search

graycatlabs / PyBBIO

A Python library for Arduino-style hardware IO support on the Beaglebone
https://github.com/graycatlabs/PyBBIO/wiki
MIT License
251 stars 89 forks source link

BMP183_test.py does not give correct pressure output #79

Closed EngrPradeep closed 8 years ago

EngrPradeep commented 8 years ago

bmp183comparison I am not able to get the correct output from the BMP183 pressure sensor. However the temperature readings are correct. But if I use the Adafruits library using on arduino for the same connections, same sensor, I get the correct readings. I tried to modify the code in the similar fashion as Adafruit library but it still gives me the same (wrong pressure reading) output. I am using BBB with python 2.7 on debian.

Any suggestion will be very much helpful.

Thank you in advance!

EngrPradeep commented 8 years ago

Apparently looks like there is a bug in the BMP183 library. The program runs and gives the output even after BMP183.py file removed from the BMP183 library folder.

I guess there is some error in the following lines of BMP183_test.py file:

from bbio import * Import the BMP183 class from the BMP183 library: from bbio.libraries.BMP183 import BMP183

Still trying to figure out how the code is written. Any help will be greatly appreciated. (I am novice to python and BeagleBone)

alexanderhiam commented 8 years ago

Sorry, haven't had a chance to look into this yet, but will check it out ASAP. Not sure what's going on, I'm pretty certain I was getting correct values when I tested it, or else I wouldn't have pushed it.

When you install PyBBIO it installs the package to /usr/local/lib/python2.7/dist-packages/PyBBIO-*.egg, so you'd have to delete the file from there. If you did that then it would throw an import error.

Mausy5043 commented 8 years ago

I have the same problem. I get perfectly normal temperatures. But the pressure is at 25700 Pa (nearest waetherstation says 1027hPa) and I'm pretty sure I'm not almost out in space :wink:

I'm on a BBB using /usr/local/lib/python2.7/dist-packages/PyBBIO-0.10-py2.7.egg-info

Mausy5043 commented 8 years ago

I've done a very quick regression of my data and the numbers from @EngrPradeep and found this relation:

P = -0.1 * p +105000  (roughly)

Where P is the expected pressure and p the reported pressure. The gain and offset look interestingly familiar :wink:. Hope this is helpful.

I'm trying to grasp your code; is this all the code that is used to communicate over SPI and get the result? Or is there more? EDIT: yes it is :grin:

Mausy5043 commented 8 years ago

I think I found the problem.

The calculation of B3 is incorrect.

If you follow the calculation using the data from the datasheet your code results in a value for B3 which is off by a factor of approx 4.3.

The formula on the datasheet reads exactly the same as in your code: b3 = ((self.cal_AC1*4+x3) << oversampling + 2) / 4. The interpretation by Python is to "shift left by oversampling+2" But it should be interpreted as "shift left by oversampling then add 2".

If I change the code to: b3 = (((self.cal_AC1*4+x3) << oversampling) + 2) / 4 the result is correct.

Mausy5043 commented 8 years ago

Proposed a PR (#96) to fix it.

alexanderhiam commented 8 years ago

Merged - thanks @Mausy5043!

Mausy5043 commented 8 years ago

No problem. :+1: