Might my Noritz tankless be misreading the flow rate?

Users who are viewing this thread

Fitter30

Well-Known Member
Messages
2,860
Reaction score
457
Points
83
Location
Peace valley missouri
Flow control is read by a magnetic sensor item 412
411 is the complete sensor
 

Attachments

  • Screenshot_2022-08-12-08-06-59.png
    Screenshot_2022-08-12-08-06-59.png
    189.2 KB · Views: 33
  • Screenshot_2022-08-12-08-05-44.png
    Screenshot_2022-08-12-08-05-44.png
    179.3 KB · Views: 33

Reach4

Well-Known Member
Messages
35,932
Reaction score
3,758
Points
113
Location
IL
it appears to use an impeller for sensing flow.
Hard water deposits on a turbine blade could cause low-reading, I would think.

Is there a symptom other than the measuring discrepancy? I would think that symptom might be to require more flow to start the WH. You might have said that, and I missed it.
 

Nathan Meyers

New Member
Messages
19
Reaction score
1
Points
3
Location
San Diego, CA
Hard water deposits on a turbine blade could cause low-reading, I would think.

Is there a symptom other than the measuring discrepancy? I would think that symptom might be to require more flow to start the WH. You might have said that, and I missed it.
Thanks for responding, Reach4.

The water is softened, so deposits are unlikely. I don't know if burner trigger points have changed - this is something I haven't paid attention to until recently. Hot showers were fine for 1-1/2 years, until about a month ago. That's when we returned from a vacation during which the water supply was turned off. Coincidence? Tech says the WH will not be affected. I've cleaned and descaled the Speakman showerhead, which did not help. The discrepancies are strange, and appear to put our showers much closer to the trigger point than they really should be.
 
Last edited:

Nathan Meyers

New Member
Messages
19
Reaction score
1
Points
3
Location
San Diego, CA
Here's the latest in the saga of my tankless cold showers...

I have measured much more carefully, using a gallon bucket, and confirmed that the unit is consistently reporting flow rate about 30% low, leading to unfortunate heating decisions. The plumbing tech's response in a revisit is: it's not a precision instrument.

30% error couldn't possibly be within tolerance, could it? Am I being unreasonable to expect heating at .95 gal/minute? The good news is that the plumbing tech is arranging for me to talk to the area Noritz rep, so maybe we'll make some progress.
 

Fitter30

Well-Known Member
Messages
2,860
Reaction score
457
Points
83
Location
Peace valley missouri
Would think 30% is way off. The magnetic flow meter technology has been around for 20 years at least. If it takes 64 oz / .5 gpm to make flow switch 30% difference 83.2 oz. The difference in the flow switch should only be a problem at low flow after switch is made temp sensor should take over.
 

Nathan Meyers

New Member
Messages
19
Reaction score
1
Points
3
Location
San Diego, CA
Would think 30% is way off.
Yes. The tech has been giving me excuses like "it's not a precision instrument" and "all parts have tolerances" and "inlet temp is higher in the summer", but 30% off is just way too far out of spec. Needing to pull 1 gpm to trigger the burner is way too far out of spec. Is it extremely difficult to engage Noritz in these issues? Is that why the tech is trying to convince me this is not the heater's problem?

Yes, he went through a phone call with Noritz and reported some values from diagnostic mode, The tests passed, but it's obvious they do not measure the accuracy of the flow reading. I can show the problem clearly with a bucket and a stopwatch, but apparently that counts for very little. I don't like being a screaming obnoxious customer, but things are heading in that direction.
 

John Gayewski

In the Trades
Messages
2,656
Reaction score
695
Points
113
Location
Iowa
Yes. The tech has been giving me excuses like "it's not a precision instrument" and "all parts have tolerances" and "inlet temp is higher in the summer", but 30% off is just way too far out of spec. Needing to pull 1 gpm to trigger the burner is way too far out of spec. Is it extremely difficult to engage Noritz in these issues? Is that why the tech is trying to convince me this is not the heater's problem?

Yes, he went through a phone call with Noritz and reported some values from diagnostic mode, The tests passed, but it's obvious they do not measure the accuracy of the flow reading. I can show the problem clearly with a bucket and a stopwatch, but apparently that counts for very little. I don't like being a screaming obnoxious customer, but things are heading in that direction.
Bottom line, itsy not performing as advertised you should get what you payed for.
 

