This "6 year" hot water heater failed after 3.5 years - is it bad luck (eg normal for the $429 box store 40 gal DHW heaters and how they make money off what I expect is a pro-rated warranty?), poor water and/or pressure environment for the heater, or possibly an installation issue?

The failure is a pinhole leak about 1/2 way up the tank (I could hear a hissing when the tank has water pressure.) After turning off the cold water supply, hissing stopped, but upon turning back on the leak became worse, even when supply was turned off, requiring immediate draining of the tank

Some facts about the water and system:

  • Well-pump with cold water pressure tank - which takes care of thermal expansion. However, the pressure switch is likely operating in the 20-40psi range. Could this pressure cycling, 100-500 times per day, weaken the tank? (100 times per day is probably normal, but in the summer could approach 500 times per day due to use of garden hoses.)

  • The well water is rather acidic, but a solution pump raises the pH to the low 6s. The water is naturally soft enough, with no other conditioning.

  • Never observed any excessive discoloration of the hot water.

  • The shower is over the heater, and on occasion, there is some air in the lines - don't recall if it was hot or cold. Interestingly, this hasn't been happening for the past ~1 year.

  • A sand filter was NOT used water supply for most of the duration. Possibly a fair amount of sand went into the tank - based on the amount of sand in the tank of a nearby toilet.

  • The previous two hot-water tanks had also been replaced prior to warranty expiration (under the previous owner), but the last one failed (3.5 years) due to internal gas leak instead of water. (I prefer the internal water leak failure than as gas leak on New Years Eve!)

Facts about the installation:

Connections : Flue, plugged into grounded AC outlet, gas supply, hot and cold dielectric w/flexible water pipes, tanks metal feet rest directly on the concrete floor

  • The water pipe connections are via dielectric fittings at the tank, with ~18&24" flexible hot watcher hose kit to the copper pipes. I'll put photos up.

  • Cold water pipe has nearby electrical earth ground connection, but the tank itself has no direct connection (except possibly the flexible hoses and AC plug ground)

  • After 3.5 years there was some corrosion on the hot water side, and the fittings can't be removed from the tank.

  • Previous tank was ~4' on resting on concrete blocks so I replaced with a tall (~5') and modified the plumbing using the flexible hoses instead of direct copper connection

  • Never checked the anode rod, maybe I should?

With 50 degree well water in the summer, everyone would be happy if the next one lasted longer.

  • The most likely cause is a defective weld seam. I am expecting if you remove insulation and look at the tank surface you will find the leak at a long ( axial) weld seam. It held for a time until corrosion penetrated a thin point in the weld. Jun 29, 2019 at 15:36
  • I am curious if there is electrical current passing through the tank somehow. Can/would you probe for potential applied voltages? Your description reminds me of a situation I once found for an electric tank where one of the heating elements had failed subtly. Had similar symptoms. Probe between the ground and different metallic points like the gas line, hot water outlet, cold inlet, tank exterior, etc.
    – wallyk
    Jun 29, 2019 at 15:52

1 Answer 1


Given that the last two also failed in the ~3 year timespan, I'd say either

  • you and the last guy have a strong affinity for bottom-shelf big-box cheapies
  • there is a quality problem with your water
  • there is a quality problem with your fuel

Assuming the pinhole developed in a non-combustion area, that would point to the water. I would have it tested (which water treatment sellers will often do for free).

I would also have a close look at the anode. If it is relatively intact, then changing the anode won't help. If it is massively corroded, then you'll need to make "changing the anode" an annual affair!

  • Different failure mechanisms on the last 2. Did I leave something out for the water situation? Fuel is from the gas company. Yes buy cheap unless there is some evidence that spending more on the same function provides increased value. Is there some? The new unit has a 9 year - predictions?
    – pathfinder
    Jun 30, 2019 at 22:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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