tried to use it he told us we had gotten the wrong size collar. DW was on the phone trying to find the right collar quick and the paver guy came back and said something like... never mind, I can make this work, and he got the whole thing assembled without the collar somehow. Looking at it the assembly I see a few places it could be leaking and in fact the water level just today dropped below the last seam in the upper assembly taking that out of play for the moment. Time will tell.

 

[mas985]

It is the plastic part of the skimmer the butts up to the gunite and/or tile. Usually there is some sore of grout line between the plastic and tile and sometimes the grout falls out. New grout should fix that. Also if there are any gaps between the plastic parts due to the improvised installation, you can use an underwater silicon sealant.

 

[davelinde]

OK, the saga continues.

Over the course of several days the level in the bucket and the level of the pool diverged an inch or more... so today I took the bucket out on the assumption I've got a leak. Yesterday I marked the pool level with tape and in 20 hours I've lost 3/8" WITH the pool covered by a solar blanket. (The bucket was under the blanket too and seemed to have lost essentially zero).

I am now officially unhappy...

So - moving towards a solution... now what? I'm going to let it leak down to the bottom of the tile and see if the loss rate stays the same. My prime suspects are still the 3 lines that did not pass a pressure test, which are 2 electrical conduit and the feed to my bubbler.

When is is OK to let the water line sit on my plaster and not risk a permanent stain? I'm guessing it's around the 6 week mark?

I can pump down the pool 10" and eliminate the bubbler and skimmer in one test as soon as it is OK for the water to be that low. grrr... I HATE leaky stuff.

 

[davelinde]

back looking for advice...

Now that we have the solar cover on 7x24 I'm thinking the evaporation loss should be near 0"? Over the last 72 hours we've lost about 1/2" with the cover in place, so it seems water is going somewhere, though not at a high rate. Am I right I need to find something? Is the best course of action to just let it leak down past the various penetrations until it stops and then I'll know what to fix? or is there something more proactive I can do...

 

[mas985]

Yes, evaporation should be significantly reduced with the cover on the pool. However, it may not be 0 because there may be exposed areas around the edge.

Have you inspected in and around the skimmer for any loose or missing grout? This is where the plastic meets the gunite/tile/plaster.

 

[davelinde]

Have you inspected in and around the skimmer for any loose or missing grout? This is where the plastic meets the gunite/tile/plaster.

I just looked again - the tile is cut almost flush to the skimmer so the grout line on one side is very tight ~1/16" and on the other side nearly invisible. I ran my fingernail around the perimeter and there IS grout in there all around without voids. The pebbletec around the lower part looks tight too. On the inside there are a few places where I would guess it could leak but at this point the water has dropped below all of them and they are dry now.

On a happy note - we just got our water bill for filling the pool once and topping it with 2" more initially. $18... so at least adding water to this sieve won't break the bank until I find the culprit.

hmmm... on pondering - a question here. The skimmer is set into the notch in the gunite with cement all around it - so even if there was a gap between the plastic frame and the tile, where would the water go? Even if the water was flowing behind the tile wouldn't the gunite shell hold it?

hmmmm.... OR - is the gunite so porous that if there is a path for water the gunite will absorb it and release it to the surround sand?

hmmmmmmm... would any of this imply that just living with slow leak will cause maintenance issues later?

 

[mas985]

Gunite is porous not to mention when it drys it develops cracks. Also, the plastic/gunite interface is not a very good seal. So the plaster, tile and grout provide the water proofing. However, if there is any small gap anywhere, you can seal it with a waterproof sealant just to make sure.

Once you have eliminated that as a possibility, the next step would be to pressure test the plumbing. You can plug the skimmer connections with a threaded plug and on the return side, you can use rubber pipe plugs.

Or you could just live with the leak. It is up to you. I don't think it would cause any long term damage because the ground can absorb a lot of water.

 

[davelinde]

"Once you have eliminated that as a possibility, the next step would be to pressure test the plumbing. You can plug the skimmer connections with a threaded plug and on the return side, you can use rubber pipe plugs"

I still have the plug for the skimmer. I wondered if there were plugs for the returns too - if I can cap all of those (8 total) that's an easy test to repeat (those lines held 30 psi for months during construction). Is there any way to seal off the 1" electrical conduit? I have two of those and they were never pressure tested. It seems that if I can pack something around the cord in the conduit and the leaks stops that would, in fact, be a fix!