apm 2.5 locks servos and spirals to crash randomly in loiter and rtl -

I recently purchased a batch of your apm 2.5 with the 3dr power supply and gps.
I have installed an apm 2.5 in a skywalker 1900 fully loaded with fpv osd and telemtry. There have been two crashed that seem hardware/software related. The first crash occurred on the first test flight after a 18 min flight time . The plane was equipped with a camera and pan tilt mount (manually controlled through the gimbal setup in mav (no auto stab for gimbal) - Manual worked flawlessly, and stabilize as well.. then I tested the RTL which performed perfectly in turning the plane around and returning towards me. The plane passed me and instead of circling it locked the servos at their extreme and put the motor at full throttle which caused and uncontrollable downward spiral of death. I immediately switched to manual to recover control but this did nothing.. the servos remained locked at full throttle regardless of my inputs (even after the crash). After a reset controls were restored. I tried to recover the log files from the apm 2.5 with firmware 2.7 with no success. I began troubleshooting and ended up putting firmware 2.68 on to fix the log file issue and I reset my failsafe on the rx and apm. After repairing the plane I took it out for some testing. I did one test with no osd/telemetry pan tilt or camera plugged in. It was probably a 10-15 minute flight and literally everything worked perfectly- I tested stab, loiter, rtl, and manual thoroughly. After landing I plugged in the osd/telemetry with no camera connected (i have the osd/and telmetry connected by switch so i can have either or working.. in this case I set it to osd only). Again after a 10-15min flight the plane flew perfectly in every mode. I figured I had the problem licked and assumed it must have been a firmware issue or maybe a failsafe glitch.. So I put the new gopro 3 onboard as my FPV cam and hooked it up to the osd and did an FPV flight with full osd info. The flight was perfect for 18 minutes in which I tested the osd in both manual and stab modes. I also used the tilt servo to look down with the cam. After this I came closer to home and initiated a loiter mode.. It seemed to be working perfectly on a large circular path. I panned my camera to the center of the circle and tilted it down slightly. After a complete circle the plane locked up exactly as before and spiraled down violently to a crash. I was at a good altitude so I had plenty of time to switch to manual mode.. but again controls were not returned to me. I even switched back to loiter and then back to manual with no success.. the servos were simple locked with full throttle. My fpv cam captured the whole event so I will attach a link to the crash.. but it shows that the osd did not lock up and that the mode changes were registered by the apm - it began to crash in loiter mode and it showed me switching to manual. I was able to get the log files off of the apm 2.5 and it shows exactly what I have explained (i will attach these as well). The only variables that changed between successful tests and crashes were the plugging in of the pan tilt servos and the addition of a fpv camera (which was not connected in the first crash, but in the second). The only inputs from the tx that I was controlling were throttle changes (in rtl crash and after switch to manual in loiter crash) and pan tilt motion (in loiter crash). no nudging of pitch/roll/ or yaw in either case. Flight time was slightly longer in the crashes than the successful tests. The gps appears to be working properly with no errors even in the log. I am baffled and have found no answers in the forums or troubleshooting guides. The only thing left I can think of is replacing the apm 2.5 and seeing if its a hardware problem. Can you please help because repairing the plane and replacing expensive fpv cams is getting expensive-
Again I will attach the logs of the loiter crash which occurred at event 23 loiter on the .kmz file..
and here is a link to the fpv crash vid with osd -
https://www.youtube.com/watch?v=w4rsIc8LLho

Test results as of 4-26-13-- Well I have bad news and more bad news. I enabled all logs and launched with the new configuration. I did not plug in either pan or tilt servo. I also set my switch between telemetry and osd to the telemetry option. The flight went perfectly- I even let it loiter and rtl for many minutes and my total flight time was well over 25 minutes.

In my next test I included the pan tilt servos and switched to my osd option.. I did not connect a camera to the osd. After a successful launch I put the plane in stab mode which worked like a dream. In stab mode I moved my pan tilt servos several times with no glitches. I then brought the plane to a relatively low altitude (perhaps 100-150 feet) I was cruising at about 1/3 throttle.. I engaged loiter and the plane responded by straightening out then it began to circle.. almost immediately the plane stalled and circled to the ground. I attempted to return the plane to manual mode with no success. Even though I was at a low altitude the plane was very badly damaged.. Especially when compared with other crashes from a higher altitudes (crazy). After downloading the log files I am not sure that any information was captured from the flight- the previous successful flight was recorded properly, but the next .kmz shows no real flight information just a rtl mode.. which could have been the reboot after the crash. I tested the plane after the crash but the servos did not respond to tx input and when i switched modes the servos twitched randomly. I will attempt to get the system running again after some repair and diagnosis. I will attach the logs I do have including the successful test flight.. perhaps this will help in the deduction process.

