Running Solo on Master

Toggle sidebar
Pedals2Paddles said:
So many things going on in this screenshot! Arducopter master 3.5-RC3 Beta on the Solo. Holding steady in Fly mode (Loiter), leaning into the wind. Solo's video on Mission Planner's HUD with Solex up on the phone (video only to one or the other). Terrain awareness working. Auto mission waiting to begin. Today was a good day of testing. This time I only had one CFIT incident, and only broke one prop. I have a good set of PIDs that I think will be nailed down for good with one more test day. Almost all the other parameters have been worked out. Again one more day of testing should have that nailed down. Tested follow and zip line smart shots. Tested orbit smart shot a week ago. I'm presently documenting a repeatable process for installing and configuring. Which I also think I'm very close to completing. With the externally mounted CP antennas on the solo, and only the stock antennas on the controller, I has a solid stable connection out to 3,800ft in the country. I don't believe I ever got that far on stock antennas in this location before.

The only major incompatibility so far is with EKF3. It will fly just fine with EKF3 enabled. But the compass and level calibrations will not work with EKF3 enabled. It just craps itself and loses connection with the controller. Really doesn't make any sense, and I don't know why yet. So for the time being, I have EKF3 disabled.

10Yw62v.jpg
Click to expand...


Looking good P2P, you are the man...
:-)
 
I have a question and please don't call me an English idiot lol

I have the FPVLR antennas and they are CP as such you have to point them at solo.

The antennas you have installed in your solo are also CP if I read correctly.

My question is as follows.
If they are CP antennas which way are the wifi radio waves pointing when solo is in the air?

Left to right or up and down?

I hope I am not being stupid here but if they are directional CP antennas I don't understand how they work.

(It's probably obvious to everyone else or such an easy observation no one has asked, but I'm not quite with it)

BTW I love MP with the heads up display pity I'm all mac
 
Circular polarization isn't necessarily directional too. Just like linear polarization, there are antennas designs that are directional and designs that are Omni. The CP antennas on the FPVLR are directional. The CP antennas on my solo are omnis.
 
  • Like
Reactions: speatuk
Where did you source them Matt?

I can only find RHCP I can find LHCP anywhere.

Also which pigtails did you use?

I'm intrigued and definitely considering this myself. But here in the U.K. they don't seam to be available anywhere, I suppose RC is not as big over here.
 
  • Like
Reactions: OldCoder, SteveReno, speatuk and 1 other person
speatuk said:
I have a question and please don't call me an English idiot lol

I have the FPVLR antennas and they are CP as such you have to point them at solo.

The antennas you have installed in your solo are also CP if I read correctly.

My question is as follows.
If they are CP antennas which way are the wifi radio waves pointing when solo is in the air?

Left to right or up and down?

I hope I am not being stupid here but if they are directional CP antennas I don't understand how they work.

(It's probably obvious to everyone else or such an easy observation no one has asked, but I'm not quite with it)

BTW I love MP with the heads up display pity I'm all mac
Click to expand...

One of the advantages of CP antennas is the orientation doesn't really matter. Like other Omni antennas there will be a slight null straight up and straight down, but it is pretty small.
I'm interested to see how mixing LHCP and RHCP affects the performance of the WiFi link. I'll have to order a couple LHCP antennas and conduct some testing.
 
 
 
Thx. This companion computer home location conflict is breaking my balls. Hoping the higher minds of Ardupilot can offer some insight in the next few days. Maybe a topic for the dev call tomorrow. Outstanding major operational issues are the RTH/RTM home location issue, and intermittently unreliable gimbal.

It's also worth noting that I had my first ESC reset in flight running master without the Solo version's protection hacks. So anyone who didn't believe Phillip when he said this was an issue, believe it. The new green cube or conventional Pixhawk 2.1 rectifies the matter completely.
 
So in your opinion P2P which is the one to get.
P2.1 or the green cube?
 
apparently they are identical except for a jumper that can be removed in the black one
Most likely if the test on green go well I will just grab a black 2.1 for solo 1.
I have one of those flying on my DIY with an Edison carrier running apsynch,
Having said that, if the green cube stays the same price and it contributes to the dev of ardu and solo then I would buy another green one to support the effort
People are working their butts off on this. So sad folks don't appreciate the work being done.

What is good news to me is that the solex app will still work.
That technically means there might be hope to have the solex app run on a DIY, since they did release the code for the onboard IMX6 for solo, then you would hope that a DIY with 2.1 running master and a companion computer capable of running IMX6 code could theoretically run solex
All but the mythical gimbal code
 
So...
Is the green cube just plug and play on the solo?
 
Yes, well except having to load the firmware
the black 2.1 will work the same, just remove the jumper
 
Ok then.
I will go for a green one as it contributes to the dev of arducopter then I'll do that.
Thanks for the help

Ordered. Due end of April
:-)
 
  • Like
Reactions: Pedals2Paddles
Master has been tough this week. Ran into a maddening compatibility issue that was preventing the RTH/RTM from working. That's a major reliability and safety issue. That basically stopped all other testing and flying. Pouring through lots of logs led to why it was happening. I added it as an issue for Arducopter, and a fix was just committed to Master last night to rectify the issue by one of the wise men behind the scenes :). That fix should be in the next beta (3.5-rc4) coming out any day now I think, so real testing can safely resume then.

