RC India

General Topics => Beginners Zone => Topic started by: anis jayaram on December 18, 2014, 04:56:34 PM



Title: quadcopter takeoff problem
Post by: anis jayaram on December 18, 2014, 04:56:34 PM
i assembled my quadcopter with the following configuration
rame- ST360
motor- 4*Avionic C2830 KV850
ESC- 30A Emax with simonK 
props -  10x4.5R
FPV   -  5.8G FPV Video TX RX 200mW+11db antenna +720 PAL Camera (SONY)
Flight control- APM2.6 + Ublox NEO 6M GPS Compass w/ Foldable Stand  + Power Module APM
Telemetry -  3DRobotics 3DR Radio Telemetry Kit 433Mhz Module
Battery - wolfpack 4200mah 20C
TX/RX-  Avionic RCB6i 2.4Ghz 6ch transmitter with receiver

but it is refusing to takeoff i didnt film my quad (will do soon)  but got a few online ones having the same problem as mine ,
https://www.youtube.com/watch?v=oAyggHIr_Qs (https://www.youtube.com/watch?v=oAyggHIr_Qs)
https://www.youtube.com/watch?v=_nj9vihERHU (https://www.youtube.com/watch?v=_nj9vihERHU)
can you suggest a fix ?  tried to take off in loiter,stabilized as well as alt hold mode , but the quad keeps on wobbling and spinning and it leans or tipps to one end shattered the props on that side
please help


Title: Re: quadcopter takeoff problem
Post by: subhashjk0508 on December 18, 2014, 05:00:42 PM
I had the same problems!! it was because of wrong motor connection to the apm!! i had connected the wrong motoes to the output pins!!

regards,
JK


Title: Re: quadcopter takeoff problem
Post by: Swapnil on December 18, 2014, 05:31:44 PM
I have the same basic setup without the external gps+compass module. My quad takes off, but at about 20% throttle it starts rising up by itself. I've attached a picture of the radio calibration. The throttle minimum and centre values are same. Is that the problem? I recalibrated but got the same results.

Anyone have the same issue? 


Title: Re: quadcopter takeoff problem
Post by: SideWinder on December 18, 2014, 06:05:42 PM
@Anis, Subhash has rightly pointed out one of the issues. Also check prop orientation and motor rotation direction. Meanwhile I hope you have setup rest of the stuff properly. Video would have been helpful.

@Swapnil

First of all
Which mode were you flying?
Which firmware?
Was it windy?
Did you AUTOTUNE?
Did you set mid point for your throttle in MP? Please give me the "THR_MID" & "TRIM_THROTTLE" value under Config/tuning---Full Parameter list in Mission Planner or APM planner.

Moreover if your quad is hovering up at 10%, that can also mean that its highly overpowered. Arducopter works best within the range of 30-70% throttle for hover. I am not sure what happens otherwise.

Nothing wrong with your radio calibration. It looks fine.


Title: Re: quadcopter takeoff problem
Post by: Swapnil on December 18, 2014, 06:12:29 PM
Stabilize mode, copter 3.2, not windy.

I don't have gps+compass module to auto-tune.

THR_MID = 500
TRIM_THROTTLE = 409

Is the radio calibration really okay? Are the min. and centre vallues supposed to be equal?


Title: Re: quadcopter takeoff problem
Post by: sooraj.palakkad on December 18, 2014, 06:21:03 PM
the quad keeps on wobbling and spinning and it leans or tipps to one end shattered the props on that side
please help

Your problem seems not with the APM - it is either your motor connection or wrong props.
check the motors are rotating in needed direction and all props are pushing air down to the ground.

If motors are rotating wrong way, just interchange any two connectors of the three that comes from the ESC of that motor.


Title: Re: quadcopter takeoff problem
Post by: sooraj.palakkad on December 18, 2014, 06:29:23 PM
Is the radio calibration really okay? Are the min. and centre vallues supposed to be equal?

