Notifications
Clear all

RTK distance from CORS station

21 Posts
12 Users
6 Likes
793 Views
JamesHistory
(@jameshistory)
Posts: 8
Regular Member Member
Topic starter
 

Hello! How far can you be from a CORS station for RTK?  Thanks 

 
Posted : October 13, 2022 8:06 am
spyropas reacted
MI-Other-Left
(@michigan-left)
Posts: 163
Member Member
 

That's a bit of a "loaded" and broad question.

As with any reference station, the ability to connect between base/rover is primarily a function of the data link (type, quality, etc.). Examples: Radio, Cellular, IP, etc.

The quality of the receivers may also affect the ability to resolve RTK vectors.

So, we need some more information to give an answer within the context of the question.

 
Posted : October 13, 2022 11:06 am
FL/GA reacted
Bill93
(@bill93)
Posts: 9329
Member Member
 

Somebody may come up with some numerical data.  My impression is that a few kilometers is great, 10's of kilometers not so good, and 100's of km pretty loose.

There may be times when distance isn't as important, but sometimes the iono/tropo and even weather effects are not similar across regions, resulting in poor accuracy.

 
Posted : October 13, 2022 11:09 am
OleManRiver reacted

Get the 2023 SurveyorConnect Wall Calendar

OleManRiver
(@olemanriver)
Posts: 741
Member Member
 

Bill93 summed it up along with mi other left.  Say you are 50 miles and have a receiver that can resolve the interger ambiguity at that distance. Well now you have to still figure in best case scenario the H and V and the PPM to get an idea of what your uncertainty error budget is. Thats if the troposphere and ionosphere is identical at base and rover end. The same can be said at 1 mile. But the odds are better at shorter distances of ionosphere and troposphere being similar and PPM is not as big of an influence in your error budget. 

As mi other left stated. Ya need to give us more information details to better help you. Like equipment what accuracy you need to achieve and geographic area your working in because the grass is always greener on other side like not all areas are the same. If you have a base on top of a mountain and you are surveying in a valley or vice  versa. Or a area where it rains and storms a mile down the road and sunshine on other side. Troposphere. Single base and rover assume they are equal. Now in a network this might change because it is being modeled around you even though you connect to a single base aka CORS. 

 
Posted : October 13, 2022 7:39 pm
lukenz
(@lukenz)
Posts: 397
Member Member
 

And even a single CORS at 1-2km doesn't seem to be as good as a matched receiver using radio for RTK at same distance despite the specs. Blame the internet drag on comms.

 
Posted : October 13, 2022 8:58 pm
Jitterboogie reacted
Jim Frame
(@jim-frame)
Posts: 7045
Member Member
 
Posted by: @lukenz

And even a single CORS at 1-2km doesn't seem to be as good as a matched receiver using radio for RTK at same distance despite the specs. Blame the internet drag on comms.

I haven't experienced this, but maybe it's because when I'm that close to my local reference station I always have fast Internet service.

 
Posted : October 13, 2022 9:33 pm

Get the 2023 SurveyorConnect Wall Calendar

lukenz
(@lukenz)
Posts: 397
Member Member
 

@jim-frame 

I suspect your internet is better than ours, NZ is only just out of the third world level of service! Also doesn't help that the nearest site to town (~5km) I'm in is up on a farm so isn't directly connected to fibre but has a radio link to a modern in town and then into the internet.

 

Have still found errors with other single CORS greater than the 1ppm distance error would suggest (even when scaling that up to 95/99% CI's) compared to radio RTK with matched receivers (i.e all Trimble/Leica not mix and match). Maybe our internet providers really are that bad with data latency or perhaps the PPM error isn't so linear over longer (5km+) baselines as weather can be different over those distances here with topography changes etc.?

 

For reference this is for full constellation not just GPS/GLO.