Nathan Meyers

New Member
Messages
19
Reaction score
1
Points
3
Location
San Diego, CA
It turns out Noritz is much more end-user friendly than I expected. A new flow meter is on the way. They tell me I can even DIY replacing it, which I would do if I could get the cover off. Seems I need to have a conversation with my contractor about their use of an impact driver to tighten the cover screws after the last flush. Ugh... it's been a long couple of weeks.
 

Dpenz

New Member
Messages
1
Reaction score
1
Points
3
Location
Cumming, GA
Nathan, I would be interested to learn your results with the replacement flow sensor.

I have considerable experience with flow sensors used in Rheem water heaters in my neighborhood, as this part is a common failure item. It is indeed a "low precision" device, very inexpensive to manufacture, and cannot be expected to give very precise or accurate measurements. In my case, the device consists of a plastic vaned impeller with an integral stainless steel shaft that spins in a plastic carrier. The impeller is magnetized, and as it rotates it generates a varying magnetic field sensed by a Hall effect pickup, that converts rotation into "pulses" per minute read by the control computer. Yours is probably similar in design.
Theoretically, each sensor would have a "pulses per gallon" calibration that the computer would use to convert the reading to "gallons per minute". But I measured pulse frequency and tried to correlate this with the computer reading, with little success.

I concluded that the computer may not need to know actual flow rate precisely, but rather needs to discriminate between "no flow" and "some flow". Some flow rate is a prerequisite to know that the entering and leaving water temperatures are indicative of actual flowing conditions, and also to ensure that the heater has some amount of water to heat. The computer can't control or anticipate the actual water flow rate, so it must primarily use leaving water temperature to control burner firing rate. If the burner turns down to its lowest firing rate and the leaving temperature exceeds setpoint the burner must turn off, regardless of flow rate.

In my experience, all flow sensor failures in my neighborhood have been due to fouling. I have cleaned several of these, and I believe the cleaned sensors will work just as well as new sensors, but I have not actually re-installed any to date. If you have a flow sensor assembly on the bench, rotate a small magnet around the sensor body and observe movement of the impeller. A clean or new impeller will respond quickly and smoothly to magnet motion. A fouled impeller may not move at all, or may move with a jerky motion.

The contaminant appears to be a black, powdery material, which I have surmised to be iron oxides or manganese dioxide or a combination of these. The particles are extremely fine, one micron or smaller, and cannot be filtered out by any practical means in a residential application. They are caused by trace amounts of elemental metals that leave the treatment plant and oxidize in the utility distribution piping. In the trace amounts that they appear in the water, the particles are entirely benign to humans. Unfortunately, the particles are "paramagnetic", that is, can be attracted by a magnet but are not magnetic themselves. So when they come upon the little magnetic impeller in our heaters, they stick on. They build up and eventually interfere with the magnetic signal or clog the impeller thrust bearing. The impeller can be cleaned with vigorous brushing, but if you do this, do not breathe the dust. The particles go into your lungs easily, but don't come out. I have found that the thrust bearing can be reamed out with a tiny drill bit.

I believe you indicated that you have a water softener, and maybe you have a filter downstream. My guess is that neither softener or filter would separate these particles, can you confirm this? A magnetic separator should work, but I have not been able to find any that have been certified for potable water use.
 

Reach4

Well-Known Member
Messages
35,932
Reaction score
3,758
Points
113
Location
IL
Dpenz: interesting writeup. I see that various iron sulfides will react with FeS and others to off-gas H2S. I wonder if some time in some acid solution outdoors would unstick the flow sensors. The product of the reaction may be magnetic, and the acid could even bind things tighter for all I know. But maybe not. Since you are having success cleaning the sensors mechanically, there is no need to look to a chemical method.

For filtering, the Pentek DGD-2501-20 will pass a fair flow, if you want to see if that works. But it sounds to me as if replacing the flow sensors every few years might be easier and cheaper than trying to filter out part of the particles.

I currently have the Pentek DGD-2501-20 as the last stage cartridge filter, for my well water, and it is before my softener. I expect to not continue using such a fine filter, but if I cross section my old filter cartridge, and see discoloration, I could decide to continue. The softener does have a pulse-producing plastic turbine with the Hall sensor. That turbine counts gallons to trigger the softener regeneration after an amount of gallons.

