Wookong Woes

[modles]

Here is a photo of my lower plate setup. I guess the only non standard thing is that I am pulling 5v out of the A port for my fpv camera, but this is very low draw. Object just visible under the right side is a 5.8ghz video tx, but this is not powered at present, so can't be causing interference with the gps. There is also a DJI 2.4ghz datalink TX under the mcu, but again this is not connected to sbus/power, so its a pretty basic setup.

View attachment 9267

 

[hjls3]

not sure about the WKM but i do know having GPS cable so close to NAZA gave my naza some troubles. Nor familiar enough yet with the WKM to know if it makes a diff or not.

 

[BorisS]

You can also try flying it without the GPs module. But be careful about it. I did this 3/4 of a year ago and back than it was not a problem, don't know if anything changed in the code in the meantime ! I personally think something it smacked up in the can-bus communication or the compass alone.

You should really smack on the Naza and give it a fly, than you more less know it must be the WKM.

Boris

 

[modles]

Yes I think a can bus error is likely, especially due to all the odd errors showing in list and perhaps linked to why the iosd did not work?

Will try the naza on Tuesday if I can.

 

[swisser]

I've not read every post on this thread in depth, but skimming over some of them I think you're getting some confusing views which might not be helping. You need to take a step back and keep things nice and simple. A few things leap out at me:

1) Use the setup assistant to verify that the the controller is getting the stick inputs you're making. This rules out a whole load of issues, including stuff about can-bus etc.

2) Don't use GPS mode yet. I'm not suggesting you don't want to have it working properly eventually, but whilst sorting out basic flying/control issues it just brings a whole load of extra variables in to the equation. When you're not using GPS you're not using the compas either, so you can ignore all that as well. Don't unplug it, just take off, fly and land in Atti mode. Confirm in step 1 above that you're selecting Atti mode correctly.

3) Stop worrying about gains. Apart from if you have gains set way to high and you get a rapid oscillation they really make very little difference to the basic flight characteristics. I've got a video somewhere where I show gains from 50 to 200 and you can't tell the difference in the video. Use the defaults.

Once you've got the basic control issues sorted you can then investigate GPS mode. If it flies properly in GPS mode then return to home mode should work fine. Only after all that should you experiment with changing gains. Or, I should say, you can experiment with them all you like but you're wasting your time.

If you want to talk it through on the phone send me a private message. Or I'm in Norfolk quite often perhaps we could meet up and go through it.

 

[modles]

Hey, thanks for the input, but it might be worth reading all the posts for the full picture.

she flies happily by stick in all 3 flight modes, apart from yawing to the left on rapid accent when running 3s. note that this is also the case in manual mode. With 4s no yaw issue.

all controller inputs are good.

its definitely not a gain issue as accent and decent are fine and stable in all 3 modes.

gps hold works great, took it out in heavy wind today and it held position impressively.

IOC switch allocation is correct, end points calibrated and verified in assistant software. But no influence to flight at all.

fly to home and fail safe works as in it goes up flies home etc but it does it in a terrifying and erratic manor, bucking like a bronco.

If I walk the copter away from home position and release it flies home smoothly. So the problem is only when it's running fly home routine.

what I might try next is different fly home settings. Ie no altitude increase etc.

 

[swisser]

"IOC switch allocation is correct, end points calibrated and verified in assistant software. But no influence to flight at all."

Do you get the green LED responding correctly?

 

[Aviator]

I helped someone today saying the IOC was not working? but the problem was he was not setting it by switching between off and course lock 3 times first... After he understood this of course it worked everytime.

Also out of interest have you tried failsafe using 4s lipo?

 

[modles]

Ah this may be it for IOC issue. Thanks for the heads up.

so if i reluctantly accept i can only fly on 4s for some reason that hopefully just an error on my part it just leaves this buckaroo style RTH

and yeah, RTH / fail safe is still screwy on 4s

 

[Aviator]

O.K What I tend to do for the IOC is stick it on a two position switch so that you cannot go wrong as you only have OFF and COURSE LOCK then create a mix that changes that same switch into OFF and HOME LOCK or whatever your preference when another switch is switched if that makes sense? Just so I get this right you have the stock 920kv motors and the stock DJI 10" props and 3s or using 4s and DJI 8" props?

 

[modles]

Yep that makes sense re switches.

