Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Making a substitute water flow sensor

Status
Not open for further replies.

Mosaic

Well-Known Member
Ok, as is typical in 3rd world countries; our local Stiebel Eltron tankless water heater dealer sells new units for $350USD and replacement flow sensors for $50 USD.

Also, they have none in stock and cannot say, if ever, they will restock.

Thus is born the need for a substitute.

My poking around has revealed a 3 pin plug flow sensor similar to a muffin FAN RPM capable power plug with a middle wire providing around 100 to 500Hz 5V square wave as the speed. Note the pic in the link below.
http://www.stiebel-eltron-usa.com/dhc-e.html

Now all we need is a 100Hz+ signal to enable the water heater as it's water temp is controlled by a temp sensor on the output. A fixed 100hz won't do since the temperature is measured on the copper pipe well away from the actual heater coil and could burn the heater coil with no actual water flow.

So I am considering something like this:
http://www.discovercircuits.com/DJ-Circuits/h2oflowindicator.htm
based on piezo sensing vibration.

As a note, the actual flow sensor has a little spoked water wheel spinning a cylindrical magnet with a hall sensor pulse switch in the housing detecting magnetic poles.

Advice is needed.
 
Have you confirmed that it is the flow sensor at fault and not the control module?

the actual flow sensor has a little spoked water wheel spinning a cylindrical magnet with a hall sensor pulse switch in the housing detecting magnetic poles.
There is not a lot to go wrong with a flow sensor like that.
The bearings fail and the wheel jams.
The magnet loses its magnetism or falls off and runs away down the pipe!
The hall effect sensor fails.

Assuming the magnet is still there and spinning, why not just get a hall sensor and try replacing that?
Much easier than re-inventing the wheel using a different technology.

JimB
 
We have seriously bad water at times. On average once a week we get > 10% silt (as seen in a standing glass) . Occasionally we get diesel in the water and raw sewage. It's that bad in the freaking 3rd world. BTW, no one gets 'fired' for that as the water company is a state enterprise.
https://www.trinidadexpress.com/news/Diesel_spill_shuts_down_water_plant_at_Guanapo_-159964975.html
https://www.newsday.co.tt/news/0,150184.html

That said....I also have to maintain 1000 gal. reserve as water is not 24/7 at all.

Ok, so back to a non contact sensor approach as the the 'water wheel' sensor seems to wear out quickly.
 
OK, I appreciate that you have "a bit of a problem" :eek:

JimB
 
Have you confirmed that it is the flow sensor at fault and not the control module?


JimB

Yes, via oscilloscope I can see what the sensor is doing. By taking it out, dismantling and cleaning it up i got it to work for a couple of weeks before the bad water did it in again. I am guessing a bit of sand is jamming the journal bearing which has a lot of play anyway. I can get it started by tapping on it while there is flow.
 
Have you considered a non-mechanical type of flow-switch, such as this?
 
I decided to go for an external unit that's more serviceable to deal with the water quality issues.
**broken link removed**

Perhaps a once a month cleanup maintenance schedule should keep the silt caking minimized and offset the wear.
The specific failure mode is: when dirty water is in the sensor and flow stops, it settles into the bearing which is loose tolerance. This creates a a pair of 'bumps' on either side of the spindle. Depending on the silt make up it can jam up the spindle when flow restarts.
 
Have you not considered fitting a small inline filter to clean the water up before it enters the heater? It would surely be easier to clean out a filter than to strip the heater down once a month. Apart from that, the filter is designed with maintenance in mind, the flow sensor isn't and will eventually require seals etc causing you even more headaches :)
 
If the water supply is bad enough to clog a filter in a day of normal use, and it's impractical to clean it out daily, then certainly without one you will be filling the heating tank and coating the element with sediment which will drastically shorten the life of the heater anyway, quite possibly killing it before the flow sensor even shows signs of wear. I would definitely look to clean up the water supply feeding the heater, that's the real source of your issue. Until you do, you are always going to have problems with the heater. Maybe look into feeding it from a break tank so that you can separate the worst of the sediment and then filter it?
 
The solution to all problems indeed is limited by voter choice of elected officials responsible for affecting changes.

But the technical problem root cause is contamination, not flow sensor.
I had good success with a powdered polyester, polyurethane sand filter pump when I had a 24' swimming pool. Backflush rates depend on flow, impurity and differential pressure or rise in single-ended presure relative to after backflush. It can be automated but solenoid valves may add too much cost unless you find useful ones from old washing machines. (hint)

There are many other minerals for sand which may be added for removing sulpher, diesel, solids consider adding a chlorine siphon drip feed before filter or during backflush.

