tankless water heater flow sensor fouling

David Penz

Each home in my new (<5 years old) subdivision has (or originally had) a Rheem RTGH natural gas-fired tankless water heater. Flow sensor failures have been rampant, evidenced by the "no error code, no hot water" symptom. I examined several failed sensors and found the turbine wheels to be coated with a very fine black powder. The material seems most likely to be manganese dioxide, resulting from small quantities of dissolved manganese leaving the treatment plant and oxidizing in the utility distribution piping. MnO2 is paramagnetic, and therefore attracted to the magnetized turbine wheel.

I have not found any discussions on the web concerning this type of fouling, yet it seems to me that it could occur for any brand water heater with a turbine type flow sensor. Probably many or all water systems carry trace amounts of MnO2.

Reverse osmosis, distillation and sub-micron membrane filtration would be effective, but none of these is practical for a residence. A magnetic strainer installed just upstream of the heater would probably work. However, while there are many commercial strainers available for hydronic systems, none seem to be certified for domestic water.

Have other Wall readers seen this problem? Does it occur with Navien, Rinnai and other brand heaters? Is there a practical remedy, other than replacing the flow sensor assembly or removing and cleaning the fouled sensor regularly?

Jamie Hall

Manganese is sometimes found in water supplies. However, if the water supply is properly treated and chlorinated, neither it nor iron (dissolved ferrous) should be present.

This appears from your description to be a problem with your entire subdivision, however, which suggests that it could be iron or manganese in the water supply. Either will be oxidized in the treatment process (not, as you suggest, in the distribution piping). So the first thing I would do would be to contact your water supplier and ask from a complete test including metals of the finished water.

Then verify that the problem is, in fact, manganese dioxide (which is weakly magnetic) or, equally likely, magnetite iron, which is strongly magnetic.

Either one may also cause black staining of plumbing fixtures.

There are a number of home water treatment devices which will remove iron and manganese from the water. These are usually intended for well water supplies, but they can -- obviously -- be used for public supplies. This article has some decent information: https://www.purewaterproducts.com/articles/treating-manganese-in-well-water#:~:text=Manganic manganese can be easily removed by a,very small, a cartridge-style sediment filter will serve.?msclkid=aebc139faa3211ec8d891e924c29a55c and so does this one: https://www.awwa.org/portals/0/files/publications/documents/samples/ironandmanganesech4.pdf?msclkid=aebcbbbaaa3211ec993ae4e9d20e1e53

David Penz

Jamie, thank you for your post. I did consult with the manager of my county's water treatment plant. He advised that the treatment process oxidizes manganese and iron to the particulate forms, then the water goes through two stages of filtration including 0.1micron membrane filtration. Less than 0.02mg/l manganese leaves the plant. He suggested these trace amounts could oxidize in the distribution system.

Black staining has not been a noticeable problem.

Even homes with "whole house" filtration systems have had flow sensor fouling. These filters are probably too coarse to trap the sub-micron MnO2 or FeO2 particles.

I suppose it is possible that there might have been an abnormal event at the water plant, causing an unusual release of contaminants.

I'll look into the sources you referenced, thank you.

hot_rod

Do all brands of tankless use turbine sensors? I thought some just had a paddle type flow switch, which would be less prone to scaling.
Are they adding phosphates”film providers” at the water treatment. That is becoming common in areas with old lead pipes to prevent leaching

David Penz

The proportional flow measurement from the turbine type sensor may offer some advantage with respect to heater control, for example, to enable a feed forward signal for burner control. Also, the "cut-in" and "cut-out" flow points for this Rheem heater are different. I am also curious to learn what other types of switches or sensors might be used by other manufacturers. Any flow device that is not magnetic would fare better against MnO2 or FeO2 contamination.

It is not "scaling" in the sense of a hard coating but rather sub-micron particulate, magnetically attracted to the rotor. See photo attached.



The treatment plant is doing all that can be expected of them, including an oxidation process and two stages of filtration including a final 0.1micron membrane filter. Only a tiny amount of elemental manganese (<0.02mg/l) leaves the plant.

I believe a device like the vertical/horizontal Caleffi Dirtmag would be a practical fix. It is not intended for use with domestic water, but what specifically would preclude its use? Lead content of the brass? At least in my neighborhood, there would be a market for such a device, and I don't yet understand why this issue would not be occurring many other places.