2 degree roll 8 foot in the air is 3" if the wrong direction, that is 6". 6" twice (a-b, b-a) is 12".
Just saying, the little errors add up. Setting your antenna height low will either exaggerate the error or make it better. If it exaggerates the error your antenna is likely set up correctly. If it makes it better, then that gives you something else to consider.
Wishing you well with it. Let us know what you find.
It definitely seems to be roll related.
We’ve ensured roll is zero’d, but sitting in one spot not moving the roll can jump about a degree in either direction. Is this typical/expected of a dual + no IMU setup?
In any case we drove 5-6 lines and it was as before, we were anywhere from 4" to about a foot off row to row, but driving that same line AB or BA would stay true.
Setting antenna height to 0 got us 0-2" on every pass.
So now, what the heck is going on? Does it need an IMU? What’s the solution here?
GPS info shows HDOP 0.46 and HZ of 10.
Ignore the one missing ground plane disc, we tried it in every combination to see if that was affecting things. The lightbar is not powered on for what it’s worth.
What does the roll filter setting do? Does that need adjustment? It’s at 0, which I think is correct for PANDA (Teensy/UDP)?
My roll when stationary, flickers 0.1 degrees occasionally 0.2 degrees. If your roll is out by 1 degree a 3m or 10ft antennae height will give 5cm or 2 inch error, which if you drive the AB line BA would make you 10cm or 4 inches out.
Whats about cable/Antenna/connector damage?
None. AgOpen shows 12+ satellites. We were fairly close to a building when doing roll zero. I could always grab the F9P from the RTKBase and plug that in to those antennas and watch it in uCenter.
This is how I do roll calibrations, care of @PotatoFarmer
Are both antennas running high precision?
They were both flashed/configured with AOGOMatic with the right dual left/right config. If there is a way to validate that I can try.
At this point mostly talking out loud, we’re going to park it away from buildings and see if we can get roll stable. I may reflash both receivers (even though I know they’re on 1.13) and see if I can get uCenter to give me any useful info while they’re installed. From what I can read the F9Ps should have approx 1cm of vertical accuracy when in any RTK mode.
I know you can see it in u center. Upper left box look at the number of decimal places for the elevations.
You may be able to see it in the gps sentence in agio.
So we’re fairly sure we may have solved it. I was incorrect earlier about some of the testing that occurred without me being there.
Adding the ground plane discs to both antennas seems to have resolved the roll issues, which seems to get the lines where they should be.
It’s a fiberglass roof and aluminum square tube.
Thanks everyone for the suggestions along the way.
This is great news.
Sometimes you stand in your own way …
My lovely brother in law didn’t think they would make much of a difference because they’re not magnetic themselves, oh well. We learned quite a bit about how roll affects the calculations.
You could artificially lower the antenna height in the tractor config to effectively dampen the roll input.
This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.
