Air Pulse Trigger Sensor

Hello all,

I have been searching for a sensor that trigger a response with an air pulse. I have BMP180 and BMP280 sensor that I am currently working with, but the BMP280 refresh rate is too low for my application at 1Hz( As far as I can tell it wont refresh fast enough and that is what I am reading through my searches).  
Although the BMP180 does seem to work in my test conditions I am worried if dirty air may hinder it's reading in the long run, additionally I need two sensors working simultaneously and the BMP180 library has a fixed address for I2C(I attempted to make a modified version for a different address with no success, but I believe that is my lack of knowledge).
 Ultimately I am not interested in measuring pressure itself, but a pulse of air pressure. The program is currently written to detect a sudden increase in pressure to trigger, but that is only because I am using pressure sensors for the reading. Please advise, I would be grateful. 


  • Has to respond to air pulse.
  • Faster than 50ms refresh(20Hz), but ideally lower than 15ms.
  • Low power(Will be on batteries).
  • Need 2 separate readings.


What size of pulse are we talking about here?

Hello Mike,
I would describe the pulse as a significant pressure increase, for test purposes I am blowing through a straw directly at the sensor at mild/medium strength, but I do not have an actual exact measurement. I can adjust my design to concentrate the pulse or dilute through channels, I just need a consistent reactive trigger for it. As for the way it is programmed at the moment I am just calculating the average pressure of 100 reading in hPa + threshold and using a while loop until it returns to normal conditions.

I once did some experiments on a blown instrument for disabled people so the blow could not be too hard. I use a BMP183 pressure sensor into a leaky air chamber to detect both the blow and the blow pressure. That was very sensitive and responded to low pressure blowing like a real instrument. However, the problem was that the air chamber just gathered spittle and the air became 100% saturated and the sensor stopped working. The next step I will investigate is some sort of filter / moisture trap to stop this happening.

Hope that helps.

Hotwire Anemometer.

Yes, good idea.
You can make one by breaking the glass on an incandescent flash light bulb. Do this by genitally squeezing it in a vice.

Then passing some current through it, much much less than you need to make it glow and measuring the voltage across it as you blow onto it.

I really hope that wasn't a Freudian slip!

1 Like

Well I am dyslexic and homophones are always difficult, I couldn't get the spell checker to help, so I looked it up on the internet and apparently that is the way UK suppliers spell it.
UK hardware supplier

Blowing into something is in the < 5psi range. An HVAC pressure switch will respond in that range but I don't know it they have a 20Hz rate. You can usually get them from HVAC suppliers for around $20.

Alternately, I'd check what Digikey has available in that range.

I was really referring to that :slight_smile:

Yes homophones again. Sorry but I have been dyslexic all my life and at school it was known as being thick as it wasn't invented then. While at the age of 70 I do my best to get things right but don't always succeed, like a toothless budgerigar, but I am happy if it gives you a laugh. :wink:

Oh, I certainly don't want to make fun of your dyslexia. I thought that "genital" and "vice" in the same sentence was funny, but it was pretty clear what you meant.

Thanks for the input, I think I may have found a solution using a Piezo sensor instead. I am reading the analog input and allowing the air pulse knock the sensor against a backboard which in test conditions has proven to be exactly what I need.

@jerrybaxter Could you share how you did this?
i'm also looking for the same solution, but I'm unable to find any examples (wiring / airflow) of how to do it with the Piezo sensor.


The basic concept I have found to work for me was to mount the Piezo Transducer in front of the air flow, connect the positive to an analog pin and negative to ground and then analogRead(pin). Air pressure stops the Piezo and it will to read 0 otherwise. If you add a resistor in parallel to positive and ground you can reverse this and get an increase in signal upon pressure(like how lights follow music).

You should not connect a Piezo directly to an Arduino input. You will get both positive and negative voltages out of one and they can be over 100 Volts. The resistor damps that down, and the diode (any type) prevents reverse voltage.

Yes it might work for now but it will fail after a time if you don't do this.

I'm sure you know more than me about this stuff, but I have seen similar diagrams that have been argued to whether they are necessary or not.
Ultimately I have tried using a parallel resistor and a diode in various configurations. I have included a series resistor of 150ohms to limit any current, but I want the noise from ground in my circuit and the parallel components taking it out. I am not supplying any voltage to the Piezo and at the moment seem to be getting the results I want(at my own risk I guess).

Well I have oscilloscope traces showing both excess voltage and reverse voltage. Not only that I have had projects fail in the middle of exhibitions, fail even with all that circuitry except the zener,

There is a lot of crap out there, and deluded people who think if it works it is fine. But they never use a project enough to have it fail and never look at the actual traces they get on an oscilloscope.

What happens is the ESD diodes kick in and clamp the voltages from the Piezo, but this is way too much of a spike to be repeated clamped and eventually they fail somewhere along the line.

You better believe it.
I have worked with electronics since I left school at the age of 16, I am now 70.

I have worked in all sizes of professional companies, from a place with premises made from a knocked together row of houses, to those with multiple sites all over the world. I have also worked as a University lecturer (in the US this would be called a professor). As well as having had six books published and written in excess of 400 magazine articles. Quite honestly I have lost count of the actual number of these.

However, it is clearly up to you what you do. I am just pointing out your options derived from my personal experience.

You can't repeat this enough. The ESD diodes are not there to protect the inputs from user-applied overvoltage, they're there to protect against ESD. If you need overvoltage protection add your own diodes.

I will accept that though I was not currently experiencing issues, it doesn't mean it wouldn't happen. I have attached a diagram of my potential alternative setup, could you advise if this design would prevent issues? I have a resistor to ground because the pin has a resistor on the pin and otherwise the signal only shorts to ground.