While I think I have a feel for what's up I figured I should run my issue by the forum just in case there is more that I should be looking at. 

Before I describe symptoms I think my cause is a loose receiver connection to the FC.

So here is the overall scenario

Friday flights @ tree infested lake area: 

• Flys fine, several small crashes but nothing too serious
  

Saturday flight: 

• Got it in the air, went to do a full (360°) roll. Quad rolled exactly 540° and stopped. 
  
• Flew around, tried again for half (180°) roll. Quad rolled exactly 540° and stopped. 
  
• 3rd attempt same thing. Quad rolled exactly 540° and stopped. 
  
• Shortly after, lost control quad plummets into the ground. 
  
• (I'll post video of all of this later, it's rendering this morning)
  

Other saturday flights without gopro

• Did more of the same but also noticed loosing control briefly while not doing acro moves. 
  
• Did a flip (thinking maybe something was wrong with gyros or FC) and Quad flipped exactly 540° and stopped. 
  
• Other attempts to fly resulted in odd fly offs, awkward losses of control, and crashes. 
  
• At this point I started to think that somehow something was wrong with my radio because the Ghost Priat wasn't flying right either (was actuall lack of air mode) and even the vortex had some issues (is just a sloppy flyer and issues were actually video related. The guy I was flying with was having VTX issues and I think it changed to my channel)
  

Sunday flight

• Troubleshoot of radio by flying all 3 quads revealed that only the bumblebee actually is having problems as mentioned in the last bullet of previous list.
  
• Bumble continues to lose control in mid air
  
• Radio does not warn of loosing signal at all. 
  

I think I have blackbox logs, or I tried to record them but I haven't looked yet. Since I am certain that it's related to the radio system and BB is armed with a switch there is a small possibility that something could have issues there (possibly cut offs could interrupt logs). I'll be looking at those first before taking it apart to check solder joints @ sbus connection. 

Anyway... anybody think I might need to check some other things besides receiver? This'll be the first time I've disassembled the bumble since building it so I am going to take the opportunity to clean up some of the wiring on both the receiver and video.

oh also, in case it helps:

FC=Omnibus F4 clone (the black one from Banggood)
Receiver=X4R-SB
The rest of the components in the build thread

what firmware version are you using? and what looptime are you using?
one thing that could happen is CPU load is too high and it interrupts RX signal and cause jitter.
But thinking about it it shouldn't be the case for F4 FC's.. more of a problem for F3 on 8K/8K...

Blackbox should tell you more what the issue is.. if there is problem with the RX switch, and stopped your recording, then you should be able to tell by the length of the log

(31-Jul-2017, 02:31 PM)Oscar Wrote: what firmware version are you using? and what looptime are you using?
one thing that could happen is CPU load is too high and it interrupts RX signal and cause jitter.
But thinking about it it shouldn't be the case for F4 FC's.. more of a problem for F3 on 8K/8K...

Blackbox should tell you more what the issue is.. if there is problem with the RX switch, and stopped your recording, then you should be able to tell by the length of the log

firmware is 3.1.6 I'll check looptime tonight I believe it's default, but other than swapping out a motor recently I haven't done anything new to it. This quad has been the best flyer that I have hands down since I built it, it's why I've become so critical about everything else I've been flying. So for it to go from being a good flyer to intermittent failures without any serious changes seems like there shouldn't be a difference in CPU load. I'll have a look at BB and looptime tonight.

Here is DVR footage of the rolls as mentioned above:

yea could be RX signal jitter... i have something similar in a new quad I'm currently testing, it's fixed by swapping out the FC. It seemed like the old FC had a dirty 5V which caused all those signal jitters in the RX.

https://oscarliang.com/ctt/uploads/2017/...t-2017.jpg

But until you can verify that's a RX jitter in the BB log I would also check the RX, try a different RX if you have a spare. Also try to update the firmware with erasing all the previous settings.

Maybe check the rx wires as well, could be there's something broken and loses contact on rough movements? More than once i had wires break right next to the solder joint, but i was using suboptimal wires, not really flexible, so in my case it was only a matter of time for this to happen...

If the logs recorded something, it should be visible when/if the rc signals go flat.

Gyro=8kHz
PID loop=2kHz

I don't think it's that.

