Hi all,
We’ve modified the source code to fix the issue of Longitude avariable, and i think it mai be fixed now according to our testing.
GPS was a bit too far east on the earlier firmware. I’ve managed to get a good hit on my house now 
Thanks!
Thanks @fomi,
The downloaded firmware T3, is after flashing named: RAK7200 version:3.1.0.7.T2
Currently under test. 
BR,
Jeroen
Hi @leopold ,
Are you sure as the TTN console is not Cayenne. Cayenne will only use the LPP data forwarded from TTN server.
First LPP need to support it, I’ll investigate and get back to you asap.
regards,
Jeroen
Hi, Some testing toady here with T1 version. There is progress as I can see!
here is the same route today by the same two trackers I used last sunday:
Reference tracker:
RAK7200 tracker:
I will now program T3 version.
Hi Fomi,
Below the device status in Serial Port Tool with your latest firmware:
version 3.1.0.7.T2 (T2, not T3 as .bin filename)
LPS22HD sensor is closed or not found
The MPU9250 sensor (indoor and stable on my desktop):
Acceleration(g) of X,Y,Z: 0.02,0.07,0.99
Gyro(degress/s) of X,Y,Z: -1.61,0.45,0.71
Magnetometer(uT) of X,Y,Z: 28.65,52.95,55.05
The indoor GPS position is accurate (The Netherlands) and pretty constant over time. Altitude value is constant but wrong, this is a common problem with all GPS modules I have tested so far.
Best regards, Hans.
One nit is that V3.1.0.7.T3 identifies as T2 in at+version.
The GPS data is now reporting and correct.
A feature request would be a GPS tracking mode where the device would send a packet every X seconds if the GPS position has not changed more than a given distance. If it did change it would send a packet every Y seconds. This would be useful when trying to provide data to ttnmapper.org.
Hi Jeffrey,
Like the idea but constant checking distance changes (or speed) will drain the battery I’m afraid.
Best regards, Hans.
I should say, as already provided some posts ago:
wake up> check previous GPS coordinates > movement? >
as an addition, after 30 min of no movement, transmit data once
I would envision sending a packet every 600 seconds unless the position changes, then send every minute. That won’t be much different than sending a packet once a minute.
Sending a position every 5 minutes if you are not moving vs every minute if you are not shouldn’t be significanly different change over sending a position every minute.
…or make it AT configurable as it depens how you want to use the tracker.
I do not need an update if there is no movement but like to have it in 30 min to know if the tracker is alive and to know the battery status.
Agreed, sorry I didn’t explicitly say this should be a configurable mode. It would be great for ttnmapper.org, but less useful for other things.
Dear @fomi,
For me there is no difference with regards the GPX fix. This is the data from the tracker, no movement, clear sky:
and TTN mapper:
It needs further investigations.
BR,
Jeroen
Hi,
My results with GPS accuracy:
So far no problems, knock on wood… Just wondering what the Cayenne SNR widget is about (left upper corner), anyone?
Best regards, Hans.
SNR, signal to noise ratio
After running 3.1.0.7.T2 for about 24 hours I see errors similar to those seen by Jeroen.
The longitude errors are clearly in the LPP data and not in the conversion of that data.
It is possible that the data from the GPS module is being misinterpreted and assuming that the tracker software is parsing the module’s NMEA sentences it is imperative that the software calculates and verifies the checksum of each sentence. Also if the GGA sentence is being used to check the “fix” field for validity.
BR,
Charlie
Dear @fomi,
Can you investigate in depth? The fix in T2(3) didn’t help.
Almost got again same ‘drawing’ this morning 27-9:
BR,
Jeroen
Hi all friends,
We’ve just released a T3 version:
https://github.com/RAKWireless/Update-File/blob/master/RUI_RAK7200_V3.1.0.7.T3(2).bin
Please check.
This version add a GPS data CRC filter.
