Suitable Sensor to Detect Air Flow in Closed PVC Pipes


#1

I am looking for sensor that is the simplest, cheapest or most reliable to detect air flow inside closed pipe.

So there are 2 pcs air pumps / blowers / aerators (220v 1100 watt each) for fish ponds.
These air blowers are connected with 2" PVC plastic pipes.

download

I will use:

  • Electron/Photon
  • 2 channel relay 5 volt
  • and also contactor.

relay

Day 1, 3, 5, 7 = Pump A will operate.
Day 2, 4, 6 = Pump B will operate.
So at one time, only 1 pump will operate.

Then there are 2 sensors installed near the output/mouth of air blowers to detect the air flow.
I don’t need to measure the speed/pressure of these air blowers.
I just need to know that air blowers are blowing air (or have air going thru the pipes).
They will have 1-way valve so that we are sure which one of the blowers is malfunctioning.

If sensor A cannot detect any air from air blower A, the Photon/Electron will turn on blower B and also do some notification/email. And same applies for sensor B.

Google told me to use:

  • flex sensor and water flow sensor.
    — expensive and continuous hard blowing air will damage/bend it permanently

  • BMP280 barometric pressure sensor
    — I assume air blowing = increase pressure
    — not sure how to install it inside the PVC pipes so that it won’t obstruct the airway
    — looks suitable but is there any other sensor that I should know?

bmp280


#2

First thought is the barometric sensor. You wouldn’t install it inside the pipe but rather install a small port onto one pipe with a hose of some sort connected to the sensor. I think there are also pressure sensors that you could thread into a 1/2" NPT fitting (or other size that is available on the sensor). How about this one:


or this

There’s also the idea of using something like this Inline pipe blower. Instead of using it to blow air, you would use it as a generator to sense air movement… then you could use the low voltage live wire to sense if the fan is moving or not.


#3

Oh wow, that Adafruit MPRLS ported pressure sensor breakout is very nice! That is a very new product so our sellers won’t sell it any time sooner. The shipping fee and tax to Indonesia will kill it.

I am trying to find the pressure transducer from our online marketplace. That inline pipe blower idea is similar to “water flow sensor”, which I fear the moving parts will be broken after 24/7 usage. Also some pressure/energy will be lost, I think.


#4

I don’t know the max pressure for your setup but you might find the raw components on aliexpress or similar:

https://www.aliexpress.com/item/FreeShipping-5PCS-lot-MPXV10GC6U-SENSOR-PRESSURE-PORTED-8-SOP-MPXV10GC6U/1832611990.html

You are correct about the inline fan idea, you will loose a bit of power or flow due to the restriction the fan induces. I don’t know about 24/7 usage but a pressure transducer should have the best longevity because of no moving parts.


#5

@ninjatill You could try using an automotive component - an air mass flow sensor - these will be very robust but require a 12V supply. You should be able to source these locally. I believe they measure the temperature difference as air flows over the sensor.


#6

In the past, I’ve had similar projects .
My suggestions:

Confirm that appropriate check valves are used at the discharge piping for both blowers. That’s important in a duplex system.

You may want to consider measuring the AMPS for each blower (simple split-core CT)

1/4" NPT pressure transducer, you are only looking for a few psi, depending on the submergence depth of the diffusers. Scale up the pressure range if it’s 5V output.

The cheapest (but not the best) way to confirm a blower is running is to measure the temperature of the metal discharge port, and also ambient temperature.

My vote : (2) Check Valves, (2) CT’s and (1) 3.3V pressure sensor past the manifold.


#7

Measuring AMPS for each blower (simple split core CT) = using this non invasive clamp to detect there is current going into the blower?

sensor

Can we always assume that malfunctioning/broken blower will NOT draw any current?
I know it is better to detect air movement in front of the blower mouth.
But it seems there isn’t any simple sensor to detect air movement.


#8

I typically use a CT board like this, for the Electron:

AMPS demonstrate the amount of work being performed by the blower, it’s physically moving many pounds of air and increasing the pressure. During operation (not startup), if AMPS start to increase that means the blower is moving more air - which generally means your piping system has a major leak. The Electron will also notice the drop in operating pressure from the pressure sensor.

If AMPS drop, your diffusers are clogged, somebody closed a valve, etc.

From what I’ve noticed, the typical failure mode for a regenerative blower is a catastrophic failure of the electric motor = 0 AMPS & 0 psi.

