I'm experimenting with electronic wind instrument. I have found some similar projects but all of them are using sensors that are not available in my closest electronic parts shop (Elfa Distrelec in Europe).
So I tried to apply some calculations and choose from options available to me, but still not sure if I'm not missing something.
Basically, my choices are as follows:
Honeywell ABPDLNN100MG2A3 - I2C 3v3 digital sensor, measures gage pressure 0 mbar to 100 mbar, barbless port
Honeywell ABPDANN005PG2A3 - I2C 3v3 digital sensor, measures gage pressure 005PG - 0 psi to 5 psi, barbed port
Specsheet is here:
Freescale MPX5010DP, analog, differential, 0 to 10 kPa (0 to 1.45 psi), 0.2 to 4.7 V Output
Specsheet is here:
Wikipedia reports that 9.8 kPa = 1.4 psi is lung pressure that a typical person can exert. But I'm not exactly sure how it will relate to blowing into a small silicone tube.
Also the fact that one of Honeywell sensors is measured in mbars is confusing. Why did they suddenly use mbars instead of psis?
Google gives that 100 mbars is 1.4 psi, but I'm not sure if there are no caveats here.
Which one might be better for my project and why? Has anybody used one of these and how do they react to human breath?
I've used the Freescale MPVZ5010GW7U for something like this. I was logging pressure while blowing glass. It has a 0-10 kPa measurement range. I think that 9.8 kPa is maybe a little low. I am looking at some data where I hit 10 kPa without specifically trying to blow as hard as I could. On the other hand, we can get higher pressures while blowing only from our mouth with the trachea closed off instead of direct lung pressure and likely I was doing that.
I think that the first decision is about compatibility with current standard measurement procedures. If compatibility is desired, the physical layout of the device becomes important, and the results must be compared to a standard device. In either case the measurement IMO should not only work for healthy persons, but also and in detail for persons with COPD or other disease.
Oh, I posted the same on reddit and one person pointed out significant issue - in electronic wind instruments, we actually do not blow with full force but have a small hole for some air leakage.
So, actually the sensor should be able to measure not full lung pressure but air velocity when blowing, and this might mean that the real pressure on the sensor will be much lower.
I guess, less then 1psi (4 kPa) might be expected. I saw someone mentioning MP3V5004GP 0.5 psi as being good enough for the task, but unfortunately my local store doesn't have it.
My closest available alternatives might be Honeywell again:
ABPLLND060MGAA3
60 mbar = 0.8psi
analog, 3V
ABPMAND001PG2A3
1 psi
digital, 3V
And even those might be not sensitive enough and will require harder blowing than a recorder. I just found an important bit of information:
"The air pressure in a recorder mouthpiece varies between 200 and 1000Pa (0.2kPa, 1kPa or 0.03 psi - 0.14 psi) depending on the note played with high notes having more pressure. The difference in pressure between pp and ff (loud and quiet) is about 200Pa.
An exhaust hole with 3-4mm diameter gives this kind of pressure on the sensor and feels quite nice to play."
I might be able to adjust air force to avoid overloading the sensor using the "leak hole".
Then mbar sensor probably has internal scaling so that it reports directly in mbar , sort of the way you can get temperature sensors that read either F or C.
DrDiettrich:
I think that the first decision is about compatibility with current standard measurement procedures. If compatibility is desired, the physical layout of the device becomes important, and the results must be compared to a standard device. In either case the measurement IMO should not only work for healthy persons, but also and in detail for persons with COPD or other disease.
This is a very strange reply considering this is someone's Arduino project of a DIY electronic musical instrument. No need to be concerned about "standard measurement procedures" or ADA compliance.
progmars:
unfortunately my local store doesn't have it.
I'm just amazed a local store would have any of these sensors. You're lucky! There isn't anything like this sold within 300 miles of me. One sensor you might consider is the Bosch BMP line of barometric pressure sensors. The reason being they are very cheap and very common, not because they necessarily are in the perfect pressure range. I have seen them used for a general pressure sensor. I haven't worked with one because my applications required a sensor I can attach a hose to and I decided it was too much work to try to hack that together for a BMP sensor instead of just paying extra for one with a hose nipple. But that Freescale sensor I used was $17, compared to $1 for the BMP180!
For measuring the blow pressure you need a differential sensor, like the MPV50xx sensors (also in 3.3V versions). Note: they produce an analog output. They also do cost a bit more than the BMP180, which is a barometric pressure sensor, so you'd need two of them as you probably have to compensate for ambient pressure.
1/ A trumpet player going for a high note can generate about 0.3 atm or 30kPa. Hard work! Oboeists blow hard as well.
2/ You can solder or glue ( if you're careful ) a small pipe to a BME280 and it works fine. Is it worth bothering about the small meteorological changes in absolute pressure? If so take a quick calibration reading before the experiment begins ...
I've used them for making a carburettor balancer for old multicylinder motorcycle engines.