MarkII, water-heater slow leak

[billandalita]

I did rip out the Hot water tank the other day and also the 5/8 in. thick wooden shelf that supports the water tank and batteries.  See the attached photo.   Right now I have cut out a marine plywood replacement and will reinstall it Friday with a new KUUMA hot water tank, hopefully with a magnesium anode.  Still not sure what caused the leak or exactly where it is, so I will report on that later when I have more time.  The freezing of water in the bottom sounds like a possibility since that is where I think the leak is, but still, how could ice rupture anything in the bottom of the tank?  Also attached is a photo, (I hope) of the tank's water connections.  Note that the hot water exit is about half blocked by what look to be aluminum oxide deposits.
Bill

[billandalita]

Oops.  I guess I didn't attach this picture of the shelf that holds the hot water tank and batteries.  So here it is hopefully.
Bill

[sail4dale]

Had leakage going down to my board support some time ago.  Fortunately
it was the pressure relief vale on the side of the tank.  It does drain directly down.
a replacement valve cured the problem

[reedbr]

I'll add my vote to the problem being fittings and not the tank itself, especially if your boat is only 10 years old. You can replace most fittings in-place with some cussing and about 4 knuckles, plus maybe $20 of large sockets. Otherwsie, remove the square tank (a geometry puzzle), remove the round aluminum tank from its square galvanized box and inspect everything. I ran a pressure test on my MkII water heater some years back using a hose bib and a garden hose to make sure my new fittings didn't leak. I also cut a notch in the fiberglass locker opening to make reinstallation easier.

[billandalita]

Thanks to all the folks that came up with suggestions for my leaking hot water tank.  I did purchase a new Kuma tank from West Marine, which is close in size and attachments to the Seaward water heater that came with the boat.  I installed an Aqua Pro magnesium anode into the tank's drain port and installed a T-line into the inlet port with a valve to drain the tank when needed.  The magnesium anode is 9.5 inches in length, and obviously cannot be replaced until the hot water tank is removed from its location.  My plan is to change the anode any time the batteries are removed and the water tank can then be fairly easily accessed.  Hopefully the anode will last a few years.  I have been checking the old Seaward water tank in my garage ever since.  Even under city water pressure there was no obvious leak on the fresh water or the engine coolant side.  However there was a constant very slow seepage from the underside of the tank which perhaps left about a teaspoon of water each day in a catch bucket.  There is a very small but noticeable corroded spot on the bottom of the tank which is the apparent source of the leak.  That slow leak stopped this week; perhaps because it corroded itself shut.  Currently I think this very slow but constant leak was the culprit that caused delaminating of the wood shelf holding the tank and batteries.   

[Ron Hill]

Guys : One of the most frequent water leakers (beside the In and Out lets is the pressure relief valve itself. (see the Aug 2013 Mainsheet!!)

Take a short piece of 1/2" hose, an adapter and attach it to the Out end of that PR Valve and let the hose drape down into the bilge.  To check roll up a piece of paper towel and stick it into the end of the hose.  Then check the end of the rolled-up towel and see if it gets wet(need a dry bilge!)!!

My experience is that once a pressure relief valve leaks on an old water heater - just get a new one as the old one will always leak/seep!!

it looks to me that the partial blockage is on the cold inlet and not the hot outlet!!

A few thoughts  

[billandalita]

I think the orientation of the tank in the photo may be a bit confusing.  The photo of the tank is an end view with the base on the right of the photo and the top of the tank is on the left.  Thus the partially blocked opening at the bottom of the photo is the hot water outlet.  The cold water inlet is at the top of the picture and to the right of the pressure relief valve.
Perhaps of interest are two more photos of the tank.  The first one shows the bottom of the tank shortly after it was removed and the fiberglass insulation partially removed.  Note the corrosion evident on the base of the tank, evidently caused by rust from the steel container box where it contacted moisture.  This was the only source of corrosion I could find at the time.  I subsequently applied city water pressure to the hot water side of the tank and then to the engine coolant tubes.  There was no leak from these tests.  I then filled the tank with water and stood it on its coolant end side for two months.  For this entire period as mentioned previously, a very small amount of water collected in an oil pan dish in which the tank set.  The second photo shows the tank turned upside down after the leak test of two months, with a small "pinhole" corroded spot on the left and a water mark running down to the bottom of the tank during the test.  Hope this was interesting:  Bill

[Stu Jackson]

Bill, did that leak exhibit itself as an apparent water pump pressure issue?

[captran]

In my seaward manual it recommends the nylon bushing, which gets brittle and cracks every few years.  I hadn't thought about the brass fitting, but it seems to make sense.  might want to use Teflon tape?  We came back from a short walk checking out the grounds of a new marina when we tied up and water was being pumped out of our normally dry bilge.  looked like it just happened.  glad I bought two 1/2" elbows the last time it went out, but I think I like the brass idea better.  Are they typically the same thread pattern?.

[billandalita]

Bill, did that leak exhibit itself as an apparent water pump pressure issue?

Stu, The water pump was definitely not coming on regularly due to low pressure, except once when I had an obvious hose connection problem.  The leak must have been so slow that any pressure loss was undetectable over any reasonable length of time.  Bill

[captran]

aI the only one that winterizes with the rv antifreeze rather than drain?