Hot Water Tank Failing Again
I've been regularly eating HW heaters since the water softener was installed (every 3.5 years).
After having 2 tanks fail I bought a powered anode (Corro-Protec). At 3.5 years the hot water was brownish. Corro said it could be Manganese in the water causing reaction. I flushed tank and installed standard anode for the time being. But it is starting to look like the tank is rusting so planning to replace (going from Smith to Rheem). Yeah, could be 3rd tank was just defective, but although the green light was always on, I'm doubtful the powered anode did the job (it was covered with a fine brownish sediment). The powered anode tank lasted no longer than a standard anode tank.
So, current thinking is:
- Get new tank - 40 gal. gas fired.
- Flush 1 time per year.
- Monitor anode once or twice per year, run Mg anode since I have softened water. I've got ~32" of space above tank to swap anodes.
- Add valve at water softener (Clack) to bleed in some non-softened water to the system to say ~4 grains of hardness.
Any other thoughts besides ditch water softener.
We are on city water, but they mix in some aquifer water, so hardness varies today from ~8 to 12 grains. Used to be 25 grains!
Thanks, Mike
Comments
-
If your city water is that soft... you don't need or want an additional water softener. Really soft water is incredibly corrosive.Br. Jamie, osb
Building superintendent/caretaker, 7200 sq. ft. historic house museum with dependencies in New England
1 -
Hi, Some thoughts. I like the idea of a bypass around the softener. The amount of hardness left in the water should be between 3 and 6 grains (60 to 120 ppm). Get in touch with Corro-Protec and see if they have any thoughts for you on the function of their anode. I've seen anodes down to a bare wire in six months in over-softened water, so you might check in three months just to get an idea how fast the anode is being used up. You want to replace the rod when six inches of the core wire is visible. Also, they make flex anodes, so the overhead clearance should not be a problem. You do want a full length anode that extends down close to the bottom of the tank. With 32" of clearance overhead, you might be cutting the anodes too short, which would be good reason to go with flex rods.
Yours, Larry1 -
Jamie, Larry,
Thanks for taking time to read this.
Additional info, I have a whole house sediment filter (4 x 10) feeding a whole house charcoal filter (4 x 10) which feeds the softener. Softener works well as I rebuilt the head some months back.
I'm game to put the softener in bypass and see how it goes, but another member of the house will complain so I'll try and dial it back.
I've talked with Corro-Protec several times, they agreed something is off here and suggested the Managese content. I looked up our water report and Mn was well below minimum.
32" is the standard Smith anode (Zn Al or just Al), but yep, no reason to not go with a longer flexible Mg anode.
And did not know this: You want to replace the rod when six inches of the core wire is visible.
The original tank installed when we bought the house lasted 16 years w/o a softener (not statistical data, but worth noting).
Looks like my options are:
- ditch softener or
- regular flushing and monitoring of the anode (and keeping a spare anode on the shelf).
I like to learn things and I've gotten good at swapping out tanks, but would like to move on to other things, lol.
Mike
0 -
Or a stainless tank if it is electric.
Was the powered anode protecting the tank but you just had some sort of precipitate?
0 -
Hi, I'm averse to having any aluminum rods in a tank, and replace them with magnesium even in new tanks. Is the tank electric or gas? There are different things to check when looking for potential leakage sites. Also, dissimilar metal connections can be a problem, particularly in softened water. Softened water can also be hard on stainless. A bypass around the softener will help make the problems smaller.
Yours, Larry
0 -
Guys,
We are talking 40 gal gas fired AO Smith tank nothing fancy here.
Mattmia2, I thought it was some sort of precipitate, flushed tank last week. Yesterday, drained a bit out and saw some discoloration, maybe this is nothing as it cleared after a few seconds. Not really sure what the powered anode did, but it's out of the tank right now.
Larry, I will get a 48" flexible Mg anode and install, but fear I'm headed for another new tank.
My plan is to add the valve to have some unsoftened water in, tired of replacing HW tanks.