When you drain a few gallons out of a tank WH, you may find some black water. I think that can be finely divided FeS produced in the tank, possibly by SRB (sulfate reducing bacteria). Those bacteria produce H2S. https://en.wikipedia.org/wiki/Sulfate-reducing_microorganism Then maybe the H2S reacts with the iron to produce the iron sulfides.
 
Last edited:

Nathan Meyers

New Member
Messages
19
Reaction score
1
Points
3
Location
San Diego, CA
Nathan, I would be interested to learn your results with the replacement flow sensor.

I have considerable experience with flow sensors used in Rheem water heaters in my neighborhood, as this part is a common failure item. It is indeed a "low precision" device, very inexpensive to manufacture, and cannot be expected to give very precise or accurate measurements. In my case, the device consists of a plastic vaned impeller with an integral stainless steel shaft that spins in a plastic carrier. The impeller is magnetized, and as it rotates it generates a varying magnetic field sensed by a Hall effect pickup, that converts rotation into "pulses" per minute read by the control computer. Yours is probably similar in design.
Theoretically, each sensor would have a "pulses per gallon" calibration that the computer would use to convert the reading to "gallons per minute". But I measured pulse frequency and tried to correlate this with the computer reading, with little success.

I concluded that the computer may not need to know actual flow rate precisely, but rather needs to discriminate between "no flow" and "some flow". Some flow rate is a prerequisite to know that the entering and leaving water temperatures are indicative of actual flowing conditions, and also to ensure that the heater has some amount of water to heat. The computer can't control or anticipate the actual water flow rate, so it must primarily use leaving water temperature to control burner firing rate. If the burner turns down to its lowest firing rate and the leaving temperature exceeds setpoint the burner must turn off, regardless of flow rate.

In my experience, all flow sensor failures in my neighborhood have been due to fouling. I have cleaned several of these, and I believe the cleaned sensors will work just as well as new sensors, but I have not actually re-installed any to date. If you have a flow sensor assembly on the bench, rotate a small magnet around the sensor body and observe movement of the impeller. A clean or new impeller will respond quickly and smoothly to magnet motion. A fouled impeller may not move at all, or may move with a jerky motion.

The contaminant appears to be a black, powdery material, which I have surmised to be iron oxides or manganese dioxide or a combination of these. The particles are extremely fine, one micron or smaller, and cannot be filtered out by any practical means in a residential application. They are caused by trace amounts of elemental metals that leave the treatment plant and oxidize in the utility distribution piping. In the trace amounts that they appear in the water, the particles are entirely benign to humans. Unfortunately, the particles are "paramagnetic", that is, can be attracted by a magnet but are not magnetic themselves. So when they come upon the little magnetic impeller in our heaters, they stick on. They build up and eventually interfere with the magnetic signal or clog the impeller thrust bearing. The impeller can be cleaned with vigorous brushing, but if you do this, do not breathe the dust. The particles go into your lungs easily, but don't come out. I have found that the thrust bearing can be reamed out with a tiny drill bit.

I believe you indicated that you have a water softener, and maybe you have a filter downstream. My guess is that neither softener or filter would separate these particles, can you confirm this? A magnetic separator should work, but I have not been able to find any that have been certified for potable water use.

Noritz agreed enough with my assessment to send me a replacement. Improvement was slight but not spectacular, and not enough to completely solve the problem - at least during that warm summer period of high inlet temperature. If the purpose of the impeller really is just to discriminate between flow and no flow so the heater can use other control mechanisms, I think Noritz is doing itself a disservice by reporting flow to the nearest .1 gpm and making claims about when heating will take place.

Can't say I checked the old sensor for microscopic particles when it came out, but the impeller did appear to be spinning freely.

It didn't help that the 2 GPM flow restrictor in the Speakman shower head (legal in CA back when I bought it) was clearly not delivering 2 GPM, and I hated having to solve the problem by increasing flow. But that, ultimately, restored the ability to take hot showers.
 
Top
Hey, wait a minute.

This is awkward, but...

It looks like you're using an ad blocker. We get it, but (1) terrylove.com can't live without ads, and (2) ad blockers can cause issues with videos and comments. If you'd like to support the site, please allow ads.

If any particular ad is your REASON for blocking ads, please let us know. We might be able to do something about it. Thanks.
I've Disabled AdBlock    No Thanks