Hi all -
Took my newly-completed quad outside today for it's first outdoor hover testing.. It's been working perfectly indoors (tethered) for tuning and setup.
Everything started out fine, GPS lock, telemetry ok, armed successfully in stabilize mode. Immediately after arming, it climbed about a foot off the ground, slid about 8 feet an crashed (broke 3 props, no major damage). Radio was unresponsive. I never moved the throttle stick.
The "cause" of this event is evident in the logs - snapshot graph of input RC channels attached at th bottom of this post.
I brought the quad back into the shop immediately and observed similar behavior on the radio calibration tool in MP. (erratic readings only somewhat related to stick positions).
Tested my Tx on another model (Blade mQX), no problems.
Haven't been able to reproduce this since then, and it never happened before in all of the tuning I've been doing over the course of the last 3-4 weeks.
Could this be anything except a bad Rx? Here's my setup:
Frame: DJI FW 450
motors, ESCs; telemetry: 3DR
APM 2.5, 3DR power module
Props: APC 10x4.7
3300mah batt.
Tx: DX8
Rx: HK OrangeRx (DSM2)
Connections are all tight. The short cable between Rx and APM are anchored with a spot of hot glue.
Possible mitigating factor: It was cold outsie, about 18F.
The easy culprit is the cheap Rx, but it's never given me problems before. Looking for other theories before I replace the Rx.
Thanks -
Scott
Tags:
Permalink Reply by Will Snodgrass on January 12, 2013 at 7:34pm That dive that channel 3 took could have triggered the failsafe (which by default kicks in when channel 3 goes below 975) so it's possible that it was attempting to RTL. The GPS your using is usually accurate to about 1 meter, so that could account for the distance it traveled. Check the Log and see what mode it was in just after it was armed.
Also check to see if that is indeed what happens when your RX loses signal, If so it's possible the power switch on your TX wasn't all the way in the on position, or possibly some kind of momentary interference.
Hi Scott, I should bring my laptop to the field in order to capture the data and plot some nice graph like you did. I'm also using a DX8 but with a Spektrum AR8000. My last crash was also bizard..... After I tested Loiter, RTL, etc... I switch to AUTO and the Copter goes the wrong way... not Waypoint #1... and I try to switch back to Stabilize but the Copter was so fare away I could not see the orientation of it... So it crashed... Except for props and Copter legs... no major damage.
But I have a situation with Channel 7. I can't use it. When I monitor this channel it goes from low to high for a fraction of a second every 10-15 seconds... Couple of months ago I got a hard time to make the Copter finish a Mission since Chan7 was set for RTL... No wonder why the Mission was interrupted... But I also have another issue that I can't explain. I'm using A11 on the AMP2.5 to drive a servo for pitching the GoPro. The servo connected to A11 glitch (move just a bit every second or so... like a HeartBeat). I know it is not interference with the 3DR Radio or ImmersionRC Video Tx since I still have that glitch when they are power down. Is it APM2.5/ArduCopter 2.8.1 or the AR8000?
But thanks for posting Scott, I will for sure analyse Mavlink RC channels now that I know that I can do it...
I will follow your post. ...other members using Spektrum gears? with good or bad experiences?
Looks like you've done some good diagnosis of the problem. I guess I'd recommend disconnecting the telemetry and see if the problem improves. It could also be a very uneven power supply I guess.
Maybe try moving the rx further away from all the other electronics.
The final possibility is we've seen some bad results with very fast tx/rx systems like the futaba. The issue is that the rx sends all 8 channels at nearly exactly the same time and the APM ppm decoder has a hard time catching the signal. The solution is to upgrade the ppm encoder firmware according to this wiki page. the latest encoder can be found in the downloads area (look for ArduPPM_v2.3.14_ATMega32U2.hex). I think you'll need an FTDI cable to do this...but i kinda doubt this is the problem.
Nice looking set-up by the way.
