RS2+ RTK vs Rover height difference

  1. Emlid Flow app version 9.3
  2. I have been mapping using a M3E, with 2 x RS2+ as base and rover. I use NTRIP to average fix over 20 minutes for the base, I then capture points for my GCP’s with the rover. I only average the point capture for rover for 20 seconds. And then fly the mission with RTK direct with the drone.
    Both the rover and base were set up to use GDA2020/MGA Zone 50 (As still seen in the Emlid point capture part of the app). So I can’t figure out why my GCP’s heights are shown in Ellipsoid, showing the rover origin as ‘Global’ .

I also note that for some reason the CSV for the GCPs shows antennae height incorrect to what was entered and saved when prepping for the job. I still had the antenna heights entered from this the last job which I screen captured, correctly showing 2.334m for rover and 1.634m for base.

Yet the survey point collection part of the emlid app shows what was also reflected in the csv, the rover to have an antenna height of 1.934m.

Surely the heights should correspond having used the same base and reference system. Sorry from a tired noob going over every post and Youtube trying to figure out why.

1 Like

In your screen shots it shows AVERAGE SINGLE for the BASE?

Did you mean to AVERAGE FIX or enter known point manually instead?

Choosing base setup method | Reach RS3#ways-to-set-the-base


Hi timd1971,

thanks for that.

Yes sorry you are correct, average single, not fix for Rover which I see as not being as accurate with your link. If I averaged the base over ntrip for 20mins on a baseline less than 10kms, then should I still expect the base to be accurate if set to average single, the pdop was 1.0?

I didn’t have any known survey point, they are very rare in my neck of the woods.

Still doesn’t explain the antenna height anomaly being different to what was entered, and also the CS as well?
kind regards

When you are setting up your local Base unit, if you are receiving corrections over NTRIP from a CORS or a similar correction service, you would want to select Average FIX to get absolute cm precision. Selecting SINGLE will only get you 3-5meters relative precision and averaging anything over a couple of minutes in SINGLE will not necessarily enhance your results for your Base. As for the antenna height problem, that does sound strange not showing what your input height that was entered.


thanks mark1st.john

ok, looks like I need to bite the bullet and redo the site. It’s close to home, so all good.
Thanks kindly




Are your base and rover antenna heights backwards? Typically the base is higher than the rover is best… but i d9ntvthink it matters too much… but in my case, my base is always higher on a tripod etc using a extender.

Wondering if u have an extender for the base and not being taken into account?

1 Like

Hi @lighttunnelprod,

As per your screenshots, you didn’t specify the vertical datum. If you did, the Сoordinate system bar in the point’s info would look like that:

Your screenshot shows that you specified 2.334 m antenna height in the Base settings of the rover. It’s unnecessary because you use it as a rover, not the base. The antenna height of 1.934 m is obtained by adding a 1.8 m measured pole height to the ARP offset of 0.134 m: 1.934 = 1.8 + 0.134. And you specify this in the Collector settings:

As @mark1st.john correctly mentioned, averaging the base in SINGLE gives you meter absolute accuracy. If you’re in the neck of the woods without known benchmarks nearby, AUSPOS comes in handy to get precise base coordinates in GDA2020. Once they’re configured in the Base settings, the rover coordinates will have the same GDA2020 datum.


Sorry guys for the late reply,

I really appreciate everybody’s time to respond.
The base height didn’t need to be high as it was on top of a hill, and my GCP’s had a clear line of site.
I will give it another crack soon and see how I go, I’ll look out for the screen to show what you advised Kirill. Thanks again everyone for your time. Legends

Kind Regards