|
|
|
one day leak repair
Saturday, November 13 2010
As predicted yesterday, I spent the entire day today working at fixing the main hydronic solar panel, a five by twelve foot monster I made from scratch back in 2005. In order to do any work on it, I first had to take the following steps:
- Turn off Arduino controller.
- Close circulation valves where they leave the house.
- Drain the panel (as much as possible) and collect the fluid for reuse (it's expensive).
- Disconnect a number of pipes and other plumbing connections.
- Lower the panel from its current 45 degree orientation.
- Remove bits of hardware.
- Cut away the silicone caulk and remove the 60.5 by 36.625 inch glass panel over the affected section
I did the preliminaries before Gretchen awoke and then I had her help me lower the panel, which is heavy. I've lowered it by myself in the past, but it's better to have some assistance. Unfortunately, though, I'd failed to move one of the stray pieces of plumbing sufficiently far from the panel and as it was lowered, a corner of the glass hung up on it and eventually snapped off. This took out a big enough chunk that the pane will now have to be replaced (in other words, it was a $50 mistake). One tries to be careful around such big and fragile things, but errors like this are inevitable. Another of the glass cover panels has been broken now for over a year and that will also have to be replaced. So I will be replacing both panes before this fix is done.
Once I had the cover pane removed, I had to figure out how to cut out the troubled pipe on either side of the leak. In the past, back when there was plastic instead of glass on the front of the panel, I'd tried to fix this leak by burying it beneath molten solder and sectors sawn from copper pipe. Occasionally such fixes even passed pressure tests (as they did before I buried the problem beneath a panel of caulked-in glass), but the problem keeps coming back. Part of the problem is that the pipe is impossible to completely drain at this location, so every time I heat up the solder, water vapor comes bubbling through, leaving behind voids that will eventually act as leaks (even if, occasionally, they are temporarily plugged by solder flux). So my plan for today's fix was to include a T-fitting as part of the fix, and drill a hole through the panel to its backside and install a boiler drain there. That will allow me to solder in my fix while providing an easy place for vapors to escape. It's also possible that such a tap could prove useful down the road for some future panel implementation (though it's difficult to imagine how; it taps the panel about a third of the way up from the bottom).
I ended up having to drill a hole into the bad section of pipe so as to completely drain it. This was necessary before I could remove all the surplus solder around it. But even once I'd cleaned up the location, I was still at a loss for how to cut out the bad piece of pipe. It lay on the inside of an undulation in a piece of corrugated metal (the panel's heat collector plate) which it was soldered against. There was no way to attack it with a pipe cutter or even a hacksaw. So I attached a little cutter to my Dremel that looks a little like a tiny steamboat paddle. But holding the Dremel firmly, I was able to cut into the pipe, first along the part furthest from the corrugated metal collector plate. Once I'd opened up a slot along the top of the copper pipe, I could reach down to its lower surface, the part soldered against the collector plate, and cut that part away. It took awhile and I had to be careful, but eventually I'd cut through the pipe at both ends and removing it was an easy act of desoldering.
After cleaning up both ends, I drilled a hole through the panel to support the new drain and then dropped in my replacement pipe section. This replacement included not only the T-fitting leading down to the drain, but customized slip couplings I'd made from normal couplings (I'd filed down the little bumps that keep fittings from being slid down all the way over the ends of a pipe).
Once I'd soldered everything into place and attached a boiler drain to the new pipe coming down through the panel, I gave the panel a pressure test at 60 psi. I didn't observe any leaks, but I wasn't completely satisfied; I've seen plumbing systems pass such tests before only to reveal tiny leaks later. That's what led to how I spent my day today after all.
For linking purposes this article's URL is:
http://asecular.com/blog.php?101113 feedback
previous | next |