Hi Randy. That is the question I'm asking myself for the past weeks... Do I need to upgrade PPM encoder? Is APM2.5 have the latest version?
Eagle,
The fix for the futaba issue is very new so it's very unlikely that your 2.5 has the updated ppm encoder. I think apm2.5s shipped from 3dr will have the new ppm encoder on them from next week.
By the way, you should also be careful about how you power the servo. You really want to separate the power used for the servo from the power used by the main cpu. So for example if you're attaching a camera gimbal..sure there are pins on the left side of the APM2 but the power to those pins is shared with the main cpu. the power rail at the back of the apm2 is generally not shared with the main cpu although it can be if you have one of the jumpers in place...i forget which one...maybe jp1?
Permalink Reply by Scott Reid on January 12, 2013 at 8:11pm @Will - After I brought it in and bench tested (with the same thing happening), yes, the bad signal on the mode channel did send it bouncing between loiter and into RTL modes (I have RTL at the highest mode slot based on how my switches are configured) -- that's probably what caused it to launch. I also have failsafe set to RTL (using the Rx's failsafe throttle value). If that happened, I would have expected the RC values to jump to their failsafe values (Thr @ 975)
@Randy - Is your suggestion about disconnecting the telemetry based on possible RF interference? The wireless telemetry was the last thing I installed (added a few days ago).
@Eagle-are you using the satellite antennas on the AR8000?
Thanks all
Yes Scott I'm using the satellite antennas on the AR8000... it is installed not far away from the Main..... it should be far as possible but hey! there is some spaces that we can use :) I don't think it is related. I have the glitches even on the work bench with the Tx just next to the Rx.
Scott,
Yes. I've certainly read a few times about xbees interfering with the receiver signals. I've heard it less with the 3dr radios but I'm sure it's possible especially if they're close together.
Permalink Reply by Allen Bishop on January 12, 2013 at 10:20pm I have had similar issues in the past and recently found out that there were a batch of Spektrum DX8 transmitters that had to be recalled. In the battery compartment at the rear of the transmitter, beneath the battery pack is a lable with a code. Check the code number against a listing on the Spektrum web site (Horizon Hobby) and if it is listed you will need to get an RMA from Horizon to ship it back for an update. The problem with the radio is in the joystick logic. If you monitor the outputs of the joystick channels and you have the problem - you will see a jittering of the output signal for one or more channels. This was a problem I was having for many months. As it turned out it was my radio. Spektrum DX8 transmitter.
They sent me a shipping label, updated the radio and sent it back in about three weeks. No charge to me.
This could be your issue and may be an issue for others as well.
Allen
Permalink Reply by Benny Simonsen on January 13, 2013 at 3:29am Allen,
I guess it's the problems that is described here: DSMX and DX8 Recall (includes some JR transmitters)
If I understand it right there are two unrelated problems. Running DSM2 on some transmitters (thats what you can see on the label) and the stick problem (If you see the link there is a test for that problem).
Besides from that I use the DX8 / AR8000 with TM1000 telemetry without any problems. One thing I really like is the telemetry so I can see the battery voltage and get audible/vibe warnings on my TX when battery is low.
/Benny
Permalink Reply by Scott Reid on January 14, 2013 at 10:48am I do have a DX8 with one of the recalled #s. I bought it on eBay from a seller who said it had been to Horizon and back for the update (I don't any any reason to disbelieve him, nor to confirm as far as I know), and I am using a HobbyKing "OrangeRx" DSM2 Rx. (The recall specifically addresses using the Tx in DSM2 mode).
Regardless, my confidence in the $19 HK DSM2 Rx has faded and I ordered a genuine Spektrum DSMX Rx.
I don't plan on doing any more flying with the DSM2 Rx.
Season Two of the Trust Time Trial (T3) Contest has now begun. The fourth round is an accuracy round for multicopters, which requires contestants to fly a cube. The deadline is April 14th.51 members
24 members
184 members
87 members
682 members
© 2013 Created by Chris Anderson.
Powered by
