RTKBase: a GUI for your own Gnss Base Station

Yeah, of course, the coordinates are fixed after the survey.

I tried to get a fix down to MM and failed when doing the survey for a week. Wonder if NTRIP corrections may have helped. Probably not.

I believe you can input corrections into the base but if that other basesā€™s position isnā€™t precisely surveyed then it doesnā€™t help you much.

1 Like

I just set my base surveying in to a basically unattainable level then left it until i got fed up of waiting. About 35 ish hours i think. Then just photograph the coordinates in u-centre and input them. Was down to less than 50mm iirc. Remember this doesnā€™t affect your rover accuracy, just itā€™s position relative to other bases and real world coordinates.

1 Like

Would be useful to know what was the RTCM 1008 message content when it did not work with Topcon (AGI-3) and what was the message content when it started working?

I run RTKBase (obviously with F9P) and I just set the antenna type to NULLANTENNA or ADVNULLANTENNA and this works fine with TOPCON AGI-4. Since AGI-4 only supports GPS and Glonass, I use legacy RTCM messages. I also set receiver options ā€œ-TADJ=1ā€.

At least with legacy messages the (older) Topcon receivers (I guess the same with Trimble) need to know if the base station receiver is made by Topcon, Trimble or any other company. This way the receiver can adapt to the different interpretation of RTCM specs.

Thatā€™s an easy way, but if you move the base and re-survey the new base location the same way then your global reference could be out by 0-100mm, and all your existing saved guidance lines and boundaries will shift that much.

With RTKBase you can log the position readings to a text file on the sd card. Then you can post process the file with a tool like CSRS-PPP to correct for any satellite drift. Thatā€™s about as good as its going to get.

Whenever I have to move the base station, I use RTCM from the old base station to find the precise relative location of the new base station, and then manually put those coordinates into the base when itā€™s set up at the new location. That way everything stays relative.

Iā€™ve never bothered with post-processing, and Iā€™ve never done a survey-in longer than 10 minutes. It just doesnā€™t matter to me that much. Getting coordinates that agree with other RTK systems is always going to be a challenge and involves dealing with GIS coordinate systems (NAD83 in my case) to compensate for the fact that the land is always moving. Itā€™s kind of interesting that Google maps is as close as it is, dealing with all that uncertainty. The satellite systems (SF3, RTX) cover wide areas. I wonder what coordinate system they use.

3 Likes

which, on the scale of what we do is absolutely nothing and, why would I move the base?

The narrowest tyres on any machine I use are six times that distance.

If you give a 24h log to the OPUS service, it should return the coordinates in NAD83. Isnā€™t it ?

Oh sure. Probably. But Iā€™ve never had the need to do it.

Just dug out the code for the 1008 message. It was just a dummy 1008 and this actually may well be why it wouldnā€™t work before. RTKBase is sending exactly what you are, along with the receiver option.

However, it didnā€™t work with the default output of RTKBase which was :-
1004,1005(10),1006,1008(10),1012,1019,1020,1033(10),1042,1045,1046,1077,1087,1097,1107,1127,1230.

1 Like

Thanks a lot for that information. A pity if nobody has any idea why the AGI-3 did not work from RTKBase with the default set. Not critical as youā€™ve found a working approach.

Iā€™m only sending the mandatory legacy messages for the AGI-4, 1004, 1012, 1006, 1008 if I remember right. I tend to remember others sending both legacy and MSM successfully. I think someone claimed too high message flow confusing AGI-4 (or was it some other brand) but Iā€™ve never tried to test that.

Following our conversation I might change the messages to those actually. My RTKBase caster is already sending those for my friend with the AGI-3 to test. If I can grab him for an hour or two Iā€™m going to see if I can narrow down what really stops it working.

I donā€™t quite understand, I have a server with a fixed IP, how should I install it, I downloaded the source code, but I donā€™t know how to install it

The opening message of this thread has links to instructions.

Is there a way to send different base coordinates to the two different Ntrip-services?

I know, this sounds very strang and stupid but this would be very helpful for me.

Perhaps setting different reciever options?

Thx a lot, christoph

No, itā€™s not possible from the GUI.
But if you have some skill, you can edit the files. This is how I would do it:

  • add a new position2= entry in settings.conf with the ā€œotherā€ coordinates :upside_down_face:
  • edit out_caster_B= to use position2 in run_cast.sh
  • start Ntrip B service from the gui

Thx a lot!
I hope my linux skills are good enough!

Would using a separate memory card for each base location , be a solution?

1 Like

Or two kits if both need to be running simultaneously.