1
\$\begingroup\$

I'm implementing the BME280 (datasheet here) into one of our systems. My coworker swears that pressure sensors are orientation dependent. I've found from online documentation that this is true for large flow-rate pressure sensors, but I haven't found any mention (searching for the terms 'orientation','offset' and reading the qualitative descriptions of the pressure component) of this particular sensor needing to be in a specific orientation.

Has any one found that the BME280 or similar sensors are orientation dependent, such that it causes a >1% difference in measurements between two extremes?

\$\endgroup\$
  • \$\begingroup\$ I frequently see this type of recommendation from engineers, but I haven't seen it from manufacturers on "open" pressure sensors such as this one. I have seen it on "ported" pressure sensors (ones that plumb into a system), especially when there is flow in the system. I think the only concern with this type of sensor is protecting the vent hole from direct moisture, light, etc. The easiest way to make that happen is to point the vent hole down or sideways, but you can also protect it in other ways. \$\endgroup\$ – Tom Brendlinger Aug 9 '16 at 18:26
  • \$\begingroup\$ The smaller the sensing element the less likely it is that the diaphragm mass will affect the reading. MEMS units probably don't suffer from sensitivity. A load beam and diaphragm unit I have lists the error in the data sheet if you mount it inverted. This one makes no mention in the data sheet but as mentioned a simple inverting test will demonstrate any effects. \$\endgroup\$ – KalleMP Sep 22 '17 at 7:25
1
\$\begingroup\$

A pressure sensor can be orientation-dependent (and acceleration-dependent too, by the way).

An practical way to test for orientation dependency is to compare measurements at different orientations: horizontal face up (1g), horizontal face down (-1g), vertical (0g). If you want to be extra-vigilant, you could repeat these measurements at an elevated temperature. If the datasheet doesn't mention the orientation dependence, and you don't see significant deviations in your measurements, then you are probably okay.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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