Mike
0 -
Copper is brown in certain oxidation states, maybe it is your pipes dissolving in the over softened water and ending up in the tank. Someone could always do qualitative analysis on it to figure out what it is.
0 -
I don't profess to be a copper expert, so will research this a bit and see what i can learn.
A water analysis makes sense if I could find a place with reasonable cost to do it.
Mike
0 -
Hi, What signs is the tank giving you suggesting that it's about to fail? I have some experience reading the tea leaves tanks give 😏
Yours, Larry
Ps. Pictures would be helpful 😉
0 -
Larry,
I flushed the tank last week, after removing the powered anode. Installed original anode as it was never really used.
Yesterday, I drained some water from the tank via a hose attached to the port and again saw some brownish water. It cleared pretty quick, nothing like it was. Before removing powered anode and flushing the tub or sink upstairs would fill with brown water, now its pretty clear. The tank is not leaking and we have plenty of hot water.
Would you want a photo showing the brownish water coming out of the hose or do you have something else in mind. I generally let the hose drain into a whitish utility sink in the basement.
Thanks, Mike
0 -
For info a couple of photos of the sediment on the powered anode and my finger.
0 -
city water or a private well?
Your local health department may do some testing A lab at a close by ag school?
Plenty of places you can mail a sample.
The cost will depend on what type of analysis you want done. Sounds like you want to know what minerals are in your water, not hazaderous chemicals?
Turbidity is the fine silt in water supply, basically silica, it is common in spring time with high run off water
Bob "hot rod" Rohr
trainer for Caleffi NA
Living the hydronic dream0 -
Hi and thanks for the photos. What would be most useful to see are the inside of the combustion chamber and plumbing hookups on top of the tank. Also, if any rust is showing around the base of the heater, it would be telling. Is there any steel plumbing in the supply to the tank? Thanks 😊
Yours, Larry
Ps. Getting into the combustion chamber might be too much trouble. If so, just look through the sight glass and look for any damp rust.
0 -
Larry, again thanks for reading thru all of this.
No steel in any plumbing, house built in 88, never had this issue before. Usually the tank just starts leaking. Cold water is fine.
No rust anywhere on tank, its like its new. Site glass very difficult to see thru, I waited for gas to come on, figuring flame might illuminate inside a bit. Best I can tell no rust, moisture or anything in there. No obvious way noted to remove site glass. Some photos for info. Mike
Thinking its time to ditch softener as long as county keeps providing city water.
0 -
Hi, I wouldn't just replace the tank at this point. From the outside it looks okay. If you installed a flex connector on the hot side, you could easily add a second anode in the hot port. This would give the tank better protection and let you continue to soften water in order to maintain domestic tranquility. The best way to know the condition of the inside of a tank is to "read" the anode. You don't have a sacrificial anode to read yet. At this point, I'd mix in unsoftened water, add a second anode and then check one of the anodes in 4-6 months. That will let you know what's happening. Of course share photos of the anode and we can help translate what it's telling you 🤠
Yours, Larry
0 -
Larry, thanks for the help here.
I plan on swapping the Al anode for a Mg longer anode. I'll probably ride with that, instill the mixing valve and see what hardness I end up with.
Not sure how installing a flex line on the hot will let me add another anode?
Mike
0 -
Hi, About adding an anode in the hot port… if you have some flexibility in the copper piping, you may not need a flex connector. Nipple length on the combined hot outlet/anodes varies, which is why I suggested a flex connector. Also, I'm not a fan of dielectrics unions as they fail in various ways. Using a plastic lined nipple with the dielectric built into good flex connectors gives the protection without the problems. 😎
Yours, Larry
1 -
OK, got it, its a combined outlet port/anode.
This will require unscrewing the existing port to insert the combo unit, maybe easy, maybe not.
I now understand that either a flexible line is needed or some re-plumbing or perhaps there is enough "flex" in the line to do this.
Thanks again for the insight. Mike
0 -
Quick update. I installed the Clack mixer last week and waited some days for the water to flow thru the system. Checked hardness today, I'm at 4 grains of hardness. I think this is good based on earlier comments above.
I've got a flexible Mg anode on its way to me, will install when I get it and monitor.
