Driver for BME280 and nRF51 DK.

[filip_p]

Hello,

I am trying to make BME280 sensor work with nRF51 DK. I just need temperature readings. So i searched if there was some examples using this sensor and i found this example.

https://github.com/knuta/nrf5-bme280-twi

I have tried the example and i get only one fixed temperature value on UART. Here is screenshot.

The buffers with measured data are also not changing when i debug.

I have checked the registers that are used for configuring the sensor and registers for reading, they seem to be ok.

What could be the issue, could you give me some pointers?

Setup:

-IDE: Segger Embedded Studio

-SDK: nRF5 SDK v12.3.0

-Board: nRF51 DK (PCA10028)

Thanks

/Filip

[filip_p]

The project works good in Keil, but in SES it runs but the UART output stays fixed "Temperature: 23.2d....".

I followed this guide for porting Keil projects into SES.

https://devzone.nordicsemi.com/b/blog/posts/segger-embedded-studio-a-cross-platform-ide

[Jørgen Holmefjord]

Please check if this suggestion fix your issue. Also try upgrading to latest version of SES if you are using an older version.

[filip_p]

The logging is done like your suggestion. While debugging, the variable value stays the same. I have the latest SES version 3.34a installed ~10 days ago. 

I have also tried building and running the example with GCC, the output is OK, it works good, temperature value does not stay fixed.

SES produces 45KB hex file, while gcc produces 42KB hex file. So from this i can assume that SES and GCC does not produce same output executable.

[Jørgen Holmefjord]

Did you check that the settings mentioned by Knut Auvor Grythe is set? (this was what I was linking to, not my post)

I posted this on their forum earlier (see forum.segger.com/index.php, and got a response telling me to enable "options->code->printf/scanf->Printf Width/Precision Supported".

It is not expected that two different compilers should generate the same binary output. This will depend on how the compiler works, optimization setting etc.

[filip_p]

Thanks for the reply,

yes i did try that, i suppose that solved the bad printing on UART(which had "d" at the end of the numbers), but the main problem still remains which was that the temperature variable value is fixed. 

[Nguyen Hoan Hoang]

Have you compared the calibration table between both versions ? The calibration is read from the BME and that is used to calculate the temp.  You may have wrong values.