Tuesday, September 8 2015
This morning I was having trouble getting hydronic fluid to circulate to the solar panel and back. I thought I'd eliminated all the air pockets that can act to block the flow, so perhaps another problem had spontaneously appeared. I tried disconnecting the actuators from two of the electromechanical valves (which stay open in the absence of an actuator), and when that didn't solve the problem, I was forced to do more fluid circulation in hopes of eliminating air. This eventually worked. The way you can tell the system is working is that it makes a stready trickling sound (though this sound might eventually go away once all the air is eliminated).
Over the course of the day, I put together an all-copper replacement for the rubber hose that keeps failing. (I suspect that the reason for this repeated failure is that the hose is for a section of pipe between panels that runs unusually hot, perhaps too hot even for high-quality rubber.) The result was a tube shaped like a capital-L, with a long side measuring 81 inches and a short side measuring a bit over 40 inches. I needed a 45 degree fitting at the end of the 40 inch leg to successfully meet the existing plumbing, but it turned out I had no such fittings. So I called Gretchen and had her pick me one up from Herzog's after her shift at the literacy center.
The day was unseasonably hot and humid, with lots of sun. So I waited until after the sun had set to install the pipe replacement. Since it was replacing essentially a garden hose having two female ends, I'd had to include garden hose fitting to its ends. It's actually kind of unusual for soldered copper pipe to take the place of such hoses, and there aren't a lot of hose fittings designed to be soldered onto copper pipe. So I generally have to improvise, either by lathing a brass hose fitting to size or lucking out with one that fits onto a copper pipe well enough to be soldered in place. With the circulation off from lack of sun, I was able to drain mostly just the parts of the rooftop plumbing to be replaced. And then I installed the new pipe. Unfortunately, the fitting at the end of that final 45 degree bend wasn't a great match on the existing plumbing, and it was hard to torque it down to the point where all leaking had stopped. As I fuddled with the existing plumbing, suddenly I felt it give as a solder joint failed, something that I've never experienced with a non-leaking joint. This released a momentary gush of antifreeze before I was able to get a funnel under it to collect it. Repairing this failed joint proved more difficult than expected. Though I was able to turn off all the valves on either side of it and drain it completely, for some reason I couldn't get it hot enough even with a steady blast of flaming MAPP gas. There was also the problem that the plumbing demanded to have a slight bend at that fitting, though the fittings themselves on either side of it were designed for a straight run. To deal with this problem, I decided to wrap the joint with an extra layer of copper sheet metal (scrap leftover from my recent water urn project). With this in place, and with copious solder and flux, I was able to make an ugly joint that by some miracle did not leak. By now I was working in complete darkness, holding a flashlight in my mouth. Before I opened up the valves to let fluid flow in, I tested the new joint with an air compressor at 60 psi. Though it's working for now, I think at some point I need to revisit this complex of pipes, valves, and fittings to ensure that it won't come apart in my hands on some future day.
For linking purposes this article's URL is:feedback
previous | next