I assume the exit nipple simply unscrews, correct, so if inclined I can add a 2nd anode.
The hot water is clear at all faucets/tubs, but I still get some brownish water from the tank drain as I've been keeping an eye on it. It clears after a few seconds.
Thanks, Mike
0 -
Whats the possibility there is a bad ground at the panel? If the water piping is used as a ground the bad ground could cause premature failure on the water heater. I have seen this before…
Just a thought.
2 -
I'm going back… to overly softened water. Which, incidentally, if it is — is also causing havoc on the inside of the copper pipes… just sayin'
Br. Jamie, osb
Building superintendent/caretaker, 7200 sq. ft. historic house museum with dependencies in New England
5 -
I can check the ground its right at the incoming pipe for corrosion.
I have always felt this short life HW heater issue has been due to softened water. I have not had any piping issues to date unlike friends in the next county where the water corrodes the pipes. I've replaced a few for a friend.
As I stated, now running water with 4 grains of hardness, hoping for the best. Worst case is a new HW unit.
0 -
Why i suggested doing qualitative analysis on the precipitate to figure out what it is. I suspect it is copper from the pipes.
0 -
I really appreciate all the input and can look into analysis if I can collect some of the, my term, sediment.
A little researching does indicate that soft water can cause pitting, typically what is known as Type 3. I found this:
Numerous studies have associated soft water with type 3 copper pitting. Although corrosion of the pipe takes place, type 3 copper pitting is not normally associated with the production of pin-holes, resulting in a leak. Instead, it is associated with the generation of corrosive products such as copper sulphate. This type of corrosion can be readily identified by examining the cross-section of the pipe. Copper sulphate deposits, which have a bright blue color, will be found on the inner side of pipes where type three pitting has taken place. Some of these deposits may come loose, and flow in the water. This results in blue-colored water. I have done plenty of plumbing in the place and always see the greenish/blue on the ID.
More reading says that Type 1 pitting can generate cuprous oxide which is reddish and I could see it giving a brownish tint to the water. However, this doesn't seem common.
I know a lab guy, will talk with him next week about collecting/analyzing the sediment.
In the meantime, I'll keep an eye on it.
0 -
Hi, This is my "bible" of corrosion. https://www.academia.edu/5491377/Corrosion_engineering_mars_g_fontana
It was originally written years ago, so corrosion mechanisms are not referred to as type one, type two etc. They are called things like pitting corrosion, crevice corrosion, erosion corrosion and so on. Between this book and the Copper Development Association, I think you'll get a very good understanding of what could be going on.
Yours, Larry
ps. Testing will only help in your understanding. 😎
0 -
Ahh, originally or maybe it was later Fontana had a helper named Greene. At least that was the version back in my college days. Long time metallurgist here, but aqueous corrosion and copper alloys are not my experience; however, aerospace is.
0 -
Wow, much expanded from what I recall, guess we used an earlier version.
0 -
Also, if my copper pipes were corroding/pitting I would think I'd see the brownish water on the cold side also. Just thinking….
0 -
Hi, I have the original Fontana and Greene book, second edition from 1978. I understand that Fontana carried on with the updated version.
I wonder how much fun or trouble it would be to cut out a short length of copper pipe and split it lengthwise. That would give you a very good idea of what's happening with the copper.
Yours, Larry
Ps. If all water is softened, I would expect to see something in both hot and cold, but there will be more damage to the hot side.
0 -
the current from the powered anode may be causing the copper to precipitate
0 -
Mattmia2,
Yep, that was exactly the thought! The powered anode is now gone (I should have taken it to lab at work to analyze debris on it, but its back at Corro Protec). I'm back to the standard anode for the past few weeks. Will install Mg anode when it gets here.
With powered anode water was really discolored, now it's pretty much gone, but draining a bit from tank will show a bit. It clears quickly. Thinking another drain/flush can't hurt, like I don't have anything else to do.
I,m thinking of bypassing the softener for a bit just to rule it out, got nothing to lose.
Appreciate all taking an interest in this.
