Is it possible to enclose the LPS22CH sensor in a separate housing and to generate a negative pressure via a hose connection so that the sensor serves as a negative pressure sensor?

[AMaro.2]

..

[Peter BENSCH]

Welcome, @AMaro.2​, to the community!

Pressure sensors always measure a pressure difference, either between two separate connections (differential pressure sensors), or against a reference pressure inside (absolute pressure sensor). ST currently only offers absolute pressure sensors that can be used for 260/300...1260hPa.

So the answer to your question is: no, it is not simply possible to measure negative pressures with an absolute pressure sensor.

Regards

/Peter

[AMaro.2]

Thanks for the answer!

Yes, I understand it. My idea is to use two sensors. One senses the ambient pressure and the other generates a negative pressure. The difference between them is the relative pressure.

I mean I did the same with the Bosch BMP280 and it works.

The problem with the LPS22CH that I am currently using is that if I create negative pressure on the hose connection, does the value not change?

[Peter BENSCH]

Oh I understand.

This case is of course also possible because you use it to determine the calculated pressure difference between two sensors.

The second sensor can then measure the negative pressure as long as it is in the absolute range of the sensor, e.g. 260...1260hPa. The calculated difference between the two sensors then corresponds to the negative pressure of the second sensor.

Regards

/Peter

[AMaro.2]

My question would be: why doesn't the value of the sensor change when I create a negative pressure on it?

I query the three registers  PRESS_OUT_H (2Ah), PRESS_OUT_L (29h), and PRESS_OUT_XL (28h) as described in the "interpreting pressure readings".

The value shown to me is approximately 1020 hpa. It only changes with the temperature and has a slight fluctuation, which according to the data sheet is only sometimes.

The temperature value is displayed correctly.

[Peter BENSCH]

The sensor has no nozzle or hose connection, which is why in your case it has to be exposed completely, i.e. the entire sensor, to the negative pressure.

What is the mechanical structure of the chamber in which the sensor is located and where does the negative pressure act there?

Regards

/Peter

[AMaro.2]

The structure is provisional. I press the hose against the circuit board and suck on the hose. I feel that the negative pressure is not escaping.

[Eleon BORLINI]

Hi @AMaro.2​ ,

are you sure you are sealing well the tube on the adapter?

If not, the air pressure will be almost equal to the ambient one.

You might try to seal the tube on the adapter with hot glue, if you can.

-Eleon

[AMaro.2]

I'm actually pretty sure that the pressure of the hose against the plate is tight.

For a simple query I only have to configure the CTRL_REG1 (10h) register to at least 1 hz, i.e. write a 0x10h in the register. And then the query of the three registers PRESS_OUT_H (2Ah), PRESS_OUT_L (29h) and PRESS_OUT_XL (28h) and divide by 4096?

I mean if I check the WhoAmI register and the temperature and the values are correct, I don't know what I'm doing wrong with the print?

[AMaro.2]

Hello, everyone.

At the weekend I got a STEVAL-MKI109V3 board on loan with which I can use the same STEVAL-MKI219V1 sensor.

Starting the UNICO GUI software is successful and the sensor is also correctly recognized.

The plot shows the temperature and pressure.

The temperature changes as soon as I put my finger on the sensor.

Now to the negative pressure: I started the same experiment as I did before as shown in the two pictures.

This time to confirm that the system is tight, I hung another vacuum measuring device in the line.

The negative pressure on the second measuring device is displayed correctly.

Unfortunately not from the lps22ch?

The negative pressure value only changes when the temperature changes.

Does one of you have contact to the chip manufacturer?

I would like to find out where the problem with the sensor is.

Because according to the data sheet, this is how the application should work.

It works with the BMP280 sensor from Bosch, which has similar properties.

[AMaro.2]

To confirm that my application is tight, I installed the board with the sensor in a separate housing.

Unfortunately, the value does not change in the Unico GUI.

The GREISINGER measuring device shows a negative pressure of approx. 78 mbar in the housing.

Thus the value in the Unico GUI should show approx. 931 hPa.

[Eleon BORLINI]

Hi @AMaro.2​ ,

thank you for the details and the pictures of your application.

Actually, as reported by @Peter BENSCH​ , the minimum pressure that the LPS22CH can measure is 260 hPa, or equivalently 260mbar. Does the 78 mbar refer to absolute pressure or to "ambient pressure - 78mbar"?

Did you try with pressures of, for example 500mbar? And did you have the possibility to test more than 1 device?

I also suggest you to try submitting the parts to baking process.

-Eleon

[AMaro.2]

oh yes I forgot to mention that the 78mbar are relative.

I have two of the sensors.

Which also amazes me that the two sensors will display different ones when idling.

one sensor shows 1021 hPa absolute when idling and the second sensor 996 hPa.

I don't understand why the values show such a large difference.

But nonetheless, when I create a negative pressure, both show no change.

[Eleon BORLINI]

Hi @AMaro.2​ ,

I was made aware of the OLS case by internal colleagues, and they'll in case suggest other.

Are you able to see differences in pressure when you, for example, blow on the device?

Or are the device stuck?

-Eleon

[AMaro.2]

when i don't blow but when i breathe, i.e. warm air, the value changes.

As I have already described before that the pressure only changes when I put my finger on it and the sensor assimilates my body temperature.

the device does not hang because the temperature changes yes.

[Eleon BORLINI]

Hi @AMaro.2​ ,

so it looks as the pressure is varying only with temperature, for example when the chip is heated.

This would indeed resemble a "contamination" issue like clogged holes, as suggested by OLS team.

Please refer only to them for the support, and please come back here once the ticket is closed.

-Eleon