CS8 so what am i missing?

[remyd]

http://www.youtube.com/watch?v=GCguvLRvraQ&feature=youtu.be

So to me it looks like a CoG issue but i am not having much luck fixing it. I welcome any suggestions!
A little history - i flew the platform last week with no problems just the gimbal and a single battery, i then added 1.7kg to the gimbal to simulate a DSLR re did the CoG and adjusted the values for the IMU and GPS but couldn't get off the ground without tipping much like in the video. In the process of getting a bit excited i busted a prop from a slight tip over but no other damage. Upon checking the CoG again i noticed that with the CS 2-d gimbal and weight along with the slight tips had caused the gimbal to drift to one side which caused slight changes to CoG. Since then i have spent considerable time making sure the CoG is excellent and in the centre of the frame. Adjusted the wookong and recalibrated to no avail - it still wants to tip!

I have even reverted to my original setup and settings when it flew great but now it wants to do the same tip. Even if i offset the COG to the alternate side to where it wants to tip it still seems to want to go in the tip direction shown on the video. Its almost like the motors/esc on that side are not providing enough thrust but i cannot really see any difference between the rotor rpm visually.

The IMU is located dead center of the frame, with forward pointing towards the GPS. The GPS is 21cm in the x direction fwd of the centre of the IMU and 12cm above the centre of the IMU.
My orignal settings with the gimbal and a single battery were (IMU x=0, y=0, z=0; GPS z=21, y=0, z=-12) when i added the 1.7 kg to the gimbal and balanced it so that COGis directly under the IMU/ centre plate i used the values (IMU x=0, y=0, z=-6; GPS z=21, y=0, z=-18).

My setup is Maytech 55a ESC's, Plettenberg 15-22 orbits, carbon 13x6.5 props, CS8 frame, CS 2 axis gimbal powered by Thunderpower 5s 5A.

Thanks in advance
Remy

 

[BorisS]

You didn't change anything of the motor wiring or props etc. in the meantime ?

Gains also stayed the same from the flights that were successful ?

Do you have anyone that can hold the bird over head and can't test by rolling the bird into the direction that it wants to tilt if it regulates/ tires to level, equality as if he or she rolls it into the other direction.

Do this in Attitude mode and just don't bring the throttle over mid position that it dosent want to take off. It will try to level itself before take off throttle strength.

Boris

 

[remyd]

Hi Boris,

Nothing has changed gains, wiring, motors etc. All is the same! I just replaced the prop and took the weight off to get back to my original settings. I don't think my helper will be that impressed with holding it above their head lol. Does anything look odd with the settings? I was thinking of recalibrating all the ESC's ?
The craft is balanced almost pefectly and i dont think i could get it any better. I really am stumped!
Cheers,
Remy

 

[BorisS]

Since you changed the props did you double check them, happened to me several times that i though I put them on correctly and than one was wrong.

My orignal settings with the gimbal and a single battery were (IMU x=0, y=0, z=0; GPS z=21, y=0, z=-12) when i added the 1.7 kg to the gimbal and balanced it so that COGis directly under the IMU/ centre plate i used the values (IMU x=0, y=0, z=-6; GPS z=21, y=0, z=-18).

I am confused about your settings. In the video you are flying without the gimbal and only a lipo on top which makes the bird top heavy. You z values should be zero to a plus values. But if the Z value being of by 3 or 4 cm would cause that behavior at take off i doubt.

Of coarse the ESCs could also be the issue, recalibrationg just to go sure would be a good thing.

Boris

 

[remyd]

Yes sorry in the video i used a positive z value of 2 because i took the gimbal off, but have since tested in my original configuration with the values qouted. The only thing that was changed was the GPS antenna as the older version was supposedly faulty. This happened prior to trying the heavy load. However, i am trying to take of Alt mode so GPS shouldn't be an issue. That said the heavy load was when i first noticed the tilt and it was after i replaced the GPS unit. What happens if you disconnect the GPS all together?

 

[BorisS]

I have flown my CX4 without the GPS module in attitude mode without any problems. This was a couple months ago if something changed FW in the meantime i can't tell. But i don't think so.

I remember my hexa acting the same way at take off because i had the gains on the Radio and sometimes had them to low.

It could be that the GPS module messes up the can-bus communication. Thus the IMU is delivering faulty info of the state of the bird. I had something similar when i extended one of the can-bus cables for the LED and badly isolated the single cables which resulted in a constant drift of the CX4. Using a different LED or NOne resolved it.

At are you gains at the man gains ?

Boris

 

[remyd]

Gains are 100 for everything and same as original when it worked

 

[BorisS]

hmmm I don't want to motivate you for something that you might regret, ( saying that after recommending to hold the bird over your head ) ! But being that top heavy and having the gains at 100 % for basic gains can also be the reason for the bird regulating/leveling to late at take off and rolling to one side. With 5s and the prop size I am positive that the basic gain for roll and pitch should be somewhere between 150 to 200.

Boris

 

[remyd]

Thanks Boris - i have the gimbal back on and 0.5kg weight on the gimbal so it is no longer top heavy. I have locked the gimbal and weight so it cannot move and balanced. I will start again tomorrow, double check motors, props, disconnect GPS and up the gains ( i may have had them at 120 early in the piece). I also have a naza and i am almost tempted to plug that in in place of the wookong to see if i get a similar physical response. If not it isn't physical rather wookong. I should also add that i added the 900Mhz data link (although not using it for these tests) into the mix prior to getting the tilt. Anyway getting late here in Oz so need to grab some shut eye before the kids wake me. If you think of any advice i would appreciate it for my testing tomorrow. As usual thanks for your contibutions to mine and many others problems through this forum. it is greatly appreciated.
Cheers,
Remy

 

[Siteline]

I have done this once before as well with the same results...a tip over. After loading the firmware and going back to default settings is it possible you didn't put the correct IMU direction back in? i.e backward/forward?

 

[remyd]

Ok - so does one admit that they are a fool?

Even though i made no changes to the motors or ESC and even though i thought i checked yesterday that the motors were spinning the right way after a night sleep and the are kids sent to school so i could have another look in piece and quiet, the first thing i noticed was that the motor that had the blade strike is reversed compared to what it should be. I know i didn't change the wiring because i heat shrinked the connections together and i haven't touched the ESC's programming since my first setup.

So has anyone heard of an ESC reversing after a blade strike? Anyone suggest why it may have done this? Anyway now to see if this fixes the problem.

As usual the problem came down to user error (he says before testing if this has fixed the problem) and this user was sure he had everything exactly right. But NO!
Thanks again to those on the forum pointing out the obvious. Sometimes its best just to start from scratch and make sure everything is installed the way it should be.
Remy

 

[BorisS]

cool that you found it !

Never heard of a ESC reversing after a blade strike strange one.

Boris

 

[remyd]

thanks Boris - got me stumped but anyway it seems to be working i have reloaded the gimbal with 1.7kg, reset parameters and am off to the park to do some testing. i will post video later for those that might be interested in medium-heavy weight copters.

 

[Twister773]

Do you have a pic of how you have your wkm modules mounted?

 

[remyd]

IMU is mounted dead centre where the COG for x and y would be if balanced correctly.

View attachment 4942View attachment 4943

You can sort of see the IMU under the battery tray - i used some perspex to make a small tray for it to sit on. I also used two power boards as i was worried about current draw through a single board given some of the information i had read on the forum.