And if it were really that easy, why didn't 3DR finish it, instead of backing out to older versions of ardu?
 
I believe the hardware problem is the ESCs, not the stock pixhawk. The stock pixhawk uses 3.3v PWM signalling. The ESCs work better with 5v signalling. I believe that is related to the saturation and derating down to 96%. I'm trying to verify this myself but I'm pretty sure.

The new 2.1 cube will have 5v signalling. There will be an altered version of the 3DR solo branch of Arducopter to go with it after it's released. It will likely undo the various things that were done to the normal firmware to get around the ESC saturation. It will also have what's needed to make use of the improved and additional hardware such as IMUs. Again, this is as I understand it. But I'm working to confirm this.
 
  • Like
Reactions: Jose969
And if it were really that easy, why didn't 3DR finish it, instead of backing out to older versions of ardu?
That's an easy one. :) The last rebase beta was pretty stable already. I fly it since then without any issues on my Solo.

It has one potential flaw, when you move Solo around during the startup and initialization you may compromise the EKF, resulting in uncontrollable behavior up to a flyaway. I could reproduce this behavior during beta testing and discussed it with Phu at that time. After some more incidents (reported in the fb group) 3DR decided to step back to 3.2 base instead of fixing this issue, since the rebase capabilities were not needed anyway for Sitescan. Also, the consumer dev team around Phu was fired at this point in time as we know.
 
That was unfortunate timing. That bug causing the EKF fly aways was actually in Arducopter 3.3 Master at the time too I think. Sadly fixing it on the solo branch was not in the cards since everyone at 3DR was being laid off at the time. The only practical thing to do was roll back to an older version. That older version lakes quite a bit of good stuff that it would have had if it stayed on the 3.3 version. And it's night and day compared to the current 3.4 release or 3.5 testing.

If you're running master on the stock solo pixhawk, you need to derate the throttle maximum output parameters to keep the max PWM below 1870. There are coding changes you can't replicate with parameters too. That level of detail is over my head. But I believe if you keep the PWM maximum below 1870, it will be generally safe to fly, just understand you're not a test pilot running something not designed for that purpose. The smart shots will also not be able to control the gimbal. You can control it with the paddle on the controller. But smart shots basically become useless.
 
What is the advantage of running AC3.5 over the latest 1.5.3 on the Solo?

I am currently running 1.5.3 with no issues and probably get my head on trying the AC3.5 if it is far better.
 
Someone better informed than me can comment on all the specific changes, but If you just want to use Solo to take pictures, probably minimal changes that affect the aircraft flying qualities. What they have introduced over the last few versions are the ability to use several other sensors (LIDAR, Ping, sonar, etc).
 
its a step back vs long term support forward question.
Ardu will address the EKF issues that Pedals mentioned above and allow solo to run new ardu features like boat mode, throw mode, swarming New GPS options Flow and Obstacle avoidance, but for now you will lose smart shots.
With the advent of the PH2.1 for solo and someone who can crack the solo gimbal code then we could have it all.
I honestly thing if 3DR could have released the gimbal code they would have
 
Over the last two weeks, I've been testing the latest beta Arducopter master on my second experimental Solo, which is currently version 3.5-RC2 (beta). The current release version is 3.4. But I figured lets throw myself right to the wolves on the latest beta. It is a LOT of work, and continues to be a work in progress. My testing results are being shared and reviewed by the developers behind Arducopter. And will going towards the greater effort of getting master workable on the Solo for general consumption. I think it has helped provoke more interest from the Arducopter devs, who are already really really busy to begin with. This also all goes well into the Green Cube's impending birth.

I can go on for an hour about everything here. But for now I'll summarize the key points that matter to the majority of Solo users:
  • The gimbal works in all respects! There was an incompatibility, but Phillip pushed a change into Arducopter that fixed that.

  • The smart shots work in all respects! Again, there used to be issues, but they've been addressed.

  • Boat mode! My kayak is not an ice breaker, so I can't test it on a boat. But it works for everyone else that uses Arducopter, and there's no reason it won't work. At some point, I will try arming it from in my Jeep while driving circles around the end of the road.

  • Way better handling of IMU data and handling of errors/conflicts. Stricter calibrations, but not inhibitive.

  • Ability to use RTK centimeter accuracy GPS (AKA, the AWACS mod!) for precision surveys. Supports dual GPS, and dual GPS blending for redundancy.
So I think there's a light at the end of the tunnel. I can't tell you how far away that light is. But it's there!


On a related note, it's a good thing I have spare parts and multiple Solos. I wasn't paying attention and it flew into a tree top while running auto tuning today after work. If you think you're above the trees, push it up another 30ft. I'll doing lots of transplant work this weekend. Based on the weather forecast, I might not lose any time since the weather this weekend will suck anyway. Note, this crash had nothing to do with Arducopter. It had everything to with the moron holding the controller trying to fly while shivering.