 
Posted : October 13, 2022 11:20 pm
Jim Frame
(@jim-frame)
Posts: 7045
Member Member
 
Posted by: @lukenz

For reference this is for full constellation not just GPS/GLO.

Do you use the same data format for RTN as for UHF?  My understanding is that some data formats are more compact than others, which could affect latency, especially with more than 2 constellations.

 
Posted : October 14, 2022 12:33 am
OleManRiver
(@olemanriver)
Posts: 741
Member Member
 

@lukenz yes sir Latency.  It can be on user end. Server end or both. Nice catch.

 
Posted : October 14, 2022 10:53 am

Get the 2023 SurveyorConnect Wall Calendar

JamesHistory
(@jameshistory)
Posts: 8
Regular Member Member
Topic starter
 
 
the receiver I will be using is a Reach RS2 and the CORS station is a Trimble NetR9. The terrain is flat and and no cloud coverage. I am 24km away from the CORS station. 
 
Posted : October 14, 2022 11:24 am
JamesHistory
(@jameshistory)
Posts: 8
Regular Member Member
Topic starter
 

@michigan-left

 
Posted : October 14, 2022 11:25 am
OleManRiver
(@olemanriver)
Posts: 741
Member Member
 

@jameshistory I would based on a quick look not see why it could not get a fix solution. But i see no specs on H and V values. If you have internet connection and no latency. So say it has Horizontal of 8mm + 2ppm and Vertical of 10mm +2ppm. That APC to APC. You can do the math with correct numbers. In a perfect world you have an uncertainty of hz 8mm +2ppm at that distance. Then you have the accuracy of how well you can plum your rod. The bubble has a spec as well lets say 10 degrees at 2 m now your building your error that is withen your equipment specs in a perfect world. We have not taken one shot yet or allowed for different in ionosphere or troposphere multiple path. Satellite DOP.  But at this stage if you are outside your project specs no need to go further as you need a different method. Through redundancy extra measurements at different times we can build confidence in the things we can’t control like multi path satellite geometry etc. and our in ability to measure perfect.

 
Posted : October 14, 2022 11:36 am

Get the 2023 SurveyorConnect Wall Calendar

lukenz
(@lukenz)
Posts: 397
Member Member
 

@jim-frame 

Using rtcm3.2 msm for CORS and satel 8fsk for base/rover so that is another good point but I don't know enough about the data packets of either to know if that is a factor or not.

 
Posted : October 14, 2022 12:33 pm
lukenz
(@lukenz)
Posts: 397
Member Member
 

@jameshistory 

The RS2 lists pretty typical RTK specs of H: 7 mm + 1 ppm V: 14 mm + 1 ppm which I assume to be RMS or 68% confidence interval results. For surveying I think you want to double or triple those to get 95/99% confidence interval levels.

 

So 95% CI for hz looks like 14mm + 46mm (2ppm over 23km baseline). Is ~60mm between positions ok for what you are doing? Your atmospheric conditions sound ideal and assume you have fast internet.

 

As an absolute position it would be borderline useful to establish a coordinate to set the RTK base on to look for other marks but as a relative precision I can't see it being useful for any of the surveying tasks I do other than GIS quality asbuilting or a rough topo.

 

 
Posted : October 14, 2022 12:50 pm
Mark O
(@mark-o)
Posts: 172
Member Member
 

As others have already answered, there are a LOT of factors. Another is howmuch of a canopy you have. My local CORS station is 2 km away, so I get pretty good resuts whether I use the network or the local reference station. In a wide open sky, I'm usually within about a cm or so horisontally. I recently staked out a property line for a friend in another state using a CORS reference station about 40 km in a partial canopy and while I was able to obtain a fix, the results were off by a 2 or 3 inches. OK for what I was doing (giving him a line for cutting trees). Bet thing to do it to take multiple measurements over many different times of day and you'll get an idea.

 
Posted : October 16, 2022 4:33 pm

Get the 2023 SurveyorConnect Wall Calendar

Page 1 / 2