Creating a known point

I’m almost at the point where i no longer need to keep bothering you lovely lot :rofl: :rofl:
Thank you all for your help so far in helping me to understand the Emlid and Reachview system.

I own two RS+ units and plan on using them as base and rover mainly for GCP’s and Total Station survey pegs (initial peg only set up to work from during topos) and some topos when the environment suits… clear open land, no trees etc.

I’ve got NTRIP connection sorted finally so i’m good to go with that.

So my questions are:-
If i set up a base RS+ on a tripod over an unknown point (new peg or gcp on site) and let that base unit collect data for approx 30 mins over NTRIP with base set to static and then create a project where i can input that collected data manually as well as the RS+ height (still on a tripod) and take a measurement at that point which could now be a gcp or total station peg… is that collected coordinate data sufficient for me to then set up as a base and rover by using those now known coordinates manually inputted at that base? This is not for any legally certified boundary work for example but still needs to be cm accurate for the gcp’s or total station peg… obviously then i would not average collected base data again but base any further survey work around that first set of data collected and obviously i would not need to keep NTRIP connected then.

Providing all parameters were set correctly and AR was high, low baseline in NTRIP etc would there still be a necessity for post processing or would that surveyed data be accurate enough?

I hope that makes sense but just ask if you need me to explain further.

Basically…set up one RS+ on a peg, gather data over NTRIP, record data… then use that data to set up project as a base and rover by manually inputting those coords and you’re good to go without NTRIP from then on.

1 Like

If you are doing site/project work it will work just fine. The repeatability is what you’re looking for and as long as you have some kind of benchmark to tie your GCPs your map will be relative.

If you are doing property or other work that is recordable then it will depend on your municipality and whether or not the require true NAD83 or LLH coordinates.

I feel like there is another thread just like this?

Yep as I’m learning about the Emlid/Reach system I’m finding to ask the more relative questions.
I hadn’t managed to get NTRIP up until this weekend after being stuck in SSID hell so was always working on the fact that I’d have to set an arbitrary datum and post process everything.
Now I’m on NTRIP and know ReachView 3 outputs directly to UK coords with no conversion needed, the kit is becoming more attractive for what I need it for.


It feels good to get through the technical adjustments so you can get to doing the enjoyable part of creating!


Hi Michael,

I have nothing to add to Michael’s words here :slightly_smiling_face:

Just want to confirm one more time that Reach RS+ delivers the same level of accuracy for both RTK and PPK.


If I’ve collected base station data for 30 minutes with a single RS+, I presume I just deduct the height to underside of the unit from the elevation given?
Obviously there’s no instrument height input at that stage.
I created a project to take one point, effectively using that base as a set up, and put the instrument height in as the pole height but got 14mm difference.
I appreciate that’s not such a big error in the grand scheme of things and that error would vary each time I did it.
So do I use the base set up, minus the height to underside of unit or creat a project and take a point where I can input the pole/instrument height?

Did you use the brass adapter and account for the extra offset to the actual phase center of the receiver? The measurement is not to the bottom/pole height. The RS+ receiver itself has a 65mm offset and the brass adapter is another 22mm so I am surprised your deviation was only 14mm.

I would like to confirm that the M2 is so far superior to the M+, that it completely justifies the cost.
The M+ is a fun and super accurate toy, the M2 is how I imagined GPS would work when they first described how it would work it in the 1990’s. The 2 fixes and holds like a pro :+1: :+1:.

I am sure the same relationship is true for RS+/RS2


I’m beginning to wonder about getting rid of the two RS+ and going for an RS2.

1 Like

No !!! Keep the RS+'s . You can have a very robust way to verify the RTK positions if needed. Two RS+'s as static and RS2 makes a closed polygon to PP when doubt encroaches in the RS2 position confidence. Also, short baselines to the rover are your friend. I do the same process with my 2 M2’s and Javad Triumph-2 RTN receiver !! Very excellent way to verify the “fixed” solution !

1 Like

Can you tell some more on how to do this? That could be very interesting!

Use RS2 RTN to establish positions of both RS+'s (in open area, approx 200-300m apart depending on rover baseline length). Use RS2 as RTN as rover to collect points. If you have a questionable fix, collect data for PP against RS+'s. If you use commercial PP software like I do, you’ll have a closed loop for accuracy and can adjust (visualize a triangle). I do this all the time to verify a questionable position, especially if I loose cell coverage. Accuracy of RTN vs PP usually <2cm horizontal and vertical. Short baselines are your friends.

1 Like

I think the majority that buy Emlid’s product don’t have access to a commercial solution. Do you know how this is done “in-hand”? It would be really nice to prove accuracy on a project.

This basically sounds like an after-the-fact localization…

I guess if you use only rtklib, you could process the unknown baselines, assuming you have good accuracy for the main static baseline and know the bearing and dist between both the rs+'s, you can PP the the other two baselines and compare the resultant coordinates of the RS2 rover. I would convert the computed coordinates of the RS2 to some kind of linear measurement system so you could actually know what the error is instead of geographic coords. Rtklib is fine for experiments and use in UAV use as in Emlid, but doesn’t have multiple baseline processing and adjustment routines for true geodetic work that we do in a commercial environments. Once you use a true high-end commercial PP software, you’ll be amazed in the things you can do.

Being a licensed professional land surveyor, my whole point here in this post and others in the past is to verify what is produced by not only Emlid’s products but also high end receivers such as our Javad gear. Just because you have a “fixed” solution being shown, you really don’t know what the accuracy of the position without some kind of verification.

As I’ve said before, with the low cost in Emlid’s products, there’s really no excuse not to have another receiver or two to verify your positional product whether it’s a local CORS or an on-site second or third receiver. Too many people rely on only a single baseline. It’ll get you in trouble one day


Good to hear from someone else and exactly why we refuse to fly on RTK alone…


How far of a baseline are we talking about?
In my mind, a tapemeasure isn’t precise enough if we are talking more than 10-15 meters, at least if the end-error needs to be <= 2 cm.
Then we are talking total station territory for establishing that baseline along with the bearings?

From what I understood, it’s indeed something to do with a TS if you want a bearing and distance.

1 Like

Cool that explains a few things.

I pulled up the survey plan for my place and all the measurements are a bearing and a distance from a known point (section marker at the nearest crossroads). The distances on the bearings are all in m, it also tells me the datum they are using.

So I should be able to find the error in my base position by taking the xyz of the four corner survey markers then changing the offset I find in the x and y in the base so that next time my rover will read the correct coordinates when I check at the four property survey pins?

1 Like