Sunday, December 21, 2008

Plumb Wrong

This week I solved a plumbing problem and it might have a lesson in troubleshooting that could apply to radio astronomy (electronics). I just installed a water system in my daughter's cabin. I have done small plumbing projects many times over the years but I think this was the first time I installed an entire supply and waste water system. Here in our part of Hawaii there is no municipal water system. You catch the water off of your roof and hold it in a catchment tank then pump it when you need it with a shallow well pump. My daughters cabin has an outdoor kitchen (common here) and about 30 feet away is the main cabin with a full bathroom.

When I fired up the pump (it is right behind the kitchen) the water flowed from the kitchen faucet just fine. I went into the cabin and turned on the water in tub.. a tiny trickle. Then I tried the bathroom sink, another trickle. Finally I opened the valve to the toilet and again a tiny amount of water began flowing into the tank. What? Something must be clogged in the pipe running from the pump to the cabin, I thought. But all of the pipe is brand new. The kitchen faucet was fine so it would have to be after the point where the kitchen sink tapped off of the supply line. It was dark so I went home and tried to sleep on it.

Next day, I went to the hardware store and bought a bunch of couplers to reconnect the supply line in all of the places I intended to cut it in hopes of finding the clog. I began chopping into the lines close to the pump and stepping towards the cabin, but everywhere I cut there was plenty of water pressure, even right under the fixtures! Could I have three bad fixtures, a tub faucet, a sink faucet, and a toilet inlet valve? Yes!

When I disassembled the valves I found each one clogged with little bits of PVC pipe and blue "Rain & Shine" PVC glue. I had created the problem in all three fixtures by using an electric trim saw to cut the pipe (which produced the shavings) and too much glue when putting them together. As I painstakingly unclogged each fixture with a bent bobby pin, I pondered how this related to many of the more horrific electronics problems I troubleshooted (shouldn't there be a word 'troubleshot'?) over the years.

We usually assume that ONE thing is wrong in a failed system, but you don't have to spend too many years fixing things to find out that multiple component failures are common. The electronic system outputs are our usual indications that something is wrong. No lights at all or all dim lights (trickles of light..) would immediately signal power supply problem to me, but that conclusion can be very wrong. Each of the "lights" could also be failing. Another point; in a system that you design and/or build from scratch your technique in assembly could lead to failures in multiple places, cold solder joints for example.

No comments:

Post a Comment

RF Noise Source NF-1000 SZ evaluation

  I have been working on an automated step calibrator project in recent months, so I was excited when I saw the device shown above on eBay. ...