Reach RS fixes have poor quality

[bjp]

I have a pair of Reach RS’s and I’d like to get high-precision location data from one of them. But, what I’m seeing instead is that the reported rover location moves over tens of meters (much worse than my cell phone!) even when it’s sitting still on a wooden platform many feet above a roof. What am I doing wrong? Here is my setup:

1) Firmware image and ReachView version.
v2.8.0 shows in upper right of ReachView on both base and rover.
Updater shows a green checkmark for all items on both base and rover.

2) Detailed problem description. How does observed behavior differ from the expected one?
Location dots on ReachView Status page for rover meander over many tens of meters away from its true location. I expect the location to be constant within centimeters.

3) Your step by step actions.

• Configured both units to connect to wifi
• On the rover:
  • In Base mode, turned Corrections output off
  • In Correction input, turned Base correction on, selected LoRa 868.0 MHz 20 dbM 18.23 kb/s RTCM 3
• On the base:
  • In Base mode, turned Corrections output on with LoRa 868.0 MHz 20 dbM 18.23 kb/s sending 1002 @ 1Hz, 1006 @ 0.1Hz, and 1010 @ 1Hz
  • After placing base in pictured location, clicked the refresh button on Base coordinates in Base mode and waited 2 minutes for it to average (which produced a reasonable location estimate)
• Viewed rover’s ReachView Status on a PC

4) Picture of the setup and connection scheme.
Physical setup: https://photos.app.goo.gl/Jtv6CdajBMzd7JCH2
No connections.

5) If using any additional hardware please provide exact make and model.
No additional hardware.

[Simon_Allen]

Looks like you have a clearish view of the sky, but an environment likely to lead to many multipath signals. My first check would be to set the elevation mask to 20 degrees and see of that helped.

Then I would look at the observation log file using rtkpost for both the rover and base and use the elevation vs snr plot. IN a perfect environment should be a relatively smooth curve from maybe snr20 at 10 degrees to 40+ at 45 degrees. If you have downward spikes then you have multipath issues.

until you get FIX your Reach data will jump around as it is raw, unfiltered position calculations. Your cellphone (irrespective of brand) is so heavily filtered that it doesn’t jump around. Some phones also use the motion units to add layers to the filtering.

Hope this helps

[igor.vereninov]

Could you please post a screenshot of the rover status? It looks like the rover is still in single mode and not using RTK corrections.