Let's say the Hot Water Heater tank leaks.

The tank itself, not the typical T&P valve drip. Not the drain valve.

Let's assume tank failure is from corrosion under sediment build up.

The leak wells up in the drain pan, and the sensor in the drain pan triggers the automatic solenoid actuated cold water supply line valve to close. Voila, no more water supply to the hot water heater.

But the tank is still leaking, and since the tank failure is at the bottom of the tank, the tank will continue to leak down and out of the tank by way of gravity.

As the tank empties.... what happens to the burner?

Does it stay lit?

Does it shut off?

Let's say the water heater gas control thermostat is set to 140 degrees to ward off bacteria in the tank, with 120 degree tempering valves at shower and sink points of use.

Presumably, without any new supply of fresh water entering into the tank, once the temperature of the water already in the tank reaches the 140 degree set point, the burner should shut off, right?

Now, as the volume of water in the tank decreases, due to the leak, the less burner time is needed to maintain the remaining water left in the tank. As the water level continues to diminish, so does the amount of burner on time to maintain the 140 degree set point.

However, eventually all the water will drain away. Is there any mechanism in the gas control valve of the heater itself that would prevent the burner from firing up again and "cooking" the steel tank itself, and the wet sediment inside of it, once that steel and sediment cools below 120 degrees?

Let's say the gas control valve is a Piezo ignition type, not a standing pilot. One of those fancy new style Honeywell gas controllers that has a flashing LED status light, controlling a FVIR burner assembly.

Here's the problem I'm trying to solve... most automatic leak detection water shut off valves only shut off the water. Very few of these auto water shut off valves also have an add on feature that shuts off the water heater burner. And of those very few that do, most of those involve an intercept to the thermopile of a standing pilot. I only found one that can intercept a spark ignited pilot (that plumbingsupply.com sells, made by Watts), but that was for an oil fired water heater, not a natural gas unit (not sure if that makes a difference, but still waiting on manufacturer Watts to call me back to answer if oil or gas matters for that unit).

In the presence of so many automatic shut off valves that shut off the water only, and in the absence of an equal number of automatic shut off valves that can also shut off a gas burner, it occurred to me to ask...

Do water heaters already have a safety mechanism to shut their burners down when the water is drained out of them?

Yet, the instructions always say to BE SURE TO FILL THE WATER HEATER WITH WATER FIRST BEFORE ATTEMPTING TO LIGHT THE PILOT/BURNER ASSEMBLY. With instructions in all caps like that, it makes me think no... if the water heater leaks down all the water inside, then serious damage could occur, as the burner has no means to determine there is no water in the tank. And that makes me want to get the kind of automatic leak detecting water supply shut off valve that also provisions a means and a method to simultaneously shut down the burner.

On the other hand, with today's post 2015 sophisticated Ultra Low NOX, FVIR, Piezo ignited, LDO sensored, thermostatic reactive device equipped gas control valves, I wasn't sure if a gas control device in the event of a water tank leak was being redundant, and if that redundancy could in any way explain the lack of such devices on the market, even while I can find at least a dozen different water shut off only devices.

So, what happens when a natural gas hot water heater tank drains down with no fresh water supply to refill it?

thats a good question.
in all my years i have never seen that happen.

If the tank is corroded at the bottom with a leak, the cold water supply is turned off, the hot side fixtures are not turned on, and the T&P valve in the closed position, would it not stop leaking because there is no air or water entering the space up top to replace the volume lost to the leak?

MA code requires us to install a vacuum breaker in a tee on the cold supply pipe.
this ensures that a vacuum can never form and collapse the tank under atmospheric pressure.
strictly following MA codes guarantees it's residents will live 35 seconds longer than the national average.
watts vacuum breaker...

And yes, I have those vacuum relief valves, so yes, the finger on the end of the soda straw anti drain down physics would not apply. The tank will drain.

Very slim pickings out there on automatic water valve shut offs that also shut off spark or hot surface ignited gas valves. Of the few that exist, they interrupt the thermocouple of standing pilot gas heaters. The one spark ignition interrupt device that plumbingsupply.com sells has instructions that specifically state that it will not work with the Bradford White Icon (Honeywell) Piezo ignition gas controller.

I'm currently noodling on solutions outside the box to accomplish automatic shut down of the water heater, but would rather not reinvent the wheel, especially on listed devices, so my original question still stands... what, if any, safety protections are built into today's water heaters that will shut them down automatically in the event that the tank drains dry?

Assume the building has no occupants to observe, report, or handle the problem. The goal is safe and automatic shut down with no human intervention.



Edited 1 times.

what do power vented water heaters do to stop the burner if the fan is not running ?
I know there is a tube that senses a draft but what does it connect to?
there must be some way of incorporating that shut off device into the leak detector ?

The one you found that can cut off the current generated by a thermopile, can be wired in to cut off the pressure switch signal or an overheat limit switch in a power vented gas water heater.

Unless the leak is at the very bottom of the tank, or below the thermostat, ALL the water will NOT "drain away". But, if it did the burner would ignite and heat the interior of the tank until the thermostat shut it off, or the ECO activated and shut the valve down.

[www.ebay.com]
Install this valve and that eliminates the possibility of the gas burner firing.

[www.taco-hvac.com]

==============================================

"Measure Twice & Cut Once" - Retired U.A. Local 1 & 638

A power vent water heater, and likewise a flue dampered water heater, both of which require 110vac to operate, have a built in means to shut off the gas controller in the event of a power failure. I don't know exactly what means it uses... but I did verify with AO Smith that no electricity means no gas burner firing on the various power flue damper and power vent models that AO Smith manufacturers (which also includes State, American, Reliance, and Lochnivar brands). Given that some Bradford White, Rheem, and Ruud water heaters use the same Honeywell WV4460 family of controllers, it would appear that those brands of water heaters with power venting or power flue dampers will also shut off the gas burner in the event of a loss of power.