So the elements that I think may be the cause are 1. the addition of the pan tilt servos (a problem with their power consumption or perhaps the setup is interfering in some way with a circle command <loiter/rtl> 2. The switching between telemetry and osd or maybe the fact that the osd is running with no camera plugged in. (or a power issue because both are powered up during flight) 3. A low throttle command (either coming from the tx or the loiter/rtl commands that are causing an unrecoverable stall and servo lock 4. Some sort of APM 2.5 hardware/software failure

I'm pretty bummed that the new power config didnt fix this strange issue (I had high hopes especially after the successful test). I plan on rebuilding and retesting, but I'd love to have some more input on troubleshooting and possible solutions. Any and all help would be much appreciated:)

all logs from spiral crash tests.zip

Replies

Lloyd Casey August 30, 2014 at 12:53am

Hi,

Did you or anyone else find a solution to this problem? I have had a very similar issue with my skywalker/apm setup and I just can't figure out what it is.
Daniel Echeverri December 6, 2013 at 8:34am

You have to check you CG and what is the minimum speed to fly, you stall and spin is too hard to recover look this video will help you: https://www.youtube.com/watch?v=nh8PbcFY_Gg
Maven Aircraft May 3, 2013 at 4:52pm

So after a few more crashes and alot of testing and troubleshooting I think the problem was coming from my large pan tilt servos. I am working on a power config that might allow these large servos to work properly with the apm 2.5. I am currently talking with 3dr to find a solution. In short the power draw from the pan tilts was browning out the apm 2.5 causing a signal loss.
Maven Aircraft April 26, 2013 at 1:30am

Test results as of 4-26-13-- Well I have bad news and more bad news. I enabled all logs and launched with the new configuration. I did not plug in either pan or tilt servo. I also set my switch between telemetry and osd to the telemetry option. The flight went perfectly- I even let it loiter and rtl for many minutes and my total flight time was well over 25 minutes.
In my next test I included the pan tilt servos and switched to my osd option.. I did not connect a camera to the osd. After a successful launch I put the plane in stab mode which worked like a dream. In stab mode I moved my pan tilt servos several times with no glitches. I then brought the plane to a relatively low altitude (perhaps 100-150 feet) I was cruising at about 1/3 throttle.. I engaged loiter and the plane responded by straightening out then it began to circle.. almost immediately the plane stalled and circled to the ground. I attempted to return the plane to manual mode with no success. Even though I was at a low altitude the plane was very badly damaged.. Especially when compared with other crashes from a higher altitudes (crazy). After downloading the log files I am not sure that any information was captured from the flight- the previous successful flight was recorded properly, but the next .kmz shows no real flight information just a rtl mode.. which could have been the reboot after the crash. I tested the plane after the crash but the servos did not respond to tx input and when i switched modes the servos twitched randomly. I will attempt to get the system running again after some repair and diagnosis. I will attach the logs I do have including the successful test flight.. perhaps this will help in the deduction process.
So the elements that I think may be the cause are 1. the addition of the pan tilt servos (a problem with their power consumption or perhaps the setup is interfering in some way with a circle command <loiter/rtl> 2. The switching between telemetry and osd or maybe the fact that the osd is running with no camera plugged in. (or a power issue because both are powered up during flight) 3. A low throttle command (either coming from the tx or the loiter/rtl commands that are causing an unrecoverable stall and servo lock 4. Some sort of APM 2.5 hardware/software failure

I'm pretty bummed that the new power config didnt fix this strange issue (I had high hopes especially after the successful test). I plan on rebuilding and retesting, but I'd love to have some more input on troubleshooting and possible solutions. Any and all help would be much appreciated:)

all logs from spiral crash tests.zip
Stephen Gloor April 17, 2013 at 5:18pm

Hi - you wrote this "The only variables that changed between successful tests and crashes were the plugging in of the pan tilt servos and the addition of a fpv camera (which was not connected in the first crash, but in the second)."

Where do the pan tilt servos plug in? Are they powered off the APM power supply or the ESC? I am assuming you have JP1 out and are powering everything seperately.
Søren Kuula April 17, 2013 at 7:20am

Hi,

Well the MAVLink output to OSD worked to the end.. APM was not completely frozen at least.

Im thinking in terms of some sensor or R/C defect.

Do you have a tlog file and can you describe your R/C setup? Does that use the PPM encoder or not?

Also describe the R/C failsafe - what happens if you remove the remains of the prop, start up in manual mode and turn off the tx?

Regards

Soren
Maven Aircraft April 16, 2013 at 12:57pm

updated link to osd video file-

http://youtu.be/_9tYc_itH3Q

This reply was deleted.

Discussions

apm 2.5 locks servos and spirals to crash randomly in loiter and rtl -

Replies

Activity