After sadly finding that there is a bug on 2.9.1b which is hard to find unless you check your tlog after a flight, hopefully before you crash:
I got this amazing link from one member of the forum! You gotta check it!
You can get it here!
Thanks mP1 for the link!
Sorry if this is old news, have been following the blog for a long time and this is the first time a see a finished solution for saving your investment.
Thank you, and it is always VERY useful to hear early feedback, especially from people such as yourself.
There are a lot of possible "automatic" triggers for various different actions, apart from the possibility of a direct command from the autopilot.
Triggers will (probably) include:
- Loss of uplink (as detected by loss of uplink heartbeat packet)
- Loss of RC control channel
- RC control channel itself (specific range of pulse width)
- Geofence violation (up to 2 fences)
- Geofence device failure (as detected by loss of geofence device heartbeat)
- Autopilot malfunction (as detected by loss of autopilot heartbeat)
These will all be software configurable (at least that is the intention).
Well, good to hear that it'll be able to communicate with the controller. A CAN bus would work with the PX4 so some hope there. With no IMU then you must be relying on the autopilot to tell it when to fire.
Anyway, I'll stop trying to drag too many details out from you and instead wait for it's appearance and hope to be one of your early customers.
I'll chip in a little. A difficult decision when making an external flight termination device is how much complexity to allow. Additional features (such as an IMU) do enable features such as you propose, but at the potential expense of reliability (i.e. hardware faults, unhandled/unpredicted boundary conditions, sensor deviations from ideal, interaction with centifugal forces as you mention, bugs). I tend to opt for simplicity over complexity (and hardware over firmware where possible) in order to maximise reliability, but it is always a tough call. I'm sorry to say that there is no IMU planned at present.
RS232 (for monitoring radio telemetry uplink)
CANbus (for communication with Autopilot)
SPI is included on the board (but no immediate plans for it)
NO I2C. Despite its popularity, it is just not reliable and I cannot use it in a device that absolutely must be reliable.
Yes, i had a crash two days ago in which a motor came off and I suspect a parachute would have helped me there as well. In the opale-productions instructions it says it'll slow a 2.5kg copter to about 4m/s which is still pretty quick but hopefully enough to reduce major damage.
The opale-productions video makes it look like the parachute can deploy within 3m although maybe it takes a few more meters to slowdown.
If you're going for an independent deployment system, It would be great if it had some sort of interface (like an SPI or I2C connection) so that the controller can do things like configure the parachute. I imagine there will be some config like "deploy when leaned over more than X degrees", etc. Also the controller could tell it things like "I'm armed so get ready!". Obviously it's your system, your choices, just wanted to get my wishes out there..
Also I imagine if you have an onboard IMU that is measuring lean angle it could have trouble with the centifugal forces due to very fast accelerations. I hear this issue has cropped up with the alex-mos gimbals which also have their own imu. A connection from the controller could provide that extra information.
I totally see the merits of having an independent system. When all goes wrong you may not want to trust the controller who may be the cause of the problem!
We have tested this parachute by attaching it to a 3DR quad and dropping it from another vehicle. The chute worked quite well and is much more affordable than these other options.
Darrell, I agree, 2 of my 3 crashes I have seen clearly how my copter falls (One was a lose battery, one was a lose propeller, the last one mentioned in the post). I had time for sure to trigger a release, and I was above 10 meters, so for sure I would be able to have a "better" crash. Also, the big cloth would be helpful to find the copter. Speaking also two the only two guys that I know do this professionally with a Cinestar... they didn't knew that something like that existed.
Any more thoughts or opinions on parachutes for multicopters? Maybe we should keep this conversation on topic..
You got the software for free. Lots of guys put in a lot of hard work and kindly share with us.
NASA pays crazy figures and they crash as well. Surely you know when you push technology stuff goes wrong and thats life. Just buy lots of spares and be happy that copters are relatively easy to take apart and fix.
So shut up and be happy. Its only a few dollars.
Perfect, stay with version 2.8 and fly fun! Keep reading the forums and when nobody mentions a problem in a specific version, then upgrade. If you are such a pro, you understand that's the best option. That is why I am not in 3.0rc2 yet.
Also, pop ups in the software... pros read the forums before updating the software, trying to find if the software is 100% stable before flying 60 meters... 5 times. But I remember my old AR Drone and I-tunes had it every time I upgrade... Is that what you think we need?
Please just enjoy building and flying your copter with the version you feel more confortable. I enjoy posting this as it would help people understand even "Pros" have problems with open source projects.
Nah i was not offered an RMA some guy said he would buy it off me if i were to get rid of it.
Again i had been flying for months on 2.8.1 and years with other boards and solutions with no issues like this, so saying "buy an RTF" "buy a DJI" is not helping, I am not a noob,i re-flashed the firmware and found that it would develop the issues after each successive flight and those issues would vanish with a re-flash. for the first two or three flights before rearing up again.
Also what use is a disclaimer that no one reads in section 15 under downloads, i mean somthing in missionplanner that pops up and says
"Warning New software may have bugs and issues not yet fixed which could cause issues for you, do you still wish to update?"
So stop assuming i am a noob, I told you this in the other forum, i have flown for a long time and designed several different laser cut copter frames, i have lots of experimentation experience and am proficient with electronics., what was happening was clearly a software issue as it did not occur with 2.8 and the multiple posts about issues with 2.9.1b popping up just confirm this.
The whole reason i became aggressive was from frustration over people like you who just assume i am some no body consumer with a wad of cash who thought one day that i would just dabble in multi rotors then complain when it didn't work out the box.