Water (sewer) level sensor

[Richard Crowley]

I am trying to identify some kind of sensor to detect the level of water (sewer) in the clean-out riser where the sewer exits my house into the lateral down to the sewer main line under the street.  It is a 4-inch plastic pipe with a threaded plug at the top.  I would like to make something that I could fasten to the inside of that plug and return something indicating the distance from the sensor down to the water surface level.

Reason: I am cleaning up my basement after the city sewer backed up into my downstairs shower.  It is a stinky, sloppy, and dangerous mess.  I had a lot of electronic stuff on the floor which is now contaminated and will need to be  trashed. The main sewer line in the street got blocked (by some fast-growing tree roots) and stuff accumulated and blocked the flow. (Stuff that shouldn't be flushed down the sewer!)

Yes, much of the cost will be covered by insurance, but I am still out several 100 dollars and a lot of time and mess.  I could install a back-flow prevention valve (at great cost to myself), but frankly even if I had one, I don't think it would have worked here anyway. The flow was so slow that it probably wouldn't have shut properly. 

It would be great to have something that will sense the fluid level in this riser, which turns out to be an excellent indicator of this kind of problem. I can use an Arduino or RasPi or BBB to send me a text message alert if the level rises to any abnormal level, BEFORE it starts flowing across my basement.

I don't know whether the ultrasonic and reflected light sensors would work because of the reflective surface inside the pipe. Most of the distance is below-ground so I can't use anything along the side of the pipe.  Ideally, I would like to use something non-contact and resistant to the very corrosive atmosphere inside the pipe.  I don't know how well the top is sealed (it probably is). I don't know if it would be practical to try to sense any change in air pressure in the pipe as the water level rises, etc.

[rikkitikkitavi]

If I understand it correctly you want to detect rising sewage in your outlet pipe due to a blocked communal sewage pipe (or extreme floods that stresses the systems capacity )

The rises is vertical and 4 inch ?  You have only access through the small threaded plug?

that means that given normal household sewage flows, the pipe will never fill up with outflow (you need thousands of gallons/min flow for that)

But given the fouling properties (no pun intended) of household sewage I would not consideder optical or sonic sensors, as they will be dirty after some service time and needs repetitive cleaning and this might affect their capacity to detect.

I see two options for sensing level in the riser pipe

1)  Dip tube with air bubble through it (my recommendtation)

http://automationwiki.com/index.php?title=Bubbler_Level_Measurement

This is very common in the industry when you have fouling liquids. Use any normal aquarium air pump that can push downt to say 50 cm depth or so, and a low pressure sensor. Use a thin stainless tube inserted to desired early warning depth. You might have to rinse it once in a while , but any fouling will cause false alarm
Can be built with easily obtainable items like aquarium tubing, pvc/stainless steel pipe , air pump , simple pressure sensors from you local electronics etc etc.

Protective tube surrounding it would be a good idea, also not inserting the bubble tube at the very bottom, but putting it some distance from normal liquid level in the riser pipe reduces fouling.


2) A mechanical level switch inserted through the thread. Suggestion is mechanical floater with vertical action, not a horisontal  lever arm.
Or a conductive sensor, two dip sensors. But to prevent fouling I recommend using a protective tube surrounding it, with a closed top to prevent fouling entering from above. It is less sensitive to fouling meaning cleaning is less frequent. If possible  place it oppostie side of the inlet to the riser pipe.

More meccano work I presume.

regards
Rickard ( chemical engineer with 15 years of experience and frequent electronic hobbyist...)


 

[wreeve]

"mechanical floater with vertical action" this is how our commercial sewage pump senses level.

[pickle9000]

A float sensor will fit in a 4 inch pipe is simple and reliable. You could even rig a servo to trip it on a regular basis to ensure it is working.

[tautech]

Richard, see this thread and the one linked in it:
https://www.eevblog.com/forum/beginners/low-power-not-gate/

These ideas, I imagine will use stainless probes for level detection.

[Kleinstein]

I would consider using a commercial sensor. The problem is nothing so unusuall so I would extect commercial solutions available. You might even get some mony back from the insurance as this reduces the risk.  Given the difficult enviroment, this might be just a presure sensor. I am afraid even a rather robust swimmer might not be very reliably and may just collect debirs to block the tube or at least the swimmer.