It looks like it's the receiver based on what the RC command is doing. All of the sudden all inputs go to center and throttle goes to zero. Then they pop back up again with additional input showing my gut reaction to try to correct (and each time I was able to correct effectively). 



I'll check my wiring this weekend probably to figure out if it's the wires or the solder joints or what. Either way those wires are way too long. 

I may just swap out the reciever, I have 2 more XSRs sitting around just waiting for a quad to go into. It'll kinda be bad troubleshooting because it wont tell me if the X4R has a problem or not but it would mean I can have a smaller receiver in there.

okay yea it does seems to be the faulty RX or connection keep us updated

(01-Aug-2017, 10:49 AM)Carl.Vegas Wrote: It looks like it's the receiver based on what the RC command is doing. All of the sudden all inputs go to center and throttle goes to zero. Then they pop back up again with additional input showing my gut reaction to try to correct (and each time I was able to correct effectively). 

That blackbox log shows exactly what happens in a failsafe situation.

Where in the craft is your receiver? I've had problems when putting a receiver that doesn't have any shielding too close to other components vastly reduced the receiver's range as it could hardly hear the signal over all the interference it was picking up.

it's not quite hitting failsafe... I've been worried about that. Each event is usually less than a full second so far.

It's attached to the top plate of the frame sitting about 5mm above the FC... and now that I look at it the antennas are touching the flight controller so you might have something there.

No shielding on the Rx, but it's wrapped in electrical tape to avoid shorts if it comes loose in flight but that's about it really.

I stuck something in there and found that the solder joints are probably not the problem after all so it's either a faulty receiver or interference rather than wiring I think.

I might try rerouting the antennas and put it in the air to see if it's better... That would be better than a more involved repair

I didn't mean a failsafe in terms of the flight controller considering the RX as down. What you're seeing is the RX losing signal and what it does is to drop the throttle to zero and centre the pitch, roll and yaw channels - an RX failsafe. When I had this problem, the RX failsafe lasted for no more than 1/3 second, but the sudden loss of control for an instant was very obvious when it happened.

From the pictures of your build, the RX antennas and the RX itself is very close to your VTX. That's more likely to be the problem than your flight controller unless the receiver is right above the switched mode regulator on the Omnibus. Switched mode regulators spit out loads of wideband interference.

If you can, put the RX antennas at the front of the craft.

---

Or put them facing forward from the rear arms like I did on the Rebel 220 build. It's a great place for them.

(03-Aug-2017, 09:39 PM)unseen Wrote: What you're seeing is the RX losing signal and what it does is to drop the throttle to zero and centre the pitch, roll and yaw channels - an RX failsafe.

But... wouldn't the FC do the same if it lost connection to the RX? Also from Carl's 1st post, a warning on the transmitter would be expected...

I tried to think of ways to test things out but none of my ideas seem safe... unless you manage to reproduce the issue on the bench without props. Then, you could for example change FC failsafe behaviour for throttle and log it, to see if it's FC or RX failsafe.

Good point. Yeah no "telemetry lost" message. I can do a couple of things go test it this weekend actually... first I can hook everything up to USB and shake it around to confirm my first suspicion. Next I can turn the accelerometer back on and set up cameras to watch my radio while I fly on a safe spot to see what all is happening. Of course with both of these I can tie antennas up better too...

After all of that if I am still having the problem then I'll pop one of the XSRs in it and try fhat.

Jury is in... I hooked the FC up to USB, then powered on my radio and shook it around and was able to recreate signal loss... then I wiggled wires and loss signal almost entirely. 

Here was the problem:


It was a very poor solder joint on Rx + power... in fact at the time it was the first time I had soldered this small of a joint ever before so it was just utter crap... already I am better at soldering than that and I dont exactly have a ton more experience. 

so I resoldered all 3 'cause they were rubbish, and because I needed to shorten the wires to the Rx and now it's clean. I also redid the joints to power the FC a bit just to tidy up a bit (unfortunately forgot to get a pic before putting it back together. It's cleaner you'll just have to trust me on it)

I am taking it out to fly tomorrow. I expect all will do well.

After flying 4 batteries I can confirm that it was the solder joints. I may or may not upload video of today's flights. I'm not sure if they're interesting enough or not lol.