0

I've got a dual-element, dual-thermostat electric water heater. One of the thermostats is stuck in the "on" position: it heats the water until the high limit switch trips and cuts off all power. How do I figure out which thermostat is the problem, or should I just replace both?

If it matters, the heater is a Kenmore 153.329460 with the lower element converted for 5500W operation. The power supply is 240-volt split-phase (so, no neutral wire; both incoming wires are hot).

  • Do you have a volt meter? Or even a non-contact voltage detectorr (aka hotstick)? It should be pretty easy to determine which element is still heating and which is off. – Tyson Feb 7 '18 at 13:24
  • I've got a multimeter and an NCV detector. Everything except the grounding wire registers as "hot" on the NCV (and I believe everything would even with the thermostats functioning properly), and I don't know what to check with the multimeter. – Mark Feb 7 '18 at 18:56
0

The typical dual-element, dual-thermostat water heater in the United States is a "non-simultaneous" design: first the upper element heats the top of the tank to the set temperature, then the upper thermostat switches power to the lower thermostat, which uses the lower element to heat the rest of the tank. There's never more than one element active at a time.

Upper thermostat with labeled screw terminals.  T1 is the upper-right screw, T2 is the upper-left, T5 is the third terminal from the top on the left, T6 is the bottom-left terminal, T7 is the bottom-right terminal schematic

The picture on the left is the upper thermostat for a typical water heater; the schematic on the right shows both the upper and lower thermostats. The switches are in the positions you'd find them on a hot tank with both thermostats working. Note that point "T6" in the schematic is on the right and T7 is on the left, while on most actual thermostats, it's the other way around. T6 is the terminal with a single wire connected to the upper element, while T4 connects to both elements and T7 connects to the lower thermostat.

Testing with a multimeter

If you've got a multimeter, it's easy; you don't need to disconnect any wires or anything. Let the tank heat up for a while until the water is well above the thermostat set point, then shut off power to it. If the upper thermostat is the one that's stuck, you'll have a pretty steep temperature gradient in the tank, so let the tank sit for an hour or so.

You may wish to switch both thermostats to their minimum setting. This will make it more likely that the tank is hot enough that both should be switched off, but it also carries the risk of opening the stuck contact, leaving you with a questionable thermostat and no way to identify it.

Set your multimeter for the highest AC voltage range, and measure the voltage between points T1 and T2, between T1 and the tank body, and T2 and the tank body. All three measurements should be less than one volt; if any is higher than that, power to the water heater is still live, and you should call an electrician to investigate that before you proceed with figuring out which thermostat is broken.

Now that you've verified that power is off, set your multimeter to the lowest "ohms" or "resistance" setting that will cover the 0 to 50 ohm range. Measure the resistance between points T5 and T6 (if your upper thermostat doesn't have a terminal in the T3 or T5 position, make sure the limit switch has been reset, and measure between T1 and T6). If you've set everything up right and the tank is above the thermostat set points, if the multimeter reads:

  • open circuit: Both thermostats working
  • less than 1 ohm: Upper or both thermostats stuck closed
  • 20-30 ohms or so: Lower thermostat stuck closed

If the upper thermostat is stuck, you'll need to make a second measurement to see if the lower thermostat is broken as well. Measure the resistance across the lower thermostat's terminals. If it's less than 1 ohm, the lower thermostat is stuck closed or you didn't heat the tank up enough; if it's an open circuit, the thermostat is working.

Testing with a voltmeter

If you've got a voltmeter, you can determine which thermostat is broken, but it involves working with live wires.

Heat the tank as described in the "multimeter" section, but after letting the tank sit, turn the power back on. Set your voltmeter to the highest AC voltage range. Measure the voltage across the terminals of the lower thermostat, across T5 and T6, and across T5 and T7. These measurements should be near either 0V or line voltage:

  • Lower thermostat: line voltage, T5-T6: line voltage, T5-T7: 0V: Both thermostats working
  • Lower thermostat: 0V, T5-T6: 0V, T5-T7: 0V: Upper thermostat stuck closed.
  • Lower thermostat: 0V, T5-T6: line voltage, T5-T7: 0V: Lower thermostat stuck closed.
  • Lower thermostat: 0V, T5-T6: 0V, T5-T7: line voltage: Both thermostats stuck closed.

Testing with a non-contact voltage tester

If all you've got is a non-contact voltage tester, things are harder: the results are less clear-cut, and it requires working around live wires.

