• The ublox 6M 1hz Issue - but why not at least 5hz?

    Timo Birnschein09/03/2019 at 04:15 0 comments

    Apparently, the M6 from ublox can be configured to run at 5hz as well, according to the manual.

    In the center of the screen, one can see, the Maximum Navigation Update Rate can be up to 5hz for the NEO-6M.

    The interesting part is, this GPS receiver actually comes from HobbyKing and was part of a kit I put together to return telemetry through the D8R-II Plus receivers! These things are amazing, have incredible range, fail safes and other interesting features. However, I believed the GPS module must be kinda crap. Finding a ublox under the heat shrink was a surprise and reading it can run at 5hz was extra nice!

    However, I could not figure out how to reconfigure it do actually output 5hz.

    I found an interesting post how to reconfigure this using an Arduino but I didn't have the time to try this, yet. 

    And none of the immensely detailed configuration screens in ublox center seem to have any effect on the configuration of the receiver:

    This would line up with what is later described in the manual for the T and P versions of the receiver (even though they write further up that it works also on M:

    In any case, for my testing I don't want to use the binary mode as this makes testing with universal software difficult. So 1hz it is for now...

  • Configuring the 10hz GPS Receivers

    Timo Birnschein09/03/2019 at 03:47 0 comments

    Two of these receivers are pretty bloody cool: They can run at 10hz which makes drones a lot more stable and specifically in GPS mode a lot more accurate to control.

    I used a couple of different softwares to monitor and configure them. 

    The goal is to run them at 115200 baud and at 10hz update rate.

    • Ublox Center
    • Mini GPS Tool 1.7.1 (pretty much useless)
    • Mini GPS Tool 1.4 (can configure the chipset baud rate - in order to do this manually one would need the checksum which is not mentioned in the configuration command manual)
    • hterm (I use this to confirm I have the correct baud rate and also to manually send config commands to GPS receivers)
    • Visual GPS View (It has great visualization capabilities to show how good the localization really is)
    • PMTK Command Packet Manual (Holds all commands the GTPA modules understand. Unfortunately, not all commands have all examples spelled out such that the checksum is clear. I didn't have a checksum calculator handy so I used Mini GPS Tool 1.4 to change the baud rate, for instance)

    In order to configure PMTK receivers, I ran Mini GPS Tool 1.4, 

    Went to the Setup tab and selected the baud rate:

    Please note, the Fix Update Rate can only be set up to 5 hz using this software. Not sure why, it makes no sense. However, using hterm, the command for 10 hz can be sent manually:

    $PMTK220,100*2F<CR><LF>

    Please note also, I added CR+LF under "Send on enter". That will fire the command and the receiver starts sending with 10 hz update rate!

    Power cycle brings it back to its standard settings: 9600 baud at 1hz.

  • Starting off....

    Timo Birnschein09/03/2019 at 03:36 0 comments

    I found myself with 4 different GPS receivers that I always wanted to use and never really got around to it. Specifically, the two expensive ones in the list were about $50 when I bought then and designated for drone applications. For some reason, I never really got any of them to work. I assume I just didn't get the memo that configuring a flight controller is hard. Not sure why. GNSS is not difficult to parse, but none of my 32 bit flight controllers wanted to have anything to do with them.

    The purpose of this project is to collect information about how I configured them, where to find the datasheets and what to expect in terms of localization quality.

    I will run a test on all GPS receivers to determine what the scatter of each chip is and how many satellites it tracks reliable. I need to note here that none if this will be very scientific. I have to test them one by one and can't run all tests in parallel at this point. So the comparison might not be 100% representative but should provide at least an overview of what to expect from these things.