Hi everybody , I am a newby here , but not new to what I have found to be a chronic issue with domestic hot water recirculating systems.

We built a new home 5 years ago and a good friend of ours who is in the top ranks of plumbers in our area helped with our project .

Any way we installed a hot water recirculating system due to the size of the home , all was well until last year I happened to be in the garage adjacent to our mechanical room and heard a spraying noise .

I opened the door and it sure was a spraying noise from a pin hole leak in a copper pipe at the hot water tank .

I shut the power to the recirculate pump and the hot water tanks off closed the service valves and called my friend, The first thing he asked was is the leak right there at the tanks where the pump is .

Bingo I said he said he has come across this before and it has always been within the area of the hot water tank or tanks and the pump and had no definite cause but thought it may be some type of electrolysis from stray electrical currents because of isolation areas of the copper pipes due to dielectric unions and the rubber O-rings at the pump flanges and service valves.

Being a proud I.B.E.W. wireman I realized this could be the culprit after looking at the system and got some ground clamps for water pipe bonding and ground wire and bonded every pipe as a system .

Feeling good about the grounding after replacing almost all the copper pipe in the problem area almost a year passed by and a leak in the 3/4" copper flex pipe that leaves the first hot water tank and enters the second one develops .

Being such a proud id_ot I realized there is rubber washers at each end of that flex hose and I failed in my bonding , so I replaced the the flex pipe and this time incorporated it in to complete the bonding system .

I told my friend and felt like ok this should put this issue to rest. Surprise a month or so later a small portion of the return loop that did not get replaced developed a leak . All which have started from the inside of the pipe and there is a green corrosion inside the affected pipes. I am sorry about the novel , but want to provide as much accurate information as possible for my best interest to rid this issue. After a lot of research I have found many causes that have attributed to the development of pin hole leaks in copper pipes ranging from improper water balance Ph ect, to velocity ,burs inside the pipe , excessive flux and solder but have no for sure reason found of why this pitting and corrosion occurs, how ever studies have one common factor in the many I have read and it all comes back to recirculating hot water.

It has been proven that when water is heated it changes the chemistry of the water one it lowers the Ph level and below a certain level water becomes corrosive more so chlorine that is in the water from treatment plants also has changes when water is heated and oxygen in the water are considered to be oxidants , with high temp and the presence of oxidants the corrosion rate of copper has been observed to increase fivefold when the water temp was increased from 80 to 170 Deg F.

Furthermore precipitated gas bubbles caused from the release of line pressure to atmospheric pressure evolve when water is released at the tap .

So now you have a corrosive water chemistry washing away the inside of your hot water pipes.

I have considered PEX but have heard for sure any brass fittings would not last and chlorine can break down plastic pipe also. I have reached out to all of you in search of a cure for this corrosion sickness that develops from recirculating hot water.

Thank you in advance for any insight and once again sorry for such a long drawn out post first time on the site .

Sincerely ,

Big Ken



Edited 1 times.

Are you using any chemicals in the water to remove oxygen? Recirculating water in a closed system, with oxygen is breaking down your copper from within. It should be obvious when the old copper is removed, by inspecting the interior of such pipe.

Best Wishes

you seem to be well read on this subject.
now you need to figure out which of the causes of pinholes relate to your situation.
could be more than one.
so, start by having your water tested for PH level.
figure out which pump you have, how long the recirc line is, and what are the temperatures of the water leaving the tank and the water returning. use that info to calculate velocity.
take some fittings apart to see if the plumber reamed the tubing.
was type 'L' tubing used rather than thinner wall type 'M'.
you have some detective waork ahead..

There are two items that may help your situation. The first is an aquastat which will only operate the pump when the temperature hits the low end set point and off at the high end. You would also want to insulate as much of the pipe as possible to reduce heat loss. A second step that you can take is to reduce the temperature of the hot water tank if it is higher than what you need. This will also reduce lime scale buildup in the tank or tanks. In the future if you develop leaks, I would suggest using water soluble flux. When I plumbed my own recirculating system I used water soluble flux for all the copper work, and pex on the return lines. I don't have any problems with pinholes at all. This could be due to the water in my area being chemically different from your area as well. Have you ever had the water tested for pH and the like?

Thank you for your input every bit counts , I have done a portion of the detective work as for the Ph level of the water is 7.2 hot and cold it is my understanding that below 7 corrosion occurs so that is one concern . I have saved and noted the position of the pipes top , bottom ect. as I have removed them from my system for review and observation purposes. By that being done I have confirmed the quality of the workmanship was performed to beyond satisfactory , as I mentioned my good friend helped or I helped him but never the less it does not hurt to check .The pump is a Armstrong Astro-50B-c 1/20 Hp. 2800 Rpm. I have not determined the velocity exactly but have ball park figured about 4FPS and if I stand correct copper pipe erosion starts to take place when flow exceeds 8 - 10 FPS. also not to say erosion would not take place in a straight horizontal or vertical run of the pipe, but it would be more likely where flow is restricted at elbows and tees , which so far the straight horizontal and vertical areas have developed these leaks close to the recirculating pump so that pushes me toward the formation of the tiny air bubbles that form when the tap is opened and the pressure changes as I mentioned before and there is always tiny air bubbles present at the tap using hot water to say rinse out your milk glass after dinner.

