Time Stamp conversion Development Desktop 2.0

[Ben]

Hi all,

I'm using a BME680 shuttleboard with an Appboard V2.0.

As software I use Development Desktop 2.0 (DD 2.0).

I'm testing the board via DD 2.0. The data is saved as a textfile by the DD 2.0.

I want to export the data to plot graphs (for example in Excel). My problem is, that I'm not able to figure out how to convert the time stamps in the textfile to the actual time/date. I tried to use Unix time converter but i didnt work.

So my (rather simple) question is: What format are these time stamps written in? And how can i convert them?

Thank you for your help

- Ben

[FAE_CA1 · Accepted solution]

Hi,

Thanks for your inquiry.

The unit of the time stamp in BME680 data log CSV file through DD2.0 GUI SW is nano-seconds. For example, when BSEC library works under low power mode (LP), the IAQ value and other sensors' measurements update at every 3 seconds. Under ultra-low power mode (ULP) the update rate is 300 seconds. So you can simply ignore the time stamp values and just use 3 seconds for each row data plot in the CSV file. Please see the attached "BME680_data_log_DD2.0.xlsx" as an example that was renamed from the CSV file.

[Ben]

Hi FAE_CA1,

I tried your solution and ignored the nanoseconds and instead used three seconds for each measurement.

This works. However, when I collect data over a prolonged period (~50 hours --> ~2 days) there is a problem with this conversion. As the timer for each measurement isn’t exactly 3 seconds but a little bit more (3.0X seconds) there is a slight offset which builds up over time.

In the table below you can see that this offset results in a timing mismatch (delay) of almost 8 minutes after two days of continuous measurement.

I want to correlate the data of these measurements with other sensor data for comparison. However, as the sensor data from the BME-680 is “too slow” the offset leads to a bias in visualization with increasing time if I plot all the data of different sensors in one graph.

Is there a way to solve this without using the exact numbers (in nanoseconds) generated by the sensor?

Total measurment time:	54,74 hours
Nanoseconds (measurment date)	Seconds (calcualted from data)	Seconds calcualted with 3 sec for each measurment	Difference in measured seconds and calculated seconds (using 3 sec for each measurment)
197083392063232.0000	197083.4 sec	196605.0 sec	478.39 seconds --> 7.97 minutes