Reach for tractor guidance

Wondered if anyone had experience to share of using the Reach RTK module for steering guidance of tractors or other agricultural machinery. How are you finding accuracy of the system?

Hi! Last year i did try reach with my tractor.

The system worked very well for 70 % of time. My biggest problem was losing fix under best conditions.
I hope that most of the problems were solved with new beta.

In Austria we start with agricultural work in march/april. so i would be able to give more feedback.

you can read here:


Hi Andreas,
Many thanks for the response. I have seen some of your posts on other forums- (TFF and TheCombineForum).
Glad to hear you have had reasonable success, though obviously a bit more reliability would be preferable. Have to hope that latest updates will improve this. Do you think a better antenna on the tractor might improve matters?
I am in Scotland (UK), so same work season as yourself. Hope all goes well when spring arrives.

1 Like

I use Tallysmann TW3710 on base and rover. They are much better then the standart antennas.

I have also bought an m8t board at csg-shop.

So i can use it also on my trekker if things go wrong with reach…

what software you are using for steering?


I currently have a Trimble EZ Guide 500 lightbar which uses GPS & EGNOS. Supposed to provide 200mm pass to pass accuracy- but can experience quite a bit of drift at times. I would content myself with a more accurate guidance system to follow for now, and possibly upgrade to full autosteer later. The thread on thefarmingforum about Stef’s work on DIY autosteer was most interesting.
If I was to pursue Reach, or similar GPS RTK module as a receiver, then BrianTee’s AgOpenGPS project looks worth following. I have a dual boot Windows /Android tablet on order, so can explore the possibilities of guidance apps that exist for both OS platforms.

Hello Andreas,

I wonder how your experiences are now with the Reach for guidance? You describe losing fix under best conditions. Does this still occur? Was the problem identified or fixed?
In the guiding sessions I did so far, I lost fix sometimes, I have started a topic for it here:


Losing fix under best conditions was solved by emlid and rtkexplorer- there was a error in coding of rtklib…


Hi, I use 2 reach rs, one as rover the other as base on my rooftop, i can tell you with 1 month of tractor field experience with cerea, that fix is very strong, and goes immediatly when i turn on my rs. I think it depends whether you are LoRa or TCP which i find much better.

1 Like

Hi, do you have any tilt compensation?

Imu brick 2.0 yes

1 Like

Within how many cm does your tractor drive on the field now? Do you have tilt compensation?

1-2 cm in best conditions…

yes i have tilt compensation with MMA8452

What’s the minimum accuracy in worse conditions? What’s the comparison between your tractor’s driving accuracy compared to a John Deere Starfire 6000 or the best Trimble GPS while on hilly or rough terrain? Is the tilt compensation module very accurate? Does it contain a compass? Do you still use a motor on the steering wheel for the autosteer? Or are you using valves or a motor on the steering column?
Where can I find such parts as you are using for the autosteer and tilt compensation and the different parts for the whole GPS system?

Thanks in advance!

1 Like

start here…


Ok. What should i do with this software?

Could you send your phone number to this email ( so i could contact you more quickly via WhatsApp?

Finally started using AgOpenGPS earlier this year with my Reach modules (original 1st generation ones).
Base module connected to an antenna mounted on the roof of my house. It is logged into an account with RTK2Go, sending NTRIP correction data over wifi to my 4g mobile router (only usable broadband connection in my locality- no fibre, and ADSL is useless).
Rover module is connected to GPS antenna mounted on the roof of my tractor cab. Powered by and sending position data via USB to a Windows 10 tablet. Smartphone provides a mobile WiFi hotspot which reach rover uses to obtain RTK correction from RTK2Go.
Rover usually obtains a fix within a few of minutes of being powered up, and whilst it maintains a fix, combined with the AgOpenGPS software provides a top rate guidance system. I am only using it as a lightbar at present- I intend to put tilt compensation and autosteering in place some time in the future.
Sadly though on occassions, the rover loses fix, and drops to float with very much poorer accuracy. Having moved to float condition, usually (but not always) it is unable to restore a fix without rebooting the module.

