Leaks!

OK, we had a leak. Or maybe two. Or maybe three... If we were methodical, and just kept working on it, surely the leaks could be patched...


2 Days after the plaster was done and the water was filled, we knew we were in trouble. It LEAKED! The plaster company diving team checks the integrity of the upper part of the shell that is out of water. They performed a few small patches that were cosmetic, but knew they would have to return with dive gear.	"In the last picture, you can see how low the water was. This much was leaking out every FEW HOURS!"	Leaks are usually found by diving with a small squeeze bottle of blue dye. The diver goes to the suspected location, lets the water calm down, and then squeezes a bit of dye near the plaster and watches to see if any of it appears to be flowing out of the shell.	The plaster company diver enters the water with his dye bottle. The most likely places for leaks to occur is around pipes. If the pipes were loose in the Shotcrete, and then if the plaster doesn't seal them completely, water leaks from the shell following the pipe. He found and patched (with more plaster) one small leak.


Unfortunately, that didn't solve the problem! So Todd borrowed dive equipment from our good buddy Randy - the dive instructor - and set off to find the leak! Todd choose to gear up in the big bog, and then squeeze by the rock rather than going down the ladder.	"Karen mixed Todd up a bunch of dye using food coloring from her cake decorating hobby. Turns out that the process of finding leaks isn't so easy! Any disturbance in the water at all causes swirls so that the dye does all sorts of unpredictable things."	Todd holds the tank as he gets in, and then puts it on in the water. It's a tight squeeze between the rocks. After only a few moments, he finds a BIG leak. One of the last things the plasterers do is pull the sponges and covers off pipes - and they also pulled off some plaster at the bottom of a big 6" waterfall bypass pipe.	"That hole was big enough to put your arm though - and water was leaking straight out of the shell into the rock surrounding the pond. No water ever showed near the surface - which would have helped to pinpoint the leak location. A hole in black plaster under a pipe is easy to miss even when looking right at it!"


"That hole was leaking so fast, it was hard to keep the pipe under water so that we could find it! Once found, it was simple to drop the water level below the pipes so the area would be dry and easy to work with. Once the plaster had cured for a week under water, it was not too dangerous to leave it dry while the leak was being repaired, but they did wet the exposed plaster every few hours, and more when it was in the sun."	Todd works on one waterfall bypass pipe from the small bog, and Mike works on the other one from a raft. Both pipes needed to be resealed, then they needed to be fixed so that the pipe cover could be pressed on.	The pipes are 6" solid PVC, and the covers were standard PVC caps that Todd very carefully slotted. We didn't want any curious Tosai swimming up those pipes and getting stuck!	Plastering a pond wall is a pretty easy project - unless the pond is already full of water. Needless to say, you have to be VERY careful that you don't drop any bits of plaster into the water where they will end up fouling the bottom drains. It also turns out that it's hard to do when you're floating - because every time you push, you float further away from your work. Note the ropes holding the raft in. Hanging upside down from the waterfall worked well too!


Well, that didn't do it either! It took about a month to find the next problem. By changing the water level, and looking at the leak rate, we determined that we also had a leak behind the grate that blocked off the 4 big mid-level intakes.	"The plaster company diver came back to remove the grate so that we could find and patch the leak. While the grate was just screwed in, they had done such a good job with the plaster that they had sealed the grate to the wall so that it wouldn't come off!"	Here, the diver secures himself to the wall, and then uses pneumatically powered underwater cutting tools to cut the grate out from the plaster. All the screws holding the grate and the bottom drains were screwed into special stainless steel anchors that we carefully set into the Shotcrete before the plaster.	The diver makes a final check to see that the grate is removable. Then Todd mixed a batch of plaster to re-seal the area behind the grate. It was originally sealed with an epoxy-based liquid rubber - that obviously wasn't as good at sealing Shotcrete as plaster! Todd replaced the grate, and we figured we had the leaks licked!


"That was not the case. It still leaked! Now, how will they find it??? "	Meanwhile, the rest of the finishing work progressed. Here, Mike helps Todd use an ABS welder to weld a plastic grid into the large oval returns in the 1500 gallon Quarantine tank. This tank is plumbed completely independently of the pond. Again, we were concerned about Tosai swimming up the pipes when the pump wasn't on.	"Todd did some more diving with his dye bottle, and the plaster company diver dove several more times with his dye bottle. Todd found one small leak in a crack that had developed from screwing the bottom drains down too tightly. We figured maybe THAT was the last leak. Nope..."	Meanwhile, we found another interesting problem. While all the diving was taking place, we discovered that the slide valves for the waterfall bypass weren't sealing properly - and the valves were too hard to operate. Turned out to be a manufacturing defect in the valve seals, along with not properly supporting the pipe on either side of the valves. Mike dug a hole in the already planted bed, and carefully replaced the valves.


