Has anyone had any luck using Glonass correction on early VarioGuide using AGI-4?
It works fine with 1004, 1005, and 1033.
As soon as I add 1012 it drops all RTK. It then tries to reconnect, occasionally fixing for a second then dropping again.
I wondered if the 1033 was missing a critical descriptor but surely the receiver would just ignore the Glonass correction if it was not correct?
I have been successfully using this message set on RtkBase with my AGI-4:
1004(1),1006(10),1008(10),1012(1),1033(10)
As above, you need to add 1008 to let the rover know the base is not a Topcon base that uses an incorrect interpretation of the RTK message format.
There is a way to configure the rover to work without 1008. An expert from AGCO factory visited us 10+ years ago and did this but I don’t know what was changed. Overwritten by a later SW update and I now run the base with 1007 and 1008 added (with proper antenna information).
Many, thanks for this.
Is the timing of 1033 and 1005 in relation to the other messages important for AGI-4?
As 1008 and 1033 are being generated by snip any idea what i could put in the base information tab?
Michael.
Are you using a paid version of SNIP? I have no experience about that, only using Stefal RTKbase where RTKLIB builds the 1008 message (usually with the ADVNULLANTENNA description). I’m also using my Novatel base that builds the 1008 message natively.
Yes i am using the paid version to generate Legacy RTCM for older receivers.
Have done a bit more research with message timing but makes no difference.
Works fine with 1004,1005,1008, and 1033.
As soon as you add 1012 it drops all corrections.
I can only assume that the AGI-4 is looking for some information in 1033 or 1008. Still mystified as to why it’s happy with gps but gives up as soon as i add glonass.
Sounds like your 1008 is not correct. How does SNIP itself decode the 1008 content? You might test with 1007 too (both 1007 and 1008) if there is some SW release of AGI-4 that needs that 1007 but most often 1008 is enough, also for many other brands that do not work without 1008.
No matter with the order. It anyway changes when most messages are sent infrequently and at a different rate (see how sk239 had them configured).
It must be somehow clumsy implementation by Topcon. The Glonass satellites do not fit into the RTK fix with the incorrect interpretation of the correction signal. Instead of ignoring satellites that do not fit with the GPS based fix, Topcon drops all off. Then again, you would want all Glonass satellites all the time which works as long as AGI-4 adapts to the base brand.
Have tried 1007, with and without ADVNULLANTENNA ,no different.
Tried my neighbours MSM base using snip to translate to legacy RTCM, exactly the same issue, which i guess proves my base is ok,(works fine with msm.)
Begining to wonder if there is an issue with the AGI-4 that i am testing.
Be interesting to know what information sk239 is sending with 1008.
Yep I’m using ADVNULLANTENNA in the antenna info part on Stefals RtkBase. Also left receiver options as the default: TADJ=1.
I actually had a very similar problem with it dropping glonass and losing fix until I plugged my base directly to my router. I had it in a different part of my yard before using a Ubiquity Unifi mesh network which just caused all kinds of grief with it.
After many hours testing various configurations of 1033 and 1008 to no avail, I’ve found the issue. For some reason snip does not convert 1074 to 1012 correctly. After changing F9p to 1077 it works straight away and the AGI-4 is processing GPS and Glonass.
2 Likes
Should read 1077 and 1087.
Just to make clear to myself 
. What are all the messages you send to get it working? I’m struggling with Fendt Topcon AGI-4.
1 Like
For my AGI-4 and RtkBase I’m successfully using:
1004(1),1006(10),1008(10),1012(1),1033(10)
Thanks! I will try that set of messages. I have started to suspect that my base station location is not accurate enough, if that has any effect.
The issue doesn’t seem to be the messages. Still not getting full accuracy. Did put the file service on in the base to have a go with a new base location. Any other ideas where the problem could be?
Neighbours mountpoint gives full accuracy within seconds (STR;Ventala;Isokyro;RTCM 3.3;1004(1),1005(10),1008(10),1012(1),1019(3),1020(2),1033(10),1042(3),1046(2),1077(1),1087(1),1097(1),1107(1),1127(1),1230(30);2;GPS+GLO+GAL+BDS+SBS;SNIP;FIN;62.97;22.30;1;0;sNTRIP;none;B;N;11580;). With the full accuracy I mean 5 bars visible in varioterminal which indicates the accuracy an stability to my understanding. My base station with same messages gives only two bars, sometimes three 
.
I guess gps+glo should still be ok and rtcm versio should’t cause this either.
For some reason that I have never understood I have to leave out 1012, try it.
So I have 1004, 1006, 1008, 1033.
Have tried using 1012 with or without 1230, but it just caused problem’s .
Non Agco AGI-4 s work happily with 1074 and 1084.
Tried to leave out the 1012, but it didin’t solve the accuracy. Still 0,10m in the best case and three or less bars.