Consider a Chinese unit with patented hand rotated comb-like top aerator rotated by hand for more effective backflush.
Bigger is better in cu.ft. capacity. Pressure sensor of any type is mandatory.

If water supply is not 24/7, toss TOU heaters and only use biggest glass lined tanks with brass gear pump on output of tank.
 
Last edited:
I would also explore a spin down filter (in your case, as large a housing as you can find). Here is one example: https://www.amazon.com/Rusco-Separator-Sediment-Additional-Replacement/dp/B00G030LZK

I have one and works well.

They work by a cyclone type effect. They spin the water as it comes in the heavier dirt goes to the outside and the cleaner water is drawn from the center. The dirt can then be drained/washed out the bottom. I would suggest you add an electric solenoid valve to the bottom hooked to a timer that will automatically wash every x hours.

From here you can add finer filters downstream without them clogged frequently and possibly irreparably.
 
I would also explore a spin down filter (in your case, as large a housing as you can find). Here is one example: https://www.amazon.com/Rusco-Separator-Sediment-Additional-Replacement/dp/B00G030LZK

I have one and works well.

They work by a cyclone type effect. They spin the water as it comes in the heavier dirt goes to the outside and the cleaner water is drawn from the center. The dirt can then be drained/washed out the bottom. I would suggest you add an electric solenoid valve to the bottom hooked to a timer that will automatically wash every x hours.

From here you can add finer filters downstream without them clogged frequently and possibly irreparably.
The unit in the link is nothing more than a screen, which is still a useful thing to have, but doesn't use the cyclone effect. For cyclone type separation, you need a true hydrocyclone:
https://en.wikipedia.org/wiki/Hydrocyclone
 
The unit in the link is nothing more than a screen, which is still a useful thing to have, but doesn't use the cyclone effect. For cyclone type separation, you need a true hydrocyclone:
https://en.wikipedia.org/wiki/Hydrocyclone

I was a little skeptical, but the one I have is made by American Granby, a reputable plumbing manufacturer and was sold to me by a reputable plumbing supply house (ie., not Lowe's/Home Depot helpers). It DOES work by cyclone effect. They are designed to work at a minimum water flow rate. If you have a trickle running in the kitchen faucet or when you first open the bathtub faucet, the screen does the work until the flow rate reaches the minimum. If I have the ability I would post a video to show you. Please trust me, it does work and works by cyclone-type effect (perhaps centrifugal effect may be a more apt description) .
 
Ok, I am considering this spindown right after the pump.
https://www.amazon.com/gp/product/B00LV02NUS/ref=ox_sc_act_title_4?ie=UTF8&psc=1&smid=AXIP0A1I8BZRS

and this 3 stage to feed the washing machine and water heater.
**broken link removed**

What do u think?

I would put a high pressure switch inline between the pump and the filter. The filter housings tend to be acrylic, with a lower than average burst pressure and definitely lower than the pressure most pumps can supply. The pressure switch is nothing more than you standard pump switch with the low (on) setting near the upper end of the pressure you have now, and the high (off) pressure lower than the pressure capacity of the spin down filter. A unit similar to this **broken link removed**. The spindown filter will remove every above x microns due to the screen.

The ebay unit you link to is commonly referred to as a 10 x 2.5 filter. If the water is of the poor quality I envision by your description, I would look at doing a build our own from 10x4 filters more. Especially if you know you to solder copper pipe. The is a pi*d and pi*r^2 increase in the amount of filter material and/or filter surface, but less than a proportional increase in the cost of the replacement filters. Additionally, building your own allows you to change and use the number and type of filters to best suit your need. I would do it for the whole house, with the exception of the water line used for watering the lawn. Just don't forget to add shut off valves before and after the filter bank so you can change them without having to drain all the plumbing. You may want to even include a bypass loop in case something happens to the filter bank, you won't be without water.
 
My pool just used sand in the large tank , which could cheaply be replaced if I wanted with a few bags. Backflush was done weekly for a minute every week, but then circ pump was continuous. Better quality "sand" improves removal efficacy of different contaminants. In my case, when I moved in the pool was a science project with Soylant green water that turned milky white with Chlorine which then was discharged by backflushing..
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top