Todd went back to check that none of plaster on the leaks we had already patched had broken loose. Not one leak we have ever patched has leaked again. However...	"It still leaked. And it was getting COLD! By October, the water was too cold to dive even with a drysuit, so they just had to be satisfied with leaving the leak until the water warmed in 2003..."	While we weren't diving, it seemed like a good time to 'age' the plaster by lowering the pH, holding it low for a few days, and then raising it back up. We did this with a 30 gallon barrel of high grade Muriatic acid to lower the pH, and LOTS of baking soda to raise it back up. While this makes the plaster a bit less smooth, it keeps the pH from going over 9 after the fish are added!	The 2002 season came to a close, and the pond was covered to reduce the leaves and junk from accumulating over the winter. With none of the fish swimming in the pond - we could treat it just like a swimming pool. The tarp made a cheap temporary cover. It was held down with water bags, just like a pool cover.


"As the water warmed in the spring of 2003, they decided to keep the pond chlorinated like a swimming pool to facilitate diving to find the leak. They also decided that Todd would either have to buy expensive scuba gear, or find another way...	Todd found another way. The compressor used to pump air 12 feet down to the fish puts out enough air to breath while diving! Todd modified an old scuba regulator to take low-pressure air, and we connected a long air line to the line that was supposed to run air into the goldfish pond.	Here, Todd enjoys an unlimited supply of air, and can stay on the bottom as long as he wants to search for leaks. It's still not a heated swimming pool, so that time is limited even with a wetsuit due to cold!	"Todd methodically dye tested every pipe over a over again. No leaks were found, but the pond was still leaking. So they set up an accurate leak gauge in the basement, and then lowered the pond 1 foot at a time and recorded the leak rate. It was leaking about 1000 gallons per week, and the leak rate slowed at a linear rate as the water level dropped."


We decide the leak must be in the bottom 4 feet. So Todd carefully checks all the pipes entering the pond below that level.	"Todd checked the vortex for possible leaks by sealing it off completely, and then opening the valves. They also noted that the pit around the valves was completely dry."	The plaster company diver came back and did another check of all the pipes penetrating the shell. He didn't find any leaks either!	"Todd checked the pumps and piping for leaks by sealing them off, and then opening them. The leak rate never changed."


"Well - it MUST be somewhere! They decide to test to see if it's evaporation by putting the cover back on the pond. The cover touched the water, and was sealed as well as possible to the edge rocks to eliminate evaporation. The leak rate of 8mm per day only changed by about 1 1/2 mm. The 1 1/2 mm was the evaporation rate in the worst of the summer - so the leak rate was still 6 1/2 mm/day. Doesn't sound like much - but still comes out to about 1000 gallons per week!"	Mike does some cosmetic work, and then does the final sealing of the rocks between the Koi pond and the goldfish pond.	"Todd decides that what he needs to be able to do is to seal plastic to a small area of the pond wall, and then put dye under the plastic for a more accurate test."	The plaster company diver checks for any leaks that may have redeveloped around the waterfall bypass pipes and the skimmers.


Todd creates a device for sealing to the wall of the pond so that we can check for leaks around the lights and other pipes that penetrate the wall.	"The water had already started to get cold, and Todd could only dive for about 20-30 minutes at a time which only let him check two small areas per dive."	Todd kept at it. But the final consensus was that none of the lights had any leaks at all! At that point, that was not the news we wanted to hear - we just wanted to FIND IT!	"By the fall, Todd had an earache and about 100 hours of 'bottom time' for the year! What we needed was someone to sprinkle some fairy dust and give us some hint of where to look!"


We decided that the most likely place to concentrate our efforts was the bottom drains and their associated plumbing. Karen made these very heavy sand-filled snakes to seal a piece of plastic around the bottom drains.	"Todd then put dye under the plastic and watched to see if there was any movement in the dye. NOTHING! "	To further test that the method WOULD find a leak if there was one, Karen allows a tiny trickle of water past a bottom drain valve, and it instantly sucked the dye out. OK - the system worked - but WHERE WAS THE LEAK???	"The 2003 diving season came to an end without finding the leak! They know it's there - it had been steady all summer! The pond was netted, then covered for winter, and it sits under the ice and snow, waiting. Everyone hopes that the spring of 2004 will bring the lucky day on which the last leak is found and patched!!!"

