1
\$\begingroup\$

Hi all posted question on reddit but giving it a shot here too, I've been working on a board that requires a pressure sensor to detect when there is a differential air pressure or not. The board needs to be quite small, and more importantly very cheap.

Even some of the cheapest pressure sensors are $5-$10 which is simply too much for the application. They also provide analog output dependent on the value of the differential pressure, whereas I only need it to detect when pressure is above a threshold.

I have seen something similar in e-cigarette/vape pens. They use what looks like a unidirectional (vented) electret mic as a static differential pressure sensor to detect when you are pulling. I was wondering if any of you have a better idea of how this works. From what I know, electret mics can only be used to measure dynamic differential pressure (sound) and will not work for static pressure.

My assumption is that to use the unidirectional mic as a pressure sensor they might instead measure the capacitance of the electret. I'm not sure how well this works considering that the ones I have seen have an internal JFET that wont allow for this measurement

Hoping some of you have some better answers as to how the e-cig companies use them for pressure sensors.

\$\endgroup\$
2
  • \$\begingroup\$ What is the exact situation (or, if you must be circumspect then a situation that is in every important respect identical) for this sensor, what is the threshold, and what is the response time? I don't see how anyone can suggest a rational "very cheap" design working on nothing more than "differential air pressure." It might be used on Jupiter for all the information that's available. \$\endgroup\$
    – jonk
    May 4, 2021 at 1:23
  • \$\begingroup\$ Just a suggestion, look at sensors used in home washing machines, they measure the level of water via pressure. I believe they are diaphragm based and cheep. NXP has a line of pressure sensors, I used the MXP series and they are great. Price is volume dependent. \$\endgroup\$
    – Gil
    May 15, 2021 at 1:54

2 Answers 2

Reset to default
1
\$\begingroup\$

Some of the e-cigs use the sound of the rush of turbulent air flowing past a conventional microphone to trigger the ASIC.

Some use a capacitive pressure sensor in the 10-20pF range that changes about 10%. There are ASICs to detect the capacitance change.

\$\endgroup\$
3
  • \$\begingroup\$ Hi thanks, I was aware of the ones that used sound. Most good ones I have seen use a "pressure sensor". This can be easily tested if the e-cig doesn't activate when you blow into it. If it used noise it would simply activate on push or pull. \$\endgroup\$
    – andreas smith
    May 3, 2021 at 20:19
  • \$\begingroup\$ On the topic of the capacitive pressure sensors, I'm assuming it's just the electret without the JFET. The problem is I can't seem to find anything on digikey, mouser etc. selling them without the JFET integrated. \$\endgroup\$
    – andreas smith
    May 3, 2021 at 20:20
  • \$\begingroup\$ They use a capacitive pressure sensor, not an electret. More like a condenser microphone if it was made very cheaply. Probably measured by applying an AC voltage across the sensor. \$\endgroup\$
    – Spehro Pefhany
    May 3, 2021 at 20:27
1
\$\begingroup\$

Smoking is dynamic, isn't it? For example if you detect an event and you do some action for a specified timeout, it can work. So for e-cig just a simple detector is needed (my presumption), not a true measuring device.

If your app. demands to detect a differential pressure over time, I think is not going to work, since the charge will be leaking, thus not suitable for static measurement. But if this is a device with fast changeovers, maybe it is usable.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.