Bummed. New P3A Crashed. Pls Advise
- Thread starter McVenturesRC
- Start date
- Joined
- Sep 21, 2014
- Messages
- 79
- Reaction score
- 3
Do NOT reuse that battery from the crash. It'll be the death of your NEXT drone. I say this from experience
One of you guys determined the .DAT file I provided didn't match up properly to the .txt file. I've gone back in to download another file. Check this out and see if it matches the .txt (healthy drone) file. Thanks.
Dropbox - FLY014.csv
Dropbox - FLY014.csv
Do NOT reuse that battery from the crash. It'll be the death of your NEXT drone. I say this from experience
Thank you for the advice.
- Joined
- Oct 16, 2015
- Messages
- 548
- Reaction score
- 123
- Age
- 58
Budwalker, I checked flylog.info and could not find the tool DatConverter. Can you send a link?
Are there tools to retrieve the GPS data in (.GPX or .KML)?
To be honest, DJI does not make very clear how tight the props should be. Sometimes they use two hands, sometimes only one. The amount of force is not mentioned. And yes, the wrench is also mentioned to tighten the props. See DJI Phantom @ t=7:22. DJI Forum|Phantom 3 Professional/Advanced Tutorial Collection
DJI implemented a special startup scheme to prevent loose props. It seems unlikely that this is an OP mistake. Power loss to the motor still looks a valid explanation.
Attachments
It's in the "Offline Tools" section. Second paragraph, at the end of the first sentence. I'd send you a link but you need to read that section.
Yes... You are correct. No-one has ever produced that before. Pre-flight Checklist video.. At the 7:22 mark.
Thank you
The motor startup routing really has no affect on prop tightness. I tested this in the past when I had an electric plane. And I can make a propeller come off by stopping the motor if the motor startup is all that is depended on for tightness. setting a prop on the motor and starting the motor leads to a more tight prop, but I can still get that situation to let a prop fly off.
I cant reproduce a flying situation to test what active breaking does, and how it affects a "loose" prop without a crash. Its just not smart.
Could you test it my holding onto the phantom landing gear while it tries to hover and rocking it around, seems like it would do whatever it could to try to hover....
- Joined
- Sep 24, 2015
- Messages
- 47
- Reaction score
- 17
- Age
- 57
I think that's similar to how people get hurt hand catching - if you try to move the bird by hand it's going to try to resist, possibly tipping and hurting you.
ianwood
Taco Wrangler
That looks very much like a bad calibration. Resulting value was way out of range. This is a good example of not being able to depend on the P3 to tell you when the compass is not calibrated properly.
No it doesn't match. Top is the txt bottom is theFly014.csv.
When you convert the dat file into csv check the column gpsAltitude in excel - if its same height through out its not the right one
If you look at the amps in the last post, top part, the battery had to be connected until the end to draw that much...and record data. It also rules out CSC since that would also have killed the current.
100% of the information we have says this crash had power all the way till AFTER the initial impact, and was applying thrust to either 3 or 1 propellers. There is no doubt that the a prop came off in flight, and the prop falling was the cause of the crash. With the prop coming off it release the balance of flight, as well as torque to airframe, forcing the direction of rotation that is shown in the log file Its pretty much in DJIs hands as to whether they will, and to what degree, they might help.
Turns out we were looking at the wrong file. The file dates on the P3 often are incorrect (see the "Retrieve .DAT file" section on flylog.info) but this one was incorrect in a whole new way. Anyway, the flight recording lasted for 378 seconds. I peeled off the last 14 seconds to show here. Bottom line is that it would seem the battery was intact all the way to the ground. I made some plots but don't have time right now to look closer. If anybody else wants to take a closer look please do; I've attached that 14 second .csv.
Note: the yaw swings from +180 to -180 because it's passing through 180
There's all kinda cool plots to look at. These are just a few. It'd be interesting to see if anything can be learned by looking closer when the event occurred. Did one corner pitch down like what would happen if the prop came loose. The .csv should be inside that .zip that's attached.
Note: the yaw swings from +180 to -180 because it's passing through 180
There's all kinda cool plots to look at. These are just a few. It'd be interesting to see if anything can be learned by looking closer when the event occurred. Did one corner pitch down like what would happen if the prop came loose. The .csv should be inside that .zip that's attached.
Attachments
Initially, a KML producer wasn't high my to-do list. But, I can see now that it'd be useful and it's not to hard to do.
We knew that from the first bit of data we got. The cant be recorded data after the impact if there was no power. And there is no way for the 4 pints to have contact and not the 2 large connectors on the battery because the 2 large blade connectors have a deeper penetration into the female slots than the pins even stick out.
Not trying to knock your work, or your tools @BudWalker.. just saying that your repeating what was always known.
Here is the best chart I could make, it would appear that the pitch and roll diverge at the same time. Based on earlier data in the flight and assuming the craft flew mostly forward, it would appear that positive pitch is nose down, which seems opposite to me. Anyway, if that is the case I would guess the intact prop was one of the front props. At the same time roll went very negative. I am not sure if that is bank left or bank right, but if it follows normal aircraft convention negative would be lowering left side which would indicate the left front prop. If someone knows the sign convention it would be interesting to see if it matches the intact prop. I also checked all the accelerometers and gyros to see if there was anything that occurred prior to this (collision) and found nothing. As a result I believe I agree with a lost prop during flight....
Attachments
You can tie this same point in time to the sudden change in compass data. Even more clearly showing a loss of lift in one corner. The direction of rotation defines the color of blade that was lost.
Similar threads