Heat the tank as described in the "multimeter" section, but after letting the tank sit, turn the power back on. Check the wire connecting the upper and lower thermostats (the "runner") with the NCV. If it doesn't show a voltage, the upper thermostat is stuck. If it does, proceed with the next step.

Shut off power to the water heater, disconnect the runner from the upper thermostat, cover the exposed end with something insulating (a wire nut or electrical tape), and turn power back on. Check the runner again with the NCV. If it doesn't show a voltage, both thermostats are working. If it does, then the lower thermostat is stuck (or the NCV is picking up induced voltage -- I said this was less clear-cut than using a multimeter).

  • it's even easier than that, just measure voltage across the thermostat. If the thermostat is closed (call for heat) voltage will read near zero. If open (successfully shut off) voltage will read as line voltage. A stuck thermostat will never read line voltage. – Harper Feb 8 '18 at 9:00
  • @Harper, if the upper thermostat is stuck closed, the lower thermostat will read 0V because the "runner" wire isn't connected to anything. A reading of line voltage across the thermostat's terminals means the thermostat isn't stuck closed, but a reading of 0V doesn't mean that it is. – Mark Feb 8 '18 at 9:45
  • Then you would have to authenticate the upper thermostat via the T5/T6 leg first. – Harper Feb 8 '18 at 9:55
-1

You have a burned out element. It is shorted to ground and continues to run away. Probably the lower element is bad. Replace it. Clean out the lime deposits and you are golden.

  • What information in my question tells you that it's a burned-out element and not a stuck thermostat? – Mark Feb 8 '18 at 6:20
  • See answer below. – Paul Logan Feb 9 '18 at 17:42
  • Replacing the lower thermostat fixed the tank's overheating problem. Seems to me that's a pretty good indication that it was a stuck thermostat and not a shorted element. – Mark Feb 10 '18 at 5:29
-1

The thermostats on the modern day tank style water heater is extremely reliable, #1. #2 I've just seen this situation enough time to know. Here's what happens: The elements are direct immersion. When the water reaches the temp of around 130* the minerals separate and fall to the bottom of the tank. They build up. the lower element become encrusted and covered. It can't dissipate heat. It overheats the the sheath splits. The thermostats break only one leg of the 220-volts. The lower element always has 110-volts available. It goes to ground. Nothing to shut this off but the tank safety. Part of that element heats as long as there is power to the tank.

-2

Try excluding the elements one-by-one and test (setting the temp low, ex: 50°C) which one is always on (see if the temp reaches 80°C), that's the guilty one.

5.5 kw electric water heater... that's a crazy waste of energy! if it's failed suggest you to replace it with a heat pump one (it'll cut the energy usage by 66%)

  • It's pretty clear you don't understand thermodynamics (a 1-watt and a 1-megawatt electric heater both use the same amount of power; the megawatt heater just spends less time heating and more time at the set temperature), so I don't know how much I'd trust your suggested testing procedure. And spending hundreds or thousands of dollars to replace a failed $10 part doesn't strike me as a good idea. – Mark Feb 7 '18 at 20:14
  • @Mark Aside from the off-topic suggestion, DDS is correct about the thermodynamics. A heat pump or AC moves heat: If we assume the coefficient of performance of the unit is 3, a heat pump water heater will move 3 kilowatts into the water with an electrical consumption of 1 kilowatt. – user71659 Feb 7 '18 at 20:43
  • @Mark, at least here in Italy, where a kWh costs at least 0,20€/kWh (1/4 USD), thermodynamics is well known, I can assure you that a 300W heat pump will heat the same amount of water of a 1,2 kW boiler, for 1/3 the consumption. Here people, at least in newer houses, are dropping natural gas for space heating because heat pump heat is more economical. (1kWh of gas may cost about 0,08€, with boiler efficiency of 107% relative to LHV for a condensing boiler). Also we have limiters on electric supply (usually 3,3kW) so a less-consuming device let us use more appliance at once. – DDS Feb 7 '18 at 21:00
  • What about the noise level of the heat pump water heater? The heat pump water heater pumps heat out of its immediate surroundings and into the water in the tank so it cannot be in an enclosed space or at least there must be allowance for heat from somewhere to flow into the space. What are the installation parameters for these heat pumps? – Jim Stewart Feb 7 '18 at 21:23
  • I don't know why people are surprised. New non-heat pump electric heaters >55 gal were banned in the US in 2015. – user71659 Feb 7 '18 at 21:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.