You can download the datasheet & pump curve for your specific Blower.
Create a psi verses cfm equation for your Election Code.
By measuring the operating pressure, the Electron can calculate the flow rate (cfm) that the blower is producing.
One thing often overlooked (but obvious on the pump curve) is that a lower operating pressure means you are moving more air.

The Electron Logic can recognize these changes over time:
Higher Amps = more Air produced
Lower Amps = less Air produced
Higher PSI = less Air produced
Lower PSI = more Air produced.

Since the Electron will calculate the cfm of Air based on the psi reading, it make a pretty good pump monitor. When the operating conditions do change, the logic can make an educated guess on what’s going wrong.
That’s why I’d recommend a pressure sensor over a “simple sensor to detect air movement”.

Check Valves:
In a multiple blower system, air will be forced backwards through the blower that’s not operating and causes it to spin backwards, if you don’t have check valves. I learned that one the hard way.

[Edit] Discharge Piping:
You may want to consider a short section of galvanized pipe for the pump discharge and manifold.
Depending on your blower and operating conditions, the discharge temperature can melt PVC. I learned that one the hard way too


#9

Thank you for sharing your experience. I really appreciate it. That’s a lot of information to digest.

I am not familiar with “pressure transducer” before. And I found out that my local seller has this “Pressure Transmitter DC 5V 0-30 Psi / 0.2 MPa Pressure Sensor” with specifications:
— Material: Stainless steel
— Voltage: DC 5V
— Pressure: 30 psi <-- I can’t find lower/smaller version
— Thread Type: 1/8" NPT
— Input: 0-30 psi
— Output: 0.5V~4.5V linear voltage output. 0 psi outputs 0.5V, 15psi outputs 2.5V, 30 psi outputs 4.5V.
tranducer

And my air blower specifications: 220v 1100watt pressure 14.5 kPa and output 1800 liter per min.

So the mentioned pressure transducer above has max input at 30 psi = 206 kPa, I assume it it can be used with my air blower? Is that the right way to choose the pressure transducer type?

And also for the check valve and discharge pipe (galvanized iron pipe), my mentor did mention it. Thanks for the reminder.

Thank you


#10

That sensor is also produced in a 5 psi version

A 5 psi Output Range would allow you to measure up to 3.5 psi without going over 3.3V on the Electron’s pin.
That’s higher than what your blower can produce, which is good.

The Bad News: I’ve used many of those exact sensors in the past and I’ve seen a high failure rate.
The ones that fail usually do so within the first week, so order several and test them for a week before you install.
In my experience, a decent pressure transducer cost $80-$100 USD minimum, but well worth it.


#11

A very cheap and dirty solution to this is to use a heated temperature sensing resistor.

The simplest way would be to bond a heating resistor to the sensing resistor and measure the sensing resistor. The result will be a rough estimate of temperature and airflow, and the smaller the surface area of the device, the more it should be dependent on airflow.


#12

#13

I would go for a simple sensor. Create a “sensor pipe” which is say 6 inches long. Now inside you put a small rectangle of plastic that is epoxied to a 12 Gauge wire which runs through the sides of the sensor pipe.

On the outside bend the wire at a right angle so you have an arm pointing down when no air is blowing through the pipe. I would then mount a rare earth magnet on the arm ( you will have to note the pole to the correct position.

Use a Hall Sensor to sense when the magnet is in the down position. I use a cheap $2.00 sensor for Melixus US5881LUA. It can operate from +5V and has an open drain output. With the open drain you can use the Photon’s internal pull-up to 3.3V and safely read the sensor output. Ada Fruit / DigiKey has these for sale an you can pull a data sheet from these sites.

So when the fan is on the rectangle of plastic will move in-line with the flow. On the outside of the pipe the arm has rotated. You can add a second sensor to detect full fan operation. This type of sensor will cause no load on the fan since the 12 gauge wire is thin.

What is good about a hall sensor is that it is a non contact to you don’t have wear over time.

If you want I can share a pcb layout that I use with these hall sensors. I use these in a garage door monitor that I build for friends. I have many that do okay in the 120F Phoenix Arizona garages all summer.

Each hall sensor is in a T0-92 3 pin part. You want to use this old type of package since it’s immune to soldering. Keep a leads long so the heat of the solder does not damage the sensor. ( I could not get the surface mount to work with my crude methods of smt.)

You should be able to find an equal to the Melixus sensor locally as these sensors are used in automotive applications.

I can send a drawing if this helps.

Message me if you need more info.

ChipMonk


#14

I am pm-ing you now.