This is the current issue. There may be others. But 0 L2 is a problem. As you can see, that’s under Rover.
Doesn’t that mean the rover can’t see any satellites? The roof could be causing that.
If it’s under or on top of the roof, it could. But if the rover is out in the open then those numbers have nothing to do with the satellites the DC sees. They are independent of each other in that regard. They both need to see the SAME satellites for RTCM to work. If the roof was causing an issue, you would notice that during an auto plan. The rover would think it is in one location when it is actually in another due to reflections causing the multipath issues with the satellites the DC is seeing. Unless I am missing something?
I don’t think your missing anything. I think you’re overlooking the fact that the roof could cause dead zones in the yard.
Maybe I’m missing something, I didn’t read the entire post. Does L2 always stay at zero no matter where it is in the yard?
Definitely not overlooking that, yet. Just troubleshooting the diagnostic screenshot and this statement first.
Fair enough. As I said earlier the roof is unlikely the cause but I based my rebuttal on this
I would say it’s unlikely but I’m struggling to see how you can say it definitely isn’t.
It definitely is not the current issue of zero L2 satellites being seen by the rover. There should be some if it truly is out in the open. Especially with an RTK status of 4. If it was seeing 15 and he was having issues with the rover jumping from one location to another in the app, then I would agree that it was the roof causing reflections and impacting the signal. That’s also why I asked about the DC Diagnostic screenshot. It will show the CNR values which will be a good indicator if the roof is causing an impact.
I guess we just have to agree to disagree on the roof definitely not being a problem.
1 Like
@bryan.wheeler is on the right track here. My DC is on my metal barn roof, and my rover sleeps right next to the barn only a foot or two away from the metal siding, and has for months. I don’t have any GPS issues. Multi path isn’t going to completely block signals and show zero L2. Even getting a low number would at least indicate the antenna are working. Defective antenna or broken wire are more likely, it’s easy to pinch the wires when attaching the antenna to the body.
2 Likes
My Yarbo mows 3 acres every other day. Cut by 99% my mowing for this summer. I only need to trim, and to mow occasionally some strip of grass Yarbo is too afraid to get close too (mainly close to large buildings).
So for sure it’ll cut mowing time massively. Now if you want a perfect land you need to play with it. After my first summer I’m quite pleased with the results but I had to cheat a bit so it zeroturns on patches that are hidden. I also vary the pathways manually by redrawing some every other week (some are on concrete or asphalt so I donc care but those on fragile lawn need to be moved if you don’t want tracks)
Also don’t mow in the rain. It’ll likely destroy too much to your liking.
And remove the guards if you don’t want it to be clogged all the time.
welcome to the club!
2 Likes
Hi there,
As Bryan suggested, you might try the following:
Check that both RTK antennas on the rover are securely fastened.
If that doesn’t help, try swapping the two antennas to see if the satellite count changes.
If the issue continues, please feel free to contact our support team with the details of the steps you’ve already taken so we can investigate further.
2 Likes
So I got out of work much later than I’d planned last night and had some errands to run so I didn’t have time to do much of anything with it last night, but this morning I took the antennas off this morning to look at the low mounts and confirm my recollection. The first picture is the right low mount, and the second picture is the left mount. When unscrewing the antennas the right antenna only took about 1-1.5 full rotations to come off, but the left antenna took 2-3 turns. Hopefully I’ll get home at a decent hour tonight and I’ll try swapping the antennas and then swapping the low mounts too. if the problem follows the antenna I’ll put in a ticket. If it follows the mounts, I’ll switch to the high mounts and see if that solves the issue.
1 Like
I’ll try to get this info tonight when I get home and post a screenshot of the details.
1 Like
Sure looks like an antenna problem to me! If you unscrew the 3 screws you can push the connector up so you can install the antenna properly. An antenna switch should not change much if it is simply not connected properly.
If those are the new style antenna amounts that have the embedded barrel connector inside them, turn it over and there may be some loose screws or screws that were put in at an angle that is causing that. You might be able to tighten it up a little bit and get some thread back. If it’s the old style connector with the clip, then what @Micke suggested is a good option too.
