Ask A Surveyor
If you are a landowner seeking advice from surveyors, please post your questions and issues here.... View more
Organizer:
- Organized by
-
Change from CMR to RTCM
Change from CMR to RTCM
Hey Guys,
We have recently had to change from CMRx to RTCM due to new machines that only use RTCM. after many changes to our system (4800 to 9600, 12.5 to 25 spacing, all of which have helped alot) We are finding that RTCM is not working as well through our terrain as CMR did.
We work in a mountainous area, 35km*25km area. We have three bases set up north, center and south.
We also have large overhead cables at most accessible high points of the job.
The issues we are experiencing in those areas are; we have overhead cables, we are experiencing alot more multipath with the RTCM then the CMRx, also needing repeaters in areas we didnt need previously on CMRx and slow starting GPS on our R12 rovers. Also working around and in state forest
We use a SPS885 unit with a Zephyr mdl2, we use a TDL450H or Hx radios for the base on 3/4m mounted antennas. our bases are not set up near the overhead cables, and they sitting at the highest points that are accessible in our area. (not near overheads)
We are also using TDL450Ls as repeaters for our cars where we are struggling.
Im wondering, is the settings and equipment we are using the best for this terrain? We have tried sticking to trimble machines but the company want the cheaper option hence the RTCM running machines. Is there anything that can help with the multipath from the overhead cables and the mountainous area? Is there better equipment we can upgrade to? and is RTCM known for the issue mention above?
Any help would be greatly appreciated
Cheers
- 14 Replies
Interesting question.
Nominally it should not make a difference which correction type you use.
But…
With the change to RTCM you will have lost any ability to track Bediou or Galileo.
You may also have turned off Glonass accidentally.
That might explain the loss of accuracy…
Have you tried switching off the bandwidth-limiting?
Does the RTCM-version factor into something like this?
Is 25 rather than 12.5 a RTCM thing (or is it a SPS thing)?
@jimcox Thank you for the reply, I set my rover to only GPS and Glonass after we changed to RTCM. They are both still ticked on for my rover, but interesting view about a lower amount of satellites. Could well be. Thanks mate!
Edit: Seems to be more the radio signal we are having issues with. Ill check how many satellites I am receiving in these areas. The issues are usually in gully areas, but previously when using CMRx, I never noticed issues if we were in gullys, and I’m still using the same base and radio for these areas. its almost as if RTCM setup is more sensitive to multipath then CMR+.
@fugarewe from what I know, 25 channel spacing allows for better transfer of RTCM data as the data is not as compressed as CMR. I may be wrong, both may work better on 25Khz spacing, but without a doubt, that bandwith worked a lot better then 12.5Kmz for our rovers and our machines as the messages were not getting cut off.
Have you tried cellular? We use cellular for RTK almost exclusively. Sometimes RTCM, sometimes CMRx. I have not noticed ANY degradation when using RTCM.
We are using RTCM3.3 (MSM), and we do get all four constellations, typically between 20 and 30 satellites. So not sure why jim.cox says you won’t get Beidou or Galileo. Probably true with RTCM2.2, I think that only supports GPS and Glonass
Have you tried cellular? We use cellular for RTK almost exclusively. Sometimes RTCM, sometimes CMRx. I have not noticed ANY degradation when using RTCM.
We are using RTCM3.3 (MSM), and we do get all four constellations, typically between 20 and 30 satellites. So not sure why jim.cox says you won’t get Beidou or Galileo. Probably true with RTCM2.2, I think that only supports GPS and Glonass
If you are using RTCM2, yes fewer measurements and less corrections. If you are using RTCM3.3 the amount of data should be comparable to CMRx (I am not a Trimble user though). If you just switched systems, there are many other thing it could be; are you collecting and sending just and only what your rover will use; have you manually picked which messages you a sending; or are you using everything (which is not necessary and will slow things down); only transmit the messages, corrections and satellites you need.
Multipath should be independent of the correction format that is being used. Multi path is caused at the receiver head and exists before the correction is solved and sent. if you are getting multipath it is a function of reflected satellite signals being picked up the receiver. Multipath can casue a bad solution but it does not matter the format the solution is then sent to the rover.
I have seen some not-too-good to bad base locations. If there is multipath at the base, you will have issues at the rover.
Why did the chicken cross the road? I don’t know. Too get better data.
It might be a bandwidth issue. 9600bps might be bottlenecking, you might want to try 19200bps, if that is an option. The CMRx data packets are a lot smaller than the RTCM3(MSM) data packets.
@john-hamilton Hey mate, we are in quite an isolated area, we have tried previously to go cellular but to no avail. Too remote. We turned on the other satellites yesterday and found the r12 rover took over 8 minutes to connect. Im guessing because we turned on all of the satellites. we have clear open space around and above our base! I was very dramatic about the placement of our bases!
@oldpacer We are using RTCM3.3, but I’m unsure of what data I need. Will do some research into this, thank you.
@lurker Cheers mate, makes sense, just found it strange we were suddenly having issues when we changed from CMRx to RTCM, thought maybe the messages were longer therefore maybe more sensitive to obstructions.
@antcrook hey mate, I can look into the 19600, but would have thought 9600 would have suffice, Ill need to go through all the different machines and make sure they are adaptable to it. Ill need to do more research into bandwidths as its something I’m not overly familiar with until we started making these changes!
I just did a test. Tracking 32 satellites, all signals turned on. CMRx is around 450 bytes/epoch, RTCM 3.3 is about 700 bytes/epoch. I know I don’t need all of the signals turned on, but I do anyway since bandwidth isn’t much of an issue over cell.
When we do radio we transmit at 9600, no problem. However, if you throw a repeater into the mix, then you have to have about twice the bandwidth, as it is transmitting twice in the same time frame (1 hz). However, when we do use radio it is usually very short distance, we use it mostly to get woods points for LiDAR QC, set the base in the open, use a rover (R10-2 with ProPoint) in the woods nearby. So no repeater involved, and strong radio signal. Actually, I have sometimes used wifi in those situations. Wifi on the data collector to the receiver running as a wifi access point
I also work in places where there is not good cell (or no cell at all). We have two of our receivers with an RTX subscription. That is not an ideal solution, it requires a much more open sky at the rover. And I recently had a bad experience on the equator (-1°S) where it would hardly work at all for the better part of a day. Hard to get convergence, then it would lose it after a minute of so. Trimble support said probably ionosphere. I had done several prior projects in the same country, never had a problem until this trip (early January). The next day it was much better. That was in an area without RTXFast.
We turned on the other satellites yesterday and found the r12 rover took over 8 minutes to connect.
Im guessing because we turned on all of the satellites. we have clear open space around and above our base! I was very dramatic about the placement of our bases!
That will be the receiver having to download the ephemerides.
Should be be much quicker next time.
I’m tending to the thought that either its the wrong RTCM messages, or that they are being munged. They are more prone to interference. And especially with repeaters. I dont think multipath is the issue here.
I wonder if looking at the age of correction and the QC stats for each satelite might tell us something. There is probably also some logging that can be turned on in the receiver which should show if bad data is being seen
Grab a simple radio scanner and listen to your frequency – you want to be hearing data with regular gaps. also a good check for other transmissions / interference
From your address I’m thinking you are in Aussie. Because of their orbits tracking Beidou can make quite a difference – did they get accidentally (politically) turned off?
One simple thing to check – what constellations are the base stations using – they are no good on the rover if the base isn’t tracking it.
Are we having fun yet?
=Jim=
.
Log in to reply.