Any suggestions on why fix is being lost?

Reach modules are running firware version 27, so almost fully up to date.

Generally I have a clear view of the sky- but I am on a North facing slope, so there will be some limitations on the number of satellites visible to the South. My mobile data signal is generally good, so shouldn’t be a problem with loss of connection to the NTRIP caster.

Could the fact I am relying on mobile data for connection to the NTRIP caster by both base and rover be an issue?

Have other user found issues with reliability of RTK2Go?

Would a dedicated mobile router in my tractor provide a more robust Wifi connection to the NTRIP caster than using a smartphone as a wifi hotspot?

If I upgraded the single band Reach module I am using as my rover to a multiband Reach M2, would this retain or regain a better/ quicker fix? Can I use a Reach M2 rover with my single band Reach Base for RTK corrections?- I dont necessarily need the extra accuracy of the multiband, just want an improved fix.

As a further point, I did once find that my base had been kicked off of RTK2Go- presumably for transmitting gibberrish. I now have a timeswitch which powers it down for a couple of hours overnight, giving it a daily reboot, and havent had any such problem since.

Any suggestions to resolve this issue would be much appreciated.

Check out PotatoFarmers work!

He has done a bunch of testing with steering.

1 Like

OK. Thanks- I’ll have a look at that.

Hi @WilM,

Very welcome back to the forum! Enjoy the community’s evolution :slight_smile:

Thank you for being precise about your workflow. This is a beneficial attitude for all of us!

It’s a bit tricky to come up with an explanation for why the Fix is lost without seeing the actual data. There may be several reasons for it. Losing the correction stream is one of them.

What about the baseline? What distances do you usually cover? Going on for longer distances than suggested may actually prevent the unit from calculating a Fix.

That depends on the strength of your network, of course. In case you have enough Internet traffic, it should be okay.

I don’t think so. Still, you can try our Emlid Caster as an alternative just to check if it makes any difference.

I believe it can provide a faster connection. It can affect the solution but I don’t think the difference is drastic.

You can get all of the advantages of the multi-band receivers only if your full RTK setup is multi-band. It means that both your rover and base should track L1 and L2 so that you can have a quicker fix, among other things.

If you use a single-band rover with a multi-band base, like Reach Module with Reach M2, your rover is still working with L1-corrections only. That’s why you can only get the perks of a single-band system.

I don’t recommend working the other way around at all - I mean using a single-band receiver as a base for a multi-band rover. The multi-band rover needs to receive the L2 corrections which a single-band one just doesn’t provide.

So, for now, we will need to understand what happens with the solution status. For that, as the first iteration, we need to check how the solution is calculated. This can be checked through the solution log from the rover recorded during the mission. The raw data log and the base correction log will show us the overall data quality and the quality of the transmitted data. You can share it with us here or via

1 Like

Many thanks for the detailed response- sorry it has taken a while for me to acknowledge. I have not been using Reach so much in the last month, so haven’t had time to investigate your suggestions further. I have taken on board your advice about not mixing single-band and multi-band units in a rover/base setup.
I have included a link at the end to a cloud drive with 3 sets of log files from one of the days I was particularly having problems maintaining Fix.
For some explanation of some of the occurences in the 10:21 log-
The periods of Single followed by Float are where the rover was driven through dense trees, so no sattelites visible.
The periods of single followed by Fix are where wifi hotspot has been taken away for a short time, so no RTK.
Not sure why fix lost around 14:00.
Real problems start around 16:30- unable to re-establish Fix, after driving through dense trees. There do appear to be plenty of sattelites received at this time.
The 16:44 file is good until Single at 17:40- after this Fix is only established for short times. There apears to be a low number of valid sattelites showing at this time,
The 18:06 file is all pretty good. Good Fix, and plenty of sattelites.

I would be interested to know if there is information in the log files which can help explain and hopefully resolve my problem.