My wife and I have a 2nd home in Williams, AZ. It’s a pretty basic place, and we don’t even have city water. I have a 250 gallon “tote” that I tow to bring water from Bearizona to our house, where I fill up a large underground cistern. When we first bought the house, it was always something of a mystery as to how much water we had left. We could open up the hatch and look down, but water is clear, you know? It’s hard to see the surface, and besides, we didn’t really know how big the tank was.
Being an Engineer, that was clearly something that I needed to fix. The first thing I did was get an idea as to how much water we could store. I taped a scale to an 8’ piece of PVC, and measured how deep the tank was, and then how much the water level rose each time I added a tote of water (250 gallons). The next thing to do was to design a better water level meter (well, better than a stick). I used the incredibly nifty time-of-flight distance IC, the VL53L0X, and an ESP32 with a color LCD to make a nice graphical display of how much water was in the tank.
It worked perfectly for two years.
Yesterday, I went to take a reading, and it was dead. I unscrewed the board, and brought it home to troubleshoot. This is a total one-off prototype, and I had made it on a protoboard, with wires soldered to act as jumpers. As I was probing around, one of the wires fell off! I started to tug on each jumper wire, and they were all either disconnected or ready to fall off. I looked at the broken end under my microscope, and the soldered ends were all corroded.
Now, that kind of failure could be caused by using water-washable flux and then not washing the board, but I doubt I had made that mistake, and besides, none of the other solder joints were corroded, only the wire jumpers. I replaced all the jumpers with a different type of wire and got the board working.
But, I was really curious as to why the wire corroded in the first place. I had used wires from some “Dupont” cables I had, because they were about the right gauge, and they were nicely colored. I thought that maybe the PVC insulation was causing the corrosion, but as I started to examine the wire, I noticed something really odd.
The strands were not solid copper.
Apparently, one of the ways that they make these things so cheaply is to use copper-clad aluminum wire (also used in Ethernet cables). That must have been why they failed. Although the no-clean flux I used was fine for copper, it caused the aluminum to corrode, and that made the wire fail from the inside out.
So, now I’m pretty sure that my repair won’t fail (or, at least not from this problem).
Excellent troubleshooting!
Yea, Cu-clad aluminum wires are a thing. I have seen it used in applications where they need to minimize weight. I’m not a fan of the stuff.
TIL … that I’m glad I’m no longer an electronic hobbyist. Cu clad Al wires in teeny gauges? That’s Chaotic Evil. Tin- and lead-free solder is already an abomination and now this! Shudders!!
Isn’t any situation where you have two different metals in contact with each other a corrosion risk? Why would anyone ever deliberately make an electrical component this way?
The jumper wires have crimped connectors installed, so that lowers the risk of corrosion. I always end up with ones that I’ve cut for various reasons, so I used those as hookup wire. I would guess that a) most of these are used for very short periods of time - just for test circuits, and b) they are almost never soldered. So, my use case was unusual.
For their intended use, they are probably pretty reliable.
(Don’t get me started on how many “clip leads” I’ve had fail by the wire breaking inside the insulation, right at the alligator clip. This failure is really, really hard to detect, and is very annoying).
This is timely. The handwarmers on my dirtbike are not working. I used 16g speaker wire to hook them up–35w at 12v should be fine. The wire has disintegrated. I can pull it apart with my hands and the strands part easier than the insulation. Now, this is an extreme environment, heat, water, vibration etc. BUT–this is also the speaker wire I used for our whole house, behind real plaster and wood ceilings. I am going to be well beyond pissed if THAT fails. Doesn’t say where it was made, from Amazon of course. Caveat fucking Emptor!
As far as I can tell, aluminum wire is just evil. I think I’ve previously mentioned the time my house nearly burned down because an idiotic appliance installer connected an aluminum house wire to a copper appliance wire using a wire-nut.
My house has aluminum wire. When I first bought it, I paid serious money to have every outlet terminated in copper, using the only UL-approved method (a special crimp connector). Whenever I have construction work done, I have to babysit the contractors, who want to lop off the ends of the Romex when that’s what is easy for them. I forbid them to do that - I make them cut the junction boxes open to remove the terminated ends and then re-run them into new boxes.
Stick will never corrode! Ha Ha! Charade You Are!
Well, went to Amazon the research the speaker wire (10,000 5* reviews). The Amazon description at the top says copper. The manufacturer description helpfully at the bottom of a well of text and pictures say copper-clad aluminum. The dirt bike is a very wet environment from washing, so I’m going to cross my fingers and hope the speaker wire is ok. It would literally cost $75,000 dollars to replace $100 worth of speaker wire.
As an aside. Fuel tankers all have pretty accurate meters to measure deliveries. They also have a wooden dipstick and it’s routine to dip before and after a delivery to confirm the meter reading.
And Hecho in Chine, of course. 14G. Supposed to be good.