Currently I am using DJI 8" props with 3s and 4s on stock 920kv DJI motors that come with f550 frame. I have not tired DJI 10" props yet. With 3s I get a reluctance to yaw right and a heavy yaw to the left on power. I have to trim about 3/4 to the right to keep it straight in normal flight. On the 4s I don't have the yaw problem and can power on hard with no induced yaw, and no trim needed.

 

[SMP]

While we're on the IOC topic. Heads up on this http://www.multirotorforums.com/showthread.php?8770-DJI-Flamewheel-550-Crash-IOC-Failure

Highlights:
Must be at least 30 odd feet away from home before it will play ball
and AUX2 to 100% on Spektrum DX7 turned the bloody motors off and crashed like a brick

 

[Aviator]

Yep that makes sense re switches.

Currently I am using DJI 8" props with 3s and 4s on stock 920kv DJI motors that come with f550 frame. I have not tired DJI 10" props yet. With 3s I get a reluctance to yaw right and a heavy yaw to the left on power. I have to trim about 3/4 to the right to keep it straight in normal flight. On the 4s I don't have the yaw problem and can power on hard with no induced yaw, and no trim needed.

O.K that makes perfect sense to me.. The reason you have the problem with yaw on 3s is because the smaller props running at a lower max RPM will have less torque than if you used the 10" props on 3s.. So if you use 3s stick the 10" on 4s use 8"
As for the instability I am still thinking about that?

Just reading through the post again... Please unplug the camera from your WKM just to eliminate this from the equation, not sure how you power the VTX but you dont want any ground loops..

 

[modles]

Ok this makes sense I think, and I take that the reason you don't see this phenomenon on the 450 quad is that there is and overall reduced amount of torque effect because of the 2 less motors? I am surprised however that no one has mentioned this before on the forum or that it is not mentioned in the manual.

I guess testing with 3s on the 10" will prove/disprove this!

I will unplug the cam from WKM and run this off a separate switching regulator. VTX is powered directly from the main BUS. I figured this will be fine as the PMU should have more than adequate filtering. Do you think instead I'm best to run this from a separate isolated supply?

 

[Aviator]

Ok this makes sense I think, and I take that the reason you don't see this phenomenon on the 450 quad is that there is and overall reduced amount of torque effect because of the 2 less motors? I am surprised however that no one has mentioned this before on the forum or that it is not mentioned in the manual.

I guess testing with 3s on the 10" will prove/disprove this!

I will unplug the cam from WKM and run this off a separate switching regulator. VTX is powered directly from the main BUS. I figured this will be fine as the PMU should have more than adequate filtering. Do you think instead I'm best to run this from a separate isolated supply?

As they say assumption is the Mother of all F ups, you may well be right that it is o.k? but I wouldn`t chance it.. Especially if you have a component go down on your cam or VTX and it sends 16v through the system.. Always isolate other systems from your main flight control, thats the principal I work to, and keep the grounds tied together so using one Lipo and feeding everything from that will be fine.. Try a switching regulator on the VTX, some are noisy and create interference others dont? if you have problems switch to linear regulation.

Qoute:
"Ok this makes sense I think, and I take that the reason you don't see this phenomenon on the 450 quad is that there is and overall reduced amount of torque effect because of the 2 less motors? I am surprised however that no one has mentioned this before on the forum or that it is not mentioned in the manual."

The manual generally keeps it simple and tells you what to do rather than cause and effect...

 

[Tomstoy2]

Actually, it's been mentioned more times than I care to remember.

Run your vtx and camera off a seperate battery, never off the main pack. This wouldn't suprise me at all if it clears up a ton of your problems. Right now, if your vtx and or camera is powered up by the main battery you are probably swamping the hell out of your system.
Always, always keep systems seperated, especially your fc system.

Keep the vtx as far away from other components, they are noisey as a kid in the dentist chair! If you can, get a piece of cf and keep it inbetween the vtx and the rest of the components. This will block any noise from going where it shouldn't.

Twist all your wires, this will help with reducing noise, also. Remember, it's all accumulative.

Your gps should be as far away as possible, also.

Keep your wires, low and high voltage seperated. Don't allow any to overlap the others.

Secure anything that is loose and can bounce or vibrate.

Props need to be balanced to a fine tee, both the blades and the hubs. This is really easy to do, but time consumming, and it has to be done.

As far as gains go, basic gains on a 550 size will run around 130-170ish. This all varies every time, even if you have 2 identical platforms.

Read the manual over and over and over again, until you have it firmly fixed in your head, especially the LED codes!

