I wish I knew what was going on ...

[Jpadie]

Those with keen memories will recall that I have recently converted a Magiline pool to a traditional filtration system. Magiline is a proprietary pool system that uses a venturi effect to filter via cartridge filters.

Since mid-season last year I have filtered the pool via a 63mm hose dropped into the pool basin. In March/April I dug out the skimmers and replumbed one so that it directly fed the pump. Below is the schematic of the skimmer/filter and so photos of the replumbing. just for context.







Life was good. Check valves not needed, suction was fine and the sight glass was full all the time. Happy days. The distance to the pump was about 2.5m head was about 30cm.

Last weekend I plumbed things in permanently. That included the previously existing 63mm suction pipe, the 63mm feed you see above and the same from the other end of the pool. I've also replumbed a line to a pre-existing line for the robot. Here's what it looks like


the near side is the same line that you saw in the first photo. There are three 90 degree angles added to the line. the total length of the pipe work is 5m. The vertical head remains the same at 30cm.

the middle feed is tapped off at the moment. that will go to the robot socket on the side of the pool. the secondary (vertical) point is to shut off the suction and enable a booster pump feed for a polaris 280. this, as said, is shut off. The intention is to be able to use the same plumbing both for a vacuum and a polaris 280.

The third feed is the other skimmer. Identical to the first above. The line is a continuous line of flexible 63mm all the way to the final riser into the pump. Total length is about 20m; 3 90 angles are in that circuit.

the feed at the top attaches to the existing (untouched) 63mm feed from the pool. The line to the original plumbing is about 150cm. Total length to the suction point is 17m. However the suction port is only 50mm even though the plumbing is 63mm. there is a reason for this (linked to the Magiline system).

there are no leaks on the flow side. beyond the odd drip from the waste pipe of the filter multi-port valve. I've put a shut off valve on that.

there are no leaks on the suction side that I have found. Done the foam test etc.

If I shut off all the valves apart from the known good line and start the pump I get almost no head. a week ago the suction was better than {insert whatever Jeremy Clarkson comment you can think of}

if I open all the valves then I get enough head to start the SWG but we're still only at the top of the inlet pipe. not the nice full pre-filter that I was getting 2 weeks ago. video of pre-filter (after 90 minutes of running the pump. Same this morning). Here is a video of the outflow where you can hear the cavitation. this is still going on today. Turning the multiport valve to waste I get slightly better but still no filling of the sight glass and still very slow to prime.

What have I done to deserve this from the Gods of swimming pools? The only aberrance I can think of is that there were a few days during which the pump and hydraulic circuit were free of water. Perhaps the seals in the pump have dried out? All the other seals (in the pipe unions) have been cleaned and newly greased.

 

[YippeeSkippy]

I'm having problems viewing the photos

 

[Jpadie]

I'm having problems viewing the photos

Hmm ... I've reuploaded the photos - i hope these work better. they are all quite large so give them time to load otherwise they might look a bit rubbish in lo-res.

another potential thought is that perhaps the shut off valves are not air-tight even if they do stop water flow. I will try plumbing those in when I've next got a moment.

 

[Jpadie]

the shaving foam test, the smoke test didn't spot any leaks. wrapping the unions with cellophane didn't find anything either.
in the end I removed the manifold with its valves and rigged up a pressure testing valve on the 63mm union to the pump. Pressurising the manifold revealed that there were significant issues with three of the four glued joints. It's annoying as the shut off valves are expensive and the manifold itself is very expensive. I've glued hundreds and hundreds of joints over the years and only once before had a problem. For three out of four joints to have problems I can only assume that either there was something wrong with the glue or that there was perhaps something slightly off in the tolerances between the manufacture of the shut-off valves and the manifold. They may be different manufacturers in fact.

I managed to get to a retained pressure of 3bar but adding a circlet of PVC (cut from another union) to one joint and gluing both the inner surface and the leading edges. for the other leaks slapping on acetone to start 'melting' the area around the joint, followed by more glue seems to have worked. I doubt it will hold though ...