0 -
Presumably the copper is still there, it just is staying in solution and the tank isn't protected as well. I wonder if flushing regularly would keep it from making it in to the outlet in noticeable quantities. Could always filter it out too, but the best idea is probably to not over-soften.
0 -
OK, I'm back to report on what I found and wrap up discussion. I don't like to leave these open ended.
I had the debris analyzes using energy dispersive analysis by a pal at work and it turned to be what I thought it was way back in April when this started.
Drum roll……! The sediment or debris is just iron oxide or rust. I think we all know what this means for my HW tank. And, although it was nice seeing the Mars Fontana corrosion book, it looks like i don't need to study aqueous corrosion for this issue.
No iron in the system unless my water softener is spewing iron, which is not likely.
Some lessons here, I think.
- The powered anode just didn't protect the tank, so stick with standard anode, preferably Mg and monitor/flush tank periodically. Of course, it's possible, the tank is defective, but I'm doubtful.
- Water softeners over-soften water, I've got it toned down so that my water has 4 grains of hardness. I may bypass the softener and see if there are any complaints, if not, I'll turn it off.
- Although I've gotten efficient at replacing HW tanks at this point and will replace this one before it leaks, I'd rather do other things than replace HW tanks.
- I will keep the whole house sediment and charcoal filters that I have.
Comments, as always, appreciated.
Mike
1 -
Hi, Thanks for getting back to us. Just for background, I've worked on many thousands of tanks over decades. Failure rate of the tanks remained at about half of one percent per year. Got over fifty years from tanks. Started with some tanks that had just a plug for the anode. So, my question is if the tank leaked, will it damage things? If not, I would be tempted to let it run with the magnesium rod and check the rod in a year. I've actually had tanks "heal" as the anode created a bandage of hardness scale over exposed steel. This reduced the need for frequent anode replacement. Just a thought. 🤠
Yours, Larry
0 -
Larry,
Thanks for the input. As discussed tank is not leaking yet.
I could easily put the Mg rod in and continue to run. If it leaks, it won't damage anything, I, like many, could clean up my basement bit though.
Basement is unfinished so not a huge concern though I don't want a mess. I also have a floor drain. These usually start as a leak, not a full-blown water hose shooting all over the place, so I guess I've got time
I'm unhappy with the powered anode and in the back of my mind, still wonder if it caused this issue.
I've got a Rheem picked out if I need to replace.
Thanks again for your insight.
Mike
0 -
If they're leaking frequently i'd probably put a pan under it and pipe it to the drain.
1 -
Mattmia2, great idea on pan, but it will require a replumb to get it under the tank.
Installed 44" long, 0.83" dia. flexible Mg anode.
The Aluminum one I removed had a month or maybe 2 of service. All intact with the white Al oxide on it.
We'll see how the Mg anode and the tank does.
1 -
what’s your water pressure set at? 50-60 psi is adequate, and a thermal expansion tank.
Bob "hot rod" Rohr
trainer for Caleffi NA
Living the hydronic dream0 -
Mg anode been in 6 days, hot water has really cleared up. Thinking this is good.
Water pressure is 55 psi or so, has a regulator, required by code, also a no push back valve so also have a thermal expansion tank.
Mike
0 -
Hi, I like the apparent results from the magnesium anode. I've not heard the term "push back valve". Is this some sort of check valve… or something teenagers carry around? 😏
Yours, Larry
0
Categories
- All Categories
- 86.3K THE MAIN WALL
- 3.1K A-C, Heat Pumps & Refrigeration
- 53 Biomass
- 422 Carbon Monoxide Awareness
- 90 Chimneys & Flues
- 2K Domestic Hot Water
- 5.4K Gas Heating
- 99 Geothermal
- 156 Indoor-Air Quality
- 3.4K Oil Heating
- 63 Pipe Deterioration
- 916 Plumbing
- 6K Radiant Heating
- 381 Solar
- 14.9K Strictly Steam
- 3.3K Thermostats and Controls
- 53 Water Quality
- 41 Industry Classes
- 47 Job Opportunities
- 17 Recall Announcements