The whole battery/prop thing is pretty much true, it's all a balancing act. Going 4s is great, less amp's consummed, but remeber you will be flying with more weight, unless you lower the capacity to equal weigh out, but this means less flight time.
It's all about experimentation to discover what works for you best, but right now your a long way from that point.

Go back thru your entire set-up, paying particular attention to what I described above. Then report back your findings. I think you'll be happy you did.

99% of the time, all problems will be operator related, not the DJI. Sure, it happens, but damn near every time it's something you did or didn't do right.

 

[modles]

I hear you on that, but its worth mentioning again that the VTX is not powered at present.

Also just tried without the camera plugged into MCU, same buckaroo effect.

So it looks like we can rule both Camera and VTX out as causes.

GPS is mounted on the longest pole, 15cm above IMU, and obvs correctly set as -15 in the assistant.

I will re-route all cables, but again this does not make sense as a cause, as this would be affecting the IMU etc at all times, not when just in RTH mode.

I understand that its hard to read through all the posts in this thread, but it's kind of the only way to get the full picture of what's happening. I do really appreciate the input though, better to have suggestions and rule them out than not have them at all!

 

[modles]

Always isolate other systems from your main flight control, thats the principal I work to, and keep the grounds tied together so using one Lipo and feeding everything from that will be fine.. Try a switching regulator on the VTX, some are noisy and create interference others dont? if you have problems switch to linear regulation.

Sounds like a plan, I have already identified good switchers that have very low levels of RF noise. But just to check, are you saying use a separate lipo for video stuff, or just use a regulator, large cap, and tie the gnds either side together?

Cheers

 

[BorisS]

I hear you on that, but its worth mentioning again that the VTX is not powered at present.

Also just tried without the camera plugged into MCU, same buckaroo effect.

So it looks like we can rule both Camera and VTX out as causes.

GPS is mounted on the longest pole, 15cm above IMU, and obvs correctly set as -15 in the assistant.

I will re-route all cables, but again this does not make sense as a cause, as this would be affecting the IMU etc at all times, not when just in RTH mode.

I understand that its hard to read through all the posts in this thread, but it's kind of the only way to get the full picture of what's happening. I do really appreciate the input though, better to have suggestions and rule them out than not have them at all!

"If" you hook your VTX and cam to the spare channel output from the MC, this would effect the BEC from the PUM. Since the MC will only give out voltage if the BEC is connected to think X1 or X2 don't remember to lazy to look. Thus the actual power supply of all the components that are powered over the can-bus
MC IMU GPS-modul should not be effected. Or lets hope so, they are not and DJI did separate the normal BEC in the PMU from the Can-bus voltage output. Nevertheless the point is a VTX can pull over 1 A and should be sperate and hooked to a separate BEC with a LC filer etc to get the power cleaner.

There are two modes to IOC corse lock and Home lock. If there wasn't a change in the FW and feature set that i am not aware of corse lock works no matter how close to or far away the copter is from you. You are setting a virtual forward point on the copter and it obeys it. Home lock unsing actually GPS data will work at the moment you push you copter 10m away from you.

At the moment you have IOC activated after startup of the copter you should get 3 been flashes indicating that it remembered a nose forward position.

Forget about POI for this time !

STEP1: Record forward direction: There are two ways: Manually and Automatically. The LED willblink quickly if recording is successful.

2. a) Automatically: The MC will record the current nose direction as forward direction, 30 secondsafter you power on the multi-rotor. **PLEASE BE AWARE OF THE FORWARD DIRECTIONWHEN YOU SWITCH ON THE MULTI ROTOR IF YOU USE COURSE LOCK**
3. b) Manually: You can switch the X2 channel between OFF and Course Lock position quickly 3
   to 5 times to record the current nose direction as the new forward flight direction at any timeafter you power on the multi-rotor for longer than 30 seconds.

STEP2: Switch on course lock: Switch the X2 channel from OFF to Course Lock position, when all
the following requirements are met the LED will blink slowly to indicate the IOC mode.

2. a) The forward direction is recorded successfully
3. b) The MC is in Atti. or GPS Atti. Mode

I am pretty sure there is a ghost in the shell or your can-bus voltage is not sufficient or it cant deliver enough power ! Also the reason why the other devices didn't work.

Boris

 

[modles]

Yep will try this. Just to be double clear, at no point was the video TX hooked up the BEC of the WKM.

Only the camera was connected to the MCU, and was pulling a verified 60ma max.