At this point, I think the main things left to address are some gimbal strangeness, the motor arm LEDs, and a repeatable user friendly factory reset procedure. Which I do not believe will be huge issues to solve, and are not safety issues at all.

Worth noting: master does not and never will have the throttle derating code baked into it that the Solo version has. That derating is a fix for the ESC flaw. It's not a perfect or 100% fix, but works for the most part today. Without it, that ESCs are prone to reseting in flight, causing flips and crashes. I've already experienced it once. This is something the Green Cube (or Pixhawk 2.1 with the jumper changed) rectifies. So all along folks like Phillip have been saying there is a safety issue with running master on a stock Solo pixhawk. And despite what some may think, they're not lying to you. I would go as far as saying that if you want to run master on Solo, you need a new Pixhawk 2.1... either the green cube variety, or a plain 2.1 and change the signalling voltage jumper. Of if you're building something DIY, different ESCs would do the trick too.
 
Last edited:
  • Like
Reactions: OldCoder, andrew.baker.142, speatuk and 2 others
Pedals2Paddles said:
Worth noting: master does not and never will have the throttle derating code baked into it that the Solo version has. That derating is a fix for the ESC flaw. It's not a perfect or 100% fix, but works for the most part today. Without it, that ESCs are prone to reseting in flight, causing flips and crashes. I've already experienced it once. This is something the Green Cube (or Pixhawk 2.1 with the jumper changed) rectifies. So all along folks like Phillip have been saying there is a safety issue with running master on a stock Solo pixhawk. And despite what some may think, they're not lying to you. I would go as far as saying that if you want to run master on Solo, you need a new Pixhawk 2.1... either the green cube variety, or a plain 2.1 and change the signalling voltage jumper. Of if you're building something DIY, different ESCs would do the trick too.
Click to expand...

Out of curiosity, did you experience the ESC reset with the "MOT_THST_MAX" parameter set to the de-rated (0.94 / 0.965) value?

I'm assuming there are some (ugly?) non-parameter based fixes (hacks?) as well, otherwise I don't see why the ArduPilot wouldn't just include the changes in master for the px4-v2 target. Nonetheless, if can identify the fixes in the ardupilot-solo codebase, it should be possible to apply them to master to create a "master-solo" version.
 
You are correct on all counts. Baked into the motor matrix code is motor output slewing. In short, this derates the PWM rate of change. This code does way more for the Solo's protection than the max PWM parameter. It is a total hack of a fix, but it was really the only way to address the problem without redesigning the ESCs. Believe me, the developers behind it hate it more than we do. It is crippling to the Solo's potential performance in the sky.

During changes in load, there is ground lifting condition on the ESC. This ground lifting results in the ESC seeing voltage drop from 3v to less than 2.5 volts. When the voltage drops below 2.5 volts on these ESCs, the signalling simply stops. The motor subsequently stops. And down you go. In most cases, the signalling comes back on within a second. Given enough altitude, the copter can often recover from the tumble. Mine did thankfully. But not before scaring the ever living crap out of me. When this happened to me, it was actually two ESCs that shut down at the same time, then came back on. Soo all that stuff the developers say about it being risky... they're not lying.

This very topic was discussed in the weekly developer's conference call on Monday again. The conference call was 2.5 hours covering all Ardupilot development matters. And in that 2.5hrs, we actually had about a 1/2 hour long discussion on all things Solo master, which was great! Lots of great input, ideas, and discussion on how to proceed. Anyway, the consensus right now is we need to put this into master, and use a parameter to enable/disable it. Default would be disabled since literally nothing but a Solo with a stock Pixhawk 2 needs it. If master is being installed on a Solo with a stock Pixhawk 2, that parameter will be enabled in the parameter file. If you've replaced it with a Pixhawk 2.1 / green cube, you can turn it off again. That's our current thought process anyway.

Doing this will allow stock solo users to safely operate master. It will allow Pixahwk 2.1 solo users to safely operate master without handicap. And it will have no effect on anyone else using Arducopter.
 
Last edited:
  • Like
Reactions: OldCoder, andrew.baker.142, David Boulanger and 1 other person
You must log in or register to reply here.

Similar threads

A
The IMU calibration stays on 0 (IMU calibration and compass calibration issues)
Replies
2
Views
1K
General Discussion
chrowe
chrowe
Flyinglawnmower
Solo refused to stay on Manual control
Replies
2
Views
2K
Solo Help
Flyinglawnmower
Flyinglawnmower
n8lbv
3.3V ESC PWM vs 5V PWM How much am I really missing out on?
Replies
6
Views
2K
Solo Discussion
n8lbv
n8lbv
J
Issues installing Open Solo 4.0 via SSH/SCP
Replies
3
Views
3K
Solo Help
jorgeposada
J
G
Question regarding transferring Firmware updates between working and reset birds - Did reset on original Solo
Replies
7
Views
3K
Solo Discussion
Swallabat
S

New Posts

Members online

Total: 239 (members: 1, guests: 238)

Forum statistics

Threads
13,164
Messages
148,288
Members
16,219
Latest member
Hanmer22