Another option might be ultrasound from the outside of the tube, possibly even detecting die direction of flow. A fully filled tube should be easily be detected by ultrasound going vertical through the tude: water is much more transparent to ultrasound than air.
 

[tautech]

Just done a Google search for "proximity sensor for liquids" and got plenty of hits.

[rikkitikkitavi]

A float sensor will fit in a 4 inch pipe is simple and reliable. You could even rig a servo to trip it on a regular basis to ensure it is working.

If it is a bladder  sensor I would agree,but sewage is fouling. TP (Toilet paper...) , waste of various kinds might stick and prevent physical movement. I have seen this with "crystallising solutions" that splashed upon high level switches placed above normal liquid level. Critical when the high level switch prevents overfilling.

Some kind of physical protection is necessary.

All solvable with cleaning, but is that something you want to do often?
In my experience a bubble dip tube is less prone to fouling and easier for remote cleaning, i e flush water down the tube instead of air.

A capacitve sensor is sensitive if the fouling matter contains water (it does...) but false larms might not be a problem.

If you can prevent fouling of the transducer head ultrasonic or radar level measurment is extremely reliable, but expensive (but not as expensive as cleaning out the mess... and trashing valuable electronics)

[Gyro]

Sorry to hear about the mess, and the losses 

What about a simple but robust conductivity sensor? A couple of substantial brass rods, maybe threaded at the top, fixed securely into your threaded cap (I assume plastic like the pipe) and protruding as far down the pipe as you need. They'd need to be well spaced but also well clear of the pipe walls. The rods would need to be rigid enough not to need any spacers lower down.

I suspect that what you're dealing with is rather more conductive than tap water. Cleaning with a hose wouldn't be too messy and would probably only be needed if the level actually reached the top of the pipe anyway, anything else should just fall off the ends of the rods.

Maybe less likely to jam up and need cleaning than a float sensor.

[C]

Would preventing the problem be better then sensing a problem.

I would guess that normal sewer main line level is not that far below your downstairs shower drain. You do not want to be the low guy that wins.

If you pump the downstairs shower drain up to a higher level before it flows in to a vented sewer pipe you can prevent the problem.

You have added cost of pumping

C

[lincoln]

Hello,
 Here is one that works well: http://www.mcmaster.com/#50195k65 It cam be converted from NO to NC by inverting the float. This was installed in the in-laws system. On the commercial side we used a 3" float in a stilling well (4" piece of pipe with an open end.  And a top side switch. The switch had a ling of all thread that ran through it to the float.

[donmr]

Look for "submersible pressure level sensor" like these http://www.dwyer-inst.com/Product/Level/LevelTransmitters/Submersible/.
These have a breather tube so they can compare the pressure under the liquid to atmospheric pressure.

[calexanian]

Fuel gauge sender unit from a car? They typically have like a 90 to 250 ohm pot in them and the pot is pretty robust. They last a long time in a fouled gas tank and are cheap. I see no reason they would not work in effluent water.,

[ez24]

Is there any chance we could have a drawing?

[nowlan]

Came to recommend the bubbler, as has been mentioned.




Doubt it would be cheap, unless surplus/used on ebay, or you build something similar.
Pressure sensors are reasonable cheap.

[ez24]

would a float switch like on this sewage pump work?  After all this is designed for submersion in sewage.
My guess is you could find the float by itself.  You may need a "basin" for room for it.  Sewage has its own challenges. (no smelly icon)

http://www.amazon.com/WAYNE-RPP50-Sewage-Tether-Switch/dp/B0007CTMRS/ref=sr_1_1?ie=UTF8&qid=1440574913&sr=8-1&keywords=sewage+pump

[Joule Thief]

I would be concerned the solids flowing through the pipe would eventually snag onto any type of float mechanism.

I have seen reflective sensors made of clear plexiglass rod where an LED light shines down the rod and bounces back up to a sensor also mounted on top of the rod. The sensing end of the plexi rod is cut and polished in a 45 degree cone shape. The cone shape allows the LED light to travel the length of the rod, hit the cone shape and reflect back to the sensor mounted to the top of the rod.

When the sensing end becomes immersed in liquid, the LED light is no longer reflected back by the cone shape as contact with a liquid will allow the LED light to keep traveling out through the liquid present in the pipe.

http://www.sensorsmag.com/sensors-mag/optical-liquid-level-sensing-eases-industrial-applications-12920