Which means that an automated water shut off valve that also outputs a trigger voltage can be wired to a continuous duty high voltage relay that could be wired to an externally boxed outlet, and the water heater blower or damper motor could be plugged into that relay controlled outlet. But that steps into the electrician's domain.

It seems to me that the gas controller, or at least the spark igniter, could be interrupted and produce the same result, and would only involve low voltage signalling modification, which would not involve wiring external boxes carrying house current. At this point, I think the unit that plumbingsupply.com sells for an oil fired heater would work, but am still stumbling on the warning that says it will not work on the BW Icon models, which use the same Honeywell controller as all the other PV and FD models mentioned above.

So I'm still trying to figure this out, and still wondering if all this figuring is even necessary.

I hate to cut into the factory wiring of water heaters, which I'm sure would void the warranty. I'd prefer a commercial plug and play type of solution, that can be removed without evidence of molestation.

That being said, the greater good is safety over warranty, and I want to protect the building from potential flood damage, and I pipe in an automated valve to shut off the water when a leak is detected, the fuel or ignition to the water heater burner needs to be shut off as well.

So as much as I would hate to take on the entire responsibility for the manufacturing quality and longevity of the water heater control systems once/if I modified it, I'd still consider it, but would prefer a clean way to do it.

I'm working with power flue damper heaters though, not power vent.

"Unless the leak is at the very bottom of the tank, or below the thermostat, ALL the water will NOT "drain away". But, if it did the burner would ignite and heat the interior of the tank until the thermostat shut it off, or the ECO activated and shut the valve down."

Let's assume the leak is at the very bottom of the tank, below the thermostat.

And, because the leak is below the thermostat that pokes through the tank from the back side of the gas control valve, the probe of that thermostat is now hovering in still air, rather being than immersed/submerged in water.

If air is a less efficient medium for transferring heat than water, how much longer would the burner burn before the probe of the thermostat were able to measure enough heat in the air to turn off the gas burner?

Now let's assume that not all the water drained away, but that there is some water still in the tank, and a lot of space in the tank to conveniently fit STEAM.

If steam builds up in the tank, is that not a well of explosive force that would overwhelm an already compromised tank from the leak?

Where is the temperature of the Emergency Cut Off measured? In the tank? Or at the burner assembly? If at the burner assembly, where combustion temperatures can be 300F, and water turns to steam at 212F, and the thermal probe of the gas controller valve is measuring air instead of water, but there is still a low level of water remaining in the tank to boil into steam, could that present a problem?

Thanks for the idea on the WagsValve, but unfortunately, the Wagsvalve is a one time only device that has no ability to reset. Once it activates, the entire valve has to be tossed, and a new one plumbed in.

Because the Wagsvalve is hard plumbed into position at the bottom of the drain pan, it cannot be moved out of the way while maintaining the water heater, such as when draining it annually. Since it is inevitable that some spillage of water might occur during the normal course of attaching or removing a garden hose to drain, or during the course of inserting a wire into a full flow ball valve drain port to loosen sediment in the bottom of the tank in between flushing or draining, there is additional risk of accidentally triggering the Wagsvalve.

Once triggered, what should have been a simple maintenance routine now turns into a full blown plumbing job, requiring the removal of pipes and the fabrication of a bypass cross pipe or some other means to restore supply water to the hot water heater until the Wagsvalve can be acquired and replaced. The Wagsvalve isn't generally available locally, and requires a wait for shipping to arrive. The Wagsvalve also costs $150 each, and while that expense is tolerable for an initial set up, it could border upon ridiculous when having to repurchase an entire valve assembly each time it was inadvertently triggered.

Finally, because it is a one shot only device, there is no way to even test the device to see if it actually works at all, and therefore it relies on as much hope and a prayer that works as not having any device at all, while hoping and praying that nothing happens.

Therefore, I can't see a Wagsvalve being useful. It is designed for standing pilot water heaters that do not use electricity. Since the high efficiency water heaters I am installing already require electricity, there is no benefit that the Wagsvalve can offer, when compared to the other automatic shut off valves that are easily testable and resettable.

I don't think Taco manufactures, sells, or even supports the Wagsvalve anymore. Taco now offers the electronically controlled "LeakBreaker" automatic shut off valve ( [www.tacocomfort.com] ) that they specifically and prominently point out is "resettable" and "testable."

It seems as if Taco separated itself from the Wagsvalve, and now a limited liability company calling themselves "Aquaguard, LLC", also in Cranston RI, is now the owner/seller/distributor of the Wagsvalve. Aquaguard appears to be a one man operation, hard to tell, but it seems to be. Perhaps that man is the original inventor, who at one time convinced Taco to distribute and support his invention, but who now is back on his own. Just conjecture here.

I've put in calls to both companies, and was only able to reach voicemails, which have not been returned.

Notwithstanding who supports or markets the Wagsvalve, for reasons further explained in an earlier post above, I do not believe that the Wagsvalve is an adequate solution

The ECO is part of the thermostat and activates at about 190 degrees, well below the temperature of steam. Your oven is full of air but it still gets hot enough to bake whatever you put in it.

I located information from Honeywell on the specific gas valve for my application.

There is an ECO, and a separately referenced TCO.

The ECO activates at 199F. Still reading to find the temp level that activates the thermal cut out.