Water heater rant

[Mike OMelia]

For the second time since we built this house, one of our water heaters conked out. One of the elements burned out. No biggie. Easy enough to replace. But both times, when I discovered it, the circuit breaker in the panel never blew. Thus, red glowing wires at the element connection. Fine if I'm at the house. But what if I'm not? Why the heck does the circuit breaker not blow?? Do I have improper breakers in my panel? Are there more sensitive ones?

Mike

[Matt Marsh]

Mike, When an element burns out, it most often opens completely, therefore no current would flow through it. The only case where the breaker would trip is if the element shorts or grounds out, which would in turn cause a high current flow. If your connection was glowing red, that means that you have a loose (high resistance) connection, but there was still flow through the element. The higher the resistance, the lower the current flow. (Current = Voltage divided by resistance.) The breaker actually sees lower current in these cases, therefore it will not trip. I'd recommend really cleaning those connections when you repair everything, and be sure that the terminal screws are tightened securely. If the wires were that hot, you should probably replace it. At least the sections that are burned.

[Mike OMelia]

I indeed did all of those things. Cut back to good copper, sanded the wire, etc. The connections were 7-8 years old. The elements were the bottom ones (most likely to go do to scaling). I just don't get why once they die, that happens. This occurred on two separate water heaters. Rheems. I guess the connections oxidized over time.

Mike

[Matt Marsh]

Environmental factors could definitely play a part. Elelectrical contact/anti-corrosion paste will help. Use it at all of the terminations, inside, under, and outside. Be careful not to be too aggressive with the sandpaper, scratches will result in less contact area at the junction, contributing to the high resistance issue. If the copper is good enough to use, you really shouldn't have to sand it. Fine steel wool or a scotchbrite pad would be a better choice. Most factory terminations in water heaters involve merely stripping back a short, straight section of the wire, then just binding it between the terminal screw and insulating barrier. Stripping the wire back a little further and actually bending it around the terminal screw in a clockwise direction makes for a better termination. Maybe the best solution would be to completely replace the internal wiring in the water heater.

[Myk Rian]

Stripping the wire back a little further and actually bending it around the terminal screw in a clockwise direction makes for a better termination.
I believe a wire lug is preferable.

[Matt Marsh]

I believe a wire lug is preferable.

If you mean a crimp style lug, it would indeed be preferable if the wire were stranded. But most water heaters are factory wired using solid wire. A crimp style lug on solid wire is trouble waiting to happen.

[David Weaver]

I would also rather have a fully soldered connection with a high current draw. No reason to create a bottleneck and generate heat a specific point.

[Matt Marsh]

I would also rather have a fully soldered connection with a high current draw. No reason to create a bottleneck and generate heat a specific point.

Boy David, soldered connections on a heating appliance is not normally a good plan.

[David Weaver]

Good point. To the extent you could, then, I guess replacement of the entire length of wire?

[Matt Marsh]

Good point. To the extent you could, then, I guess replacement of the entire length of wire?

I think its definitely the best option. If the terminal screws on the elements are iffy, I'd replace those too (entire element, not just the screws).

[Mike OMelia]

I replaced the entire element, so the screws went along with the deal. I found enough good wire to use... I do not fully understand the failure mechanism. For the lead to get glowing red hot, the entire circuit must be drawing more current than it normally does. I would have thought that would trip the breaker... but it does not. When I get home, I will check the breaker rating. Both elements in the heater are 4500 watts (that could be peak). I still have the element (old one).

[Tom Stenzel]

I replaced the entire element, so the screws went along with the deal. I found enough good wire to use... I do not fully understand the failure mechanism. For the lead to get glowing red hot, the entire circuit must be drawing more current than it normally does. I would have thought that would trip the breaker... but it does not. When I get home, I will check the breaker rating. Both elements in the heater are 4500 watts (that could be peak). I still have the element (old one).

The heating element has resistance over its entire length so the heating is uniform.

If you have a bad connection at the terminal even a few ohms can create a lot of heat since all of the heat is concentrated in a small area. That is, the space of the wire under the screw.

Did you check the old heating element with an ohmmeter to see if it went open?

-Tom Stenzel

[Lee Schierer]

If you didn't suck out the accumulated calcium under that element you will be replacing it again in sort order. I use a piece of curved 3/4" copper pipe attached to a wet dry vac to suck out the calcium chips in the bottom of the tank. We haven't had to do this since we installed our water softener more than 15 years ago.

[Matt Marsh]

For the lead to get glowing red hot, the entire circuit must be drawing more current than it normally does. I would have thought that would trip the breaker... but it does not.

Mike, As I tried to explain in an earlier post, it is actually the opposite.

Ohm's Law says: I=E/R (amps=volts divided by resistance). The water heater only uses one element at a time, so to figure the specifics for the circuit, we can use the known constants. 4500 watts, and 240 volts. Element current draw can be calculated by dividing the watts by the voltage (I=P/E). 4500/240=18.75 Amps. Element resistance is 240volts/18.75amps=12.8 Ohms. Inversely, 240volts/12.8ohms=18.75 Amps.

As you can see, any increase in resistance, actually decreases the amount of current. If we add even 5 ohms as an example, the new total circuit current is 240volts/17.8ohms=13.5Amps. Because water heaters are considered continuous duty, code requires us to size the circuit to 125% of full load. Most residential water heaters use 4500 watt elements as you stated, so 125% of 18.75 amps = 23.44 Amps. Most are fed with #10 awg copper protected with a 30 amp circuit breaker.

Maybe it might help to envision an arc welder. If you were to measure circuit current during the welding process you would see the highest current flow as you first strike an arc. If your strike is successful and you maintain an arc, current flow drops. But if your rod sticks instead of arcs, current flow remains higher. At the point where the most fire and melting is occuring, current flow is less. The reason is because the maintained arc is adding resistance to the circuit, just as the bad connection in your water heater did.