1 Like
The low mounts I received already had the barrels installed, unlike what was depicted in the manual. I also found it interesting that the way they were shipped, the antennas were pre-installed on the high mounts, but the manual didn’t mention that. Then again, it was kind of obvious that the manual hasn’t been updated recently as there were several things in the instructions that were wrong like which allen wrench was used for the side plates (or maybe it was the antenna mounts?) and the like. Also, it never mentioned that the controller was stored in the battery compartment of the robot. I had a little bit of a panic when I first unpacked the robot and was trying to check that I had everything and couldn’t find the controller. 
3 Likes
Mine were also on the high mount (But I had to mount the barrels myself). Also had the same scare with the controller (although I don’t use it anymore…).
Long story short, maybe the barrels are different like @bryan.wheeler said between you and I. But still, from the picture you showed, the connecter seems to not protrude properly compared to mine. Keep us posted if you are able to fix this!
2 Likes
So…. I’ve had a chance to look at the mounts…… And it would appear that they were assembled incorrectly….
Above you can see the disassembled low mount that was on the right side of the rover. I tried pushing on parts, flexing, rotating, and everything seemed solid and the barrel wouldn’t budge. So, being the engineer I am, I took it apart.
The first thing I notice was that the metal plate had a hole with two flats on it. Then I inspected the plastic potion for matching flats and notice this:
Not great to see as it looks like something was being pressed together that shouldn’t have, but after inspecting the barrel and seeing this:
The thought occurred to me that maybe, JUST MAYBE, the issue was something as simple as improper assembly. So I tried putting the barrel back in and intentionally misaligning it with the hole in the plate and resulted in the picture on the left. I rotated it until it lined up with the opening and felt it drop in and got the picture on the right…..
Needless to say, I put everything back together and powered up the rover, IN MY GARAGE, and got this:
I finally was able to drive it out, went out and got the charging station setup and connected and now have the setup below…
Though, interestingly as pointed out by the arrow in the picture the light on the DC has turned yellow (and occasionally goes back to green too…), which in the past couple days, I’ve not seen it do since I moved it. But, at the location with the charging station I now get this, though it occasionally blinks out for a second….
At this point, it’s a bit late in the day to try to do any actual mapping, but I’ll hopefully get a chance to try it out later this week or next.
And if y’all will forgive me, for as much as I was starting to stress about all the work I’ve put in thus far, I’m going to be a bit negative for a second.
@Yarbo-Forum this is a joke. For the price of this equipment, for such a simple failure to have ever left your manufacturing facility is appalling. Something like this should have a ZERO percent occurance rate because something as simple as aligning the barrel to the plate during assembly, but being so critical to functionality, is something that should be ONE HUNDERED PERCENT INSPECTED! I could understand if it was a component failure, or if it was something that required high precision and accuracy to assembly correctly, or was a delicate part that broke in shipping, but it’s a dang round part with flats going in a matching hole, and worse yet it’s VISIBLY OBVIOUS if you get it wrong. Unless these are being assembled by three random guys in a back room of an abandoned Chinese strip mall, this is really not cool and gives me pause to consider what level of quality the rest of the rover is built with. And it doesn’t matter that it was simple enough I could fix it in a couple minutes myself. I SHOULDN’T HAVE TO for the price this equipment cost. At the end of the day, this is just really poor manufacturing and quality control.
Now that said, now that it’s fixed and the problem was something that simple, I’m happy to go back to being positive and hopeful as I have been about Yarbo as almost everything else so far has been pretty smooth and the few other hiccups have been somewhat of my own making. But something like that does not sit well with me for a company with high asperations. Rant over…
5 Likes
So the yellow light on the DC is survey mode. If you go into the DC diagnostics it disables all DC functions and goes into survey mode to give you that nice scatterplot. Your CNR’s are very low. 36 is the minimum. You want to be in the 40’s most of the time. The roof could potentially be a factor there with reflections. Putting something over the roof in the area around the DC would prove or disprove that. Otherwise it could be a poor RTK antenna connection there too or just poor visibility of the sky. To check if it’s the antenna, swap the right one off the core (as you face the rear of the core) with the one off the DC. If your CNR values increase, then it’s an antenna issue. If not, it’s probably environmental. Ideally it needs 120 degree view of the sky from the center of the DC up like an ice cream cone.
Hi there, I’ll share this with the relevant department to help improve our quality control. Are you able to perform mapping normally now? If not, please feel free to reach our so that I can assist you further.
1 Like