OK, we had a leak. Or maybe two. Or maybe three... If we were methodical, and just kept working on it, surely the leaks could be patched...

"In the last picture, you can see how low the water was. This much was leaking out every FEW HOURS!"

Leaks are usually found by diving with a small squeeze bottle of blue dye. The diver goes to the suspected location, lets the water calm down, and then squeezes a bit of dye near the plaster and watches to see if any of it appears to be flowing out of the shell.

Unfortunately, that didn't solve the problem! So Todd borrowed dive equipment from our good buddy Randy - the dive instructor - and set off to find the leak! Todd choose to gear up in the big bog, and then squeeze by the rock rather than going down the ladder.

"Karen mixed Todd up a bunch of dye using food coloring from her cake decorating hobby. Turns out that the process of finding leaks isn't so easy! Any disturbance in the water at all causes swirls so that the dye does all sorts of unpredictable things."

Todd works on one waterfall bypass pipe from the small bog, and Mike works on the other one from a raft. Both pipes needed to be resealed, then they needed to be fixed so that the pipe cover could be pressed on.

The pipes are 6" solid PVC, and the covers were standard PVC caps that Todd very carefully slotted. We didn't want any curious Tosai swimming up those pipes and getting stuck!

Well, that didn't do it either! It took about a month to find the next problem. By changing the water level, and looking at the leak rate, we determined that we also had a leak behind the grate that blocked off the 4 big mid-level intakes.

"The plaster company diver came back to remove the grate so that we could find and patch the leak. While the grate was just screwed in, they had done such a good job with the plaster that they had sealed the grate to the wall so that it wouldn't come off!"

"That was not the case. It still leaked! Now, how will they find it??? "

"Todd did some more diving with his dye bottle, and the plaster company diver dove several more times with his dye bottle. Todd found one small leak in a crack that had developed from screwing the bottom drains down too tightly. We figured maybe THAT was the last leak. Nope..."

Todd went back to check that none of plaster on the leaks we had already patched had broken loose. Not one leak we have ever patched has leaked again. However...

"It still leaked. And it was getting COLD! By October, the water was too cold to dive even with a drysuit, so they just had to be satisfied with leaving the leak until the water warmed in 2003..."

"As the water warmed in the spring of 2003, they decided to keep the pond chlorinated like a swimming pool to facilitate diving to find the leak. They also decided that Todd would either have to buy expensive scuba gear, or find another way...

Todd found another way. The compressor used to pump air 12 feet down to the fish puts out enough air to breath while diving! Todd modified an old scuba regulator to take low-pressure air, and we connected a long air line to the line that was supposed to run air into the goldfish pond.

Here, Todd enjoys an unlimited supply of air, and can stay on the bottom as long as he wants to search for leaks. It's still not a heated swimming pool, so that time is limited even with a wetsuit due to cold!

We decide the leak must be in the bottom 4 feet. So Todd carefully checks all the pipes entering the pond below that level.

"Todd checked the vortex for possible leaks by sealing it off completely, and then opening the valves. They also noted that the pit around the valves was completely dry."

The plaster company diver came back and did another check of all the pipes penetrating the shell. He didn't find any leaks either!

"Todd checked the pumps and piping for leaks by sealing them off, and then opening them. The leak rate never changed."

Mike does some cosmetic work, and then does the final sealing of the rocks between the Koi pond and the goldfish pond.

"Todd decides that what he needs to be able to do is to seal plastic to a small area of the pond wall, and then put dye under the plastic for a more accurate test."

The plaster company diver checks for any leaks that may have redeveloped around the waterfall bypass pipes and the skimmers.

Todd creates a device for sealing to the wall of the pond so that we can check for leaks around the lights and other pipes that penetrate the wall.

"The water had already started to get cold, and Todd could only dive for about 20-30 minutes at a time which only let him check two small areas per dive."

Todd kept at it. But the final consensus was that none of the lights had any leaks at all! At that point, that was not the news we wanted to hear - we just wanted to FIND IT!

"By the fall, Todd had an earache and about 100 hours of 'bottom time' for the year! What we needed was someone to sprinkle some fairy dust and give us some hint of where to look!"

We decided that the most likely place to concentrate our efforts was the bottom drains and their associated plumbing. Karen made these very heavy sand-filled snakes to seal a piece of plastic around the bottom drains.

"Todd then put dye under the plastic and watched to see if there was any movement in the dye. NOTHING! "

To further test that the method WOULD find a leak if there was one, Karen allows a tiny trickle of water past a bottom drain valve, and it instantly sucked the dye out. OK - the system worked - but WHERE WAS THE LEAK???