search

j-kaltes / Juggluco

Android app for Freestyle Libre 1,2 and 3 and Chinese Sibionics sensors
GNU General Public License v3.0
210 stars 54 forks source link

Bluetooth continuous data off by a constant from scanned data #120

Open sijanec opened 8 months ago

sijanec commented 8 months ago

Screenshot_20240204_234348_Juggluco

Hello!

I first paired and activated the sensor (Libre 2) with the hardware receiver (FreeStyle Libre 2). It had wrongly set date and time (a couple minutes in the past), so I assume the sensor now has wrong time for its entire lifespan. I assume that's the reason why 5 minute measurements (History) are off by around 10 minutes. But what interests me the most is that instant measurements (received via bluetooth) are always off (lower) by around 0,4 mmol/L compared to manualy created scans. Scanned values (cyan dots) have the same values as values received by the hardware received (FreeStyle Libre 2), but the magenta trace is always quite lower and I have no idea why. Do you have any explanation for this by any chance?

Please see the attached screenshot.

Regards and thanks for your help Anton

j-kaltes commented 8 months ago

Measurement time is determined relative to the current time. As long as you don't change the time (in GMT) of the device Juggluco is running on, there will be no problems. The history values of Freestyle Libre 2 sensors are 15 minutes apart. History values are only known of the past, they are always more than 20 minutes old. The only thing remarkable, is the difference between the scans and stream values. If you send the data via a mirror connection to a device off my, I could look at it. If you want that, I send to your e-mail address how you need to configure the mirror connection.

sijanec commented 8 months ago

The issue resolved itself after some hours. Scans are now consistent with automated minute readings. Thank you for the quick response, I think it's safe to close this issue.