Hi people,
well this is a crappy day. I put this quad together in the last month(s) and have slowly configured more and more, going one step at a time so I could understand it all better and get familiar with everything slowly. Today I finally attempted to get my camera gimbal to be able to keep my camera stable (tilt) in flight, using a single servo. I got the servo all connected, got it to function, keeping the gimbal level as I tilted the quad back and forth. Its a simple setup, or so it should be..... I got the servo working, took off, climbed slowly, did a few rotations in the air, at about 40 feet up, and about 25 seconds into the flight, the quad just shut down, tilted to the right and fell out of the air, basically landing on its top side!!!! &^%#@&*^%@^!!!!
I will give some more details, so maybe my issue can be solved more easily. My quad is 550 sized, Has the pixhawk, uses 3DR PM, has M8N GPS, 3DR 915 Mhz telemetry. ESC is HobbyWing 20A Quattro. The gimbal is a simple tilt only and im using a HiTec HS-65mg sevo for the tilt function. For a BEC im using a quality Castle CC Bec Pro. Off the BEC im powering a tiny Spektrum ultra micro FPV camera that i got with my FatShark V4 FPV goggles. Im using a Y cable to the camera and the other end into the pixhawk to power the servo rail. with this configuration there should be absolutely no reason why the pixhawk suffered a brownout! The pixhawk is powered off the 3DR power module with the standard power wire, and the BEC is also plugged into the pixhawk PIN 6, and the servo into the tilt (aux out S1), as indicated in the 3DR (Camera Gimbal with Servos) WiKi.
i should point out the CC BecPro has two 5v outputs, so I could power the FPV camera and servo rail separately, but initially I didnt see any reason to as the camera is so low draw, and a single mini servo like that is so relatively low draw, I didnt think there would be any issue running both off of one lead from the BEC.
it seems like there is so much to digest here, but i really hope someone here can LMK what the issue would be, as I cannot let this happen again, as so far it doesn't look like I suffered any damage other than one wire being torn off one of my motors, and ill only be able to tell if that was it once I re-solder it and do a test.... And, with that in mind, how could this even happen?? I have the 3DR PM powering the Pixhawk with its special power wire right of the Module, AND there is a 5V wire coming from the BEC to the servo rail, which, should act as a power source as well, so HOW could I suffer a brown out from just one servo, and a small FPV camera running off the BEC?? makes no sense??
please LMK what you guys think, as i have no confidence in this system anymore, I cant fly anymore if there is a chance its just going to shut down and fall out of the sky with no notice.....LMK if you need any more info.
thanks in advance!
Ginnz.
Replies
The voltages for the both the servo rail and pixhawk supply look reasonable. I don't think this is a brownout situation.
Did you launch from a surface much higher than the ground?
The "Alt" data appears to show you flying above and below your launch height. You then flew until your battery went below it's minimum threshold and generated a failsafe error. The copter immediately went into land mode. Once it thought it was on the ground, it cut the motors. The last line of the log clearly shows it a landing complete message (you wouldn't see this if it was a brownout as the logging would abruptly stop mid flight). I know there have been issues with false positives when landing with older version of the software, what version are you using?
Phillip,
Wow, thats what the data tells you? jeez, that's completely crazy, s I was flying it in the spot where I have been testing all week. I took off from the ground, as per normal..... maybe the pixhawk gps only got lock, or it only started "thinking/logging" after i was airborne, thinking i only took off at altitude? it all makes sense that it thought it took off at around 25-30 feet up, as it was that high when it just shut down and fell.
also, voltage wise, that is all wierd, as I was flying on fully charged 4s packs (2-2200 mah in parallel), the packs I always use and have no issues with. Now, Ive ordered a new 3DR PM as initially it seems I was having issues with mine, as it would always shut down my 3S packs showing they were drained and going into failsafe, even though they were full. Since I started using 4S packs only, I can at least fly.
So, wha tyou seen in the log doesn't have anything to do with the servo on my camera gimbal? it was just a coincidence that on my first flight with the servo running to stabilize the gambal it went wonky and fell out of the sky? You dont think it was a receiver re-set/brownout from a current overdraw off the servo rail or anything? I should be safe to hook it back up to stabilize the camera?
I need to work on reading the logs so I can figure this stuff out myself! There is so much information that a guy can glean off them!!? And thanks alot for your help! your opinion on this stuff is invaluable to me setting this thing up further! and as far as set up, i should have the new 3DR power module, and 4 new 3400 MaH 4s- 40c packs tomorrow!
Thanks again, and LMK if you think i should be safe hooking that servo back up?
--edit--
oh, yeah, my software version is MP 1.3.31 Arducopter v.3.2.1, I believe its all up to date.
Doug (ginnz)
I'm not an expert at reading the logs but your Alt graph looks weird to me, so maybe something else is going on there. Other graphs have clearly shown all altitudes above zero (launch height) with a landing at nearly the same altitude. Your graph shows movement above and below zero, but nothing anywhere near 25-30ft of movement. Is it possible that your barometer is plugged or defective (I would expect to see an alarm if that was the case)?
Irregardless of the Alt issue, the aircraft definitely thought it was landed at the time it shut down the props. In v3.3 (soon to be released) I believe the landing detection was improved over v3.2.1.
I don't think this relates to your gimbal servo, but I could be overlooking something.
Yeah, ok, BTW, I downloaded the log, and tried to review it in MP, but I really dont know what im looking for!? I can see where it went haywire, but there is so much data, its confusing. I downloaded it in .bin format and will post it here.
2015-08-23 13-18-02 11.bin.log
Ok, I dont have a "zenier diode" wired up to the servo rail/bec input? could that have been the issue? This whole additional "zenier diode" thing kinda miffs me, why in the world could 3DR just not make it simple to power the servo's like any other system?? Or have built this into the system? it seems like sime kind of cheap after thought that a guy would have to hokey pokey rig such a thing into the system. And where can I get oine of these things?? are they popular in normal circuits? if so, what would be a good example of a normal circuit where I could salvage one? (VCR? sat. receiver? ).... maybe the lack of one of these caused my pixhawk to reset? I cant see why this would happen, I only have one simple analog servo connected to the gimbal.... if one servo could cause this, i can only imagine what three could do!!??
TIA, Ginnz
Anyone able to make sense of my log file??? can someone please help me determine what happened?
opps. I posted the .bin to my initial post... oh well, same idea. Well, I re-soldered the torn wire last night, re-mounted the Pixhawk (it came undone from its anti-vibration mounts), checked all the wiring, straightened out the GPS antenna, and test flew it. It all seems to have survived ok! i can't believe it. It basically dropped straight down from approx 40 feet! I cannot have that happen again! I think I will get rid of the Y cable and run the tiny FPV camera from one dedicated line from the BEC, and run the other line straight to the Pixhawk pin 6 for powering the servo rail, as shown in the instructions. I still cannot for the life of me figure out why it browned out if the 3DR PM power wire rund the Pixhawk, AND I had a power wire from the BEC to the servo rail?? I mean, if for any reason the servo/camera was overdrawing power, then the dedicated power wire from the PM should still power the Pixhawk?? Anyways, I hope someone can make heads or tails from the log and LMK what the deal is?
TIA, Ginnz