65xv9u3.jpg
 
P2P,
Thanks for the great info. Looking forward to flying master on my Solos
Sorry about your unintentional meeting with the tree! Looks pretty bad. Bet the cold makes the plastic more brittle.
 
I've seen crashes from higher cause less damage. So you're probably right. It also doesn't help that after hitting the tree, it fell straight down onto solid ice.
 
  • Like
Reactions: OldCoder
Wow, that sucks. There's nothing like that sinking feeling you get as you see your Solo plummeting to the ground. :(

On the bright side, you've now got a lot more spare parts. :)

Hopefully somebody will come out with a 3rd party housing we can put all of our spare parts back into...
 
why would you do a marketing ploy on a community product with little to no profit margin
The reason why is simple there are a lot of ex 3dr employees that got there feelings hurt when they got canned and are trying to bring there ideas they had while at 3dr to the table and profit from it. Solo is a very profitable drone since it's open source and because it's open source it never really ages or out dates like Dji and other drones. And besides that it's America people try to profit off every thing it's the great rat race
 
Perhaps you think it should be a charity? The developers that design and build this stuff should just give it all away for free and spend tens of thousands of their own money donating advanced hardware to people on the internet? When did profiting off your hard work as a small business building a good product become evil??

The notion that it's just a useless add on bilking people out of money for no actual benefit is completely wrong. It's an optional advanced upgrade for those who an advanced optional upgrade. Does MAS give away propellers for free? Does FPVLR give away antennas for free? Do you need any of them? No. But they're there if you want to buy them.
 
Last edited:
Over the last two weeks, I've been testing the latest beta Arducopter master on my second experimental Solo, which is currently version 3.5-RC2 (beta). The current release version is 3.4. But I figured lets throw myself right to the wolves on the latest beta. It is a LOT of work, and continues to be a work in progress. My testing results are being shared and reviewed by the developers behind Arducopter. And will going towards the greater effort of getting master workable on the Solo for general consumption. I think it has helped provoke more interest from the Arducopter devs, who are already really really busy to begin with. This also all goes well into the Green Cube's impending birth.

I can go on for an hour about everything here. But for now I'll summarize the key points that matter to the majority of Solo users:
  • The gimbal works in all respects! There was an incompatibility, but Phillip pushed a change into Arducopter that fixed that.

  • The smart shots work in all respects! Again, there used to be issues, but they've been addressed.

  • Boat mode! My kayak is not an ice breaker, so I can't test it on a boat. But it works for everyone else that uses Arducopter, and there's no reason it won't work. At some point, I will try arming it from in my Jeep while driving circles around the end of the road.

  • Way better handling of IMU data and handling of errors/conflicts. Stricter calibrations, but not inhibitive.

  • Ability to use RTK centimeter accuracy GPS (AKA, the AWACS mod!) for precision surveys. Supports dual GPS, and dual GPS blending for redundancy.
So I think there's a light at the end of the tunnel. I can't tell you how far away that light is. But it's there!


On a related note, it's a good thing I have spare parts and multiple Solos. I wasn't paying attention and it flew into a tree top while running auto tuning today after work. If you think you're above the trees, push it up another 30ft. I'll doing lots of transplant work this weekend. Based on the weather forecast, I might not lose any time since the weather this weekend will suck anyway. Note, this crash had nothing to do with Arducopter. It had everything to with the moron holding the controller trying to fly while shivering.

65xv9u3.jpg

It's about time you told people what you have been up to Matt. :)

I would be right there with you but I only have 1 spare. Thanks for taking one (or more) for the team.
 
  • Like
Reactions: Pedals2Paddles
great, in that case it would be possible to use duap gps in our Solo via accessory bay :)
 
great, in that case it would be possible to use duap gps in our Solo via accessory bay :)
Yep. Obviously it would need to be mounted above. But I believe the necessary i2c ports are broken out on the accessory bay or on the main board just above it.
 
i through that second gps should be connected via serial port at least in pixhawk 1
 
Just had to go back and look. You're right, GPS is serial and would connect to SERIAL2 (AKA TELEM2) which is on the accessory port. I was thinking of the compass that usually accompanies the external GPS units, which connects to the i2c port on the main board. Before someone asks, no I don't believe you can use two external compasses.
 
OMG..!

That took a tumble.
Think "solo skin graft" springs to mind.

Btw, do you have a glue gun?

Lol

Really really gutted for you though mate. Hope you can salvage at least something from it.

God damn trees

fc33689b2e725e169a36a543976d3ce7.jpg
 

Members online

No members online now.

Forum statistics

Threads
13,093
Messages
147,741
Members
16,048
Latest member
ihatethatihavetomakeanacc