You do not specify WHERE in the circualton system the leaks are, but they are usually a turns and tees. They are usually caused by NOT modulating the discharge, not intake, of the pump to the point where it just maintains the temperature. If this is not done cavitation occurs at turns due to the velocity causing erosion of the copper. The rubber gaskets at the pump, etc, do NOT cause electrolysis, because the bolts or nuts on the connections to the pump are NOT "dielectric insulated". However, stray currents CAN cause water heater failure due to the dieclectrics on the supply lines so you do have to jumper the hot and cold water lines going into it.

If you're using 3/4" copper, you're probably over 10 ft/sec, with that pump.

3/4" would slow the velocity more than if they were using 1/2" but it could still be enough to erode the copper at the corners.

Looking at the curve for that pump, and figuring a reasonable head, he'd have to be using 1.25" to be safe.I know....a lot of assumptions.

You do NOT increase the size of the piping, you modulate the OUTPUT of the pump to the desired level, with a simple "globe" valve.

I CERTAINLY WASN'T SUGGESTING HE TEAR EVERYTHING OUT AND INSTALL LARGER PIPES.Not when a gravity recirc is all that is needed.

Unless the system is DESIGNED for a gravity system, which is often impossible with modern buildings, it probably won't work.

Let me state it a different way.....A flow or velocity similar to a gravity recirc is all that is needed.

So KC ROTO I have considered a temp activated control for the pump and also there are on demand controllers such as my grundfos booster pump has and it is fact that lower temp will reduce the rate of corrosion. I respect your input although what you stated is true , the fact is the rate of this pitting and corrosive will be reduced but still have presence in the system and I am Hell Bent on finding the combination of factors or if it is just one factor. I am going to rid the cause whatever it takes and cost. I have a timer that I installed when the system was put in but the demand for hot water in this house is unpredictable so we just let it run , besides that we built a heated towel rack in the master bath for the little women and got me a lot of good brownie points but when she gets out of the shower to a cold towel after being pampered with a nice warm one my brownie points are almost gone and I will be paying interest on them if I fail to keep it running . This towel rack is loaded with tees and elbows and on the next floor not any sign of corrosion so once again I am not thinking velocity is an issue the problems like I said before are right by the pump at the hot water tanks and have all been in straight runs horizontal and vertical but not in the elbows or tees. I have tested the water myself with a kit I bought Monday but Tomorrow I am taking 1 quart from the cold and 1 quart from the hot water to Central Washington Water and have it professionaly tested because more and more I keep getting pushed that direction as the process of elimination and good old common sense progress. Thank ALL of You again I will keep ya all up to date as this investigation proceeds. Sincerely, Big Ken

I was rereading your first post again and I realized part of what was bothering me. You have multiple tanks, but they are not configured in a manner that works as well as it should. You stated that you had a "3/4 copper flex pipe that leaves the first hot water tank and enters the second one" I have a couple of recommendations. The first is to change the way your tanks are configured. Since you are an electrician, you I can tell you to run your tanks in parallel, not in series. Running the tanks in series simply makes an inefficient on demand heater. You would be better served by distributing the water demand equally between the two tanks so that the tanks will keep a steady flow of hot water at a lower temperature setting. In addition, the elements in the first tank will last longer since they won't be getting hit with all the cold water all the time. Set up the incoming waterlines on a T with the inlets equidistant. The return line for the recirculation loop should enter incoming cold side just above the T. The hot water outlets should be configured in the same manner. The other thing I would recommend is hard piping with L 3/4 copper. It has a heavier wall thickness and it can absorb any vibration that the recirc pump may be causing. I don't have any proof, but I suspect that the vibration on the copper flex lines may be part of the issue. The flexible lines cause more turbulence inside them, and the walls aren't as heavy as tubing. Just my 2 cents, but I haven't had any leaks in the last 4 systems that I have installed for my family in the last 9 years.

That towel rack, with its tees and elbows, is made out of completely different materials than those in the piping system.

Well, you are entitled to your opinion, but mine is that a "series" installation is MUCH MORE efficient because it does not depend on some nebulous "mirror" image piping that may or may not create equal flow through both heaters. I have NEVER install a parallel system, other than in very large builings and then it was using the manufacturer's "engineered manifold" to connect them together.

That is different because a properly placed valve WILL do that.

the output (not the intake/suction side) needs to be throttled to reduce velocity

personally, I would install a union with a 1/8" or 3/16" diameter restrictor plate IMMEDIATELY after the pump

the pump will NOT overheat

it will pump less gph

it will draw less current

the tube will no longer erode



or get a Taco pump DESIGNED for HW recirc

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

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

In every instance I said the valve has to be on the OUTLET of the pump, to avoid cavitation.

correct

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

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

hj The towel rack is made of the same type of material as the plumbing on the hot water system . I am going to adventure my not so computer savvy skills and see if I can get a couple images of the system on here .

Ken,
I am not sure when you originally posted this, but I am having the EXACT same problems as you describe.
Have you made any further progress on fixing this?

The cause of pin-hole leaks in copper hot water recirculation systems is over-pumping - causing too high a water flow velocity through the copper tubing. Check the pump make and model and look at the pump curve to get the flow in gpm. Then calculate the water flow velocity through the pump suction piping. It should be less than 2 feet per second.