I don't think the min throttle and center throttle value being equal is okay, IMO this may be the reason for your quad to take of at low throttle values. The flight controller assumes your low throttle value is center throttle, and thus generates high throttle to motors when you just start increasing your throttle. Thus your quad gets lift at 20% throttle .


Title: Re: quadcopter takeoff problem
Post by: SideWinder on December 18, 2014, 06:31:13 PM
From ardupilot.com "Mission Planner will show a summary of the calibration data. Normal values are around 1100 for minimums and 1900 for maximums." So it doesnt matter, if the values are different, thats is why we calibrate radio in the first place  I believe. BTW look at the image for demonstration over at ardupilot.com they are different

http://copter.ardupilot.com/wiki/initial-setup/configuring-hardware/#Calibrate_radio_control (http://copter.ardupilot.com/wiki/initial-setup/configuring-hardware/#Calibrate_radio_control)

Set the THR_MID between 430 to 460.  Adjust it till the time you find that when in hover the throttle is less sensitive in the center.

Basically Trim throttle is autopilots estimation of throttle required for hover. Randy, lead developer of arducopter had mentioned that autopilot calculates that data before landing hence he says its little less than the actual throttle required for hover.


Title: Re: quadcopter takeoff problem
Post by: Swapnil on December 18, 2014, 06:39:03 PM

I don't think the min throttle and center throttle value being equal is okay, IMO this may be the reason for your quad to take of at low throttle values. The flight controller assumes your low throttle value is center throttle, and thus generates high throttle to motors when you just start increasing your throttle. Thus your quad gets lift at 20% throttle .

I realize that. But, even after recalibrating many times, the same thing happens. I just can't figure out whether the problem is with my radio or with one (or more) of the parameters.


Title: Re: quadcopter takeoff problem
Post by: Swapnil on December 18, 2014, 06:47:02 PM
... So it doesnt matter, if the values are different, thats is why we calibrate radio in the first place  I believe....

I am not using mission planner, but APM Planner 2. It shows centre values as well as min. and max. The problem is that the min. and centre values are same, not different. Ideally it should be like this:

min: around 1100
centre : around 1500
max: around 1900

However, my centre value is equal to the min value (pic in #2). It can't be a radio problem as I flew my delta this morning and everything was normal.

I've disabled pre-arm check as I don't have a compass.


Title: Re: quadcopter takeoff problem
Post by: SideWinder on December 18, 2014, 06:55:57 PM
See its alright for throttle to not have a center value because thrust of different setups would be different, hence assigning  common value would not be useful. If you set the mid throttle properly, the autopilot would take care of rest of the stuff. You also need to auto tune twice so that you can get some good values, atleast thats my experience.

Also in alt hold mode, you will be using throttle assist, which will add some more expo to the center of the throttle. Throttle assist is also used in loiter/poshold/drift mode I believe.


Title: Re: quadcopter takeoff problem
Post by: Swapnil on December 18, 2014, 06:58:46 PM
How do I use the onboard compass (you said there is one one the board, only unused)? Maybe the problem is related to that.


Title: Re: quadcopter takeoff problem
Post by: sooraj.palakkad on December 18, 2014, 07:07:18 PM
@Swapnil, the problem is clearly this center throttle value - But I have no experience with APM can't help you out :(
There should be a way to manually set the middle throttle value - let me check .:)


Title: Re: quadcopter takeoff problem
Post by: sooraj.palakkad on December 18, 2014, 07:09:36 PM
How do I use the onboard compass (you said there is one one the board, only unused)? Maybe the problem is related to that.

This problem is not related to compass. However on board compass is available only on some models of APM. Which is your board version ?


Title: Re: quadcopter takeoff problem
Post by: SideWinder on December 18, 2014, 07:16:05 PM
Swapnil I would first like you to try to set mid throttle correctly. I am pretty sure your problem would be solved. Compass wont have anything to do with it because it wont be used in stabilize mode , or alt hold mode for that matter. When compass is enabled (you can enable and disable from MP) its used to ascertain heading, which is matched with acc data because both should be similar ideally. I think there is a threshold, after which your copter would go into failsafe if there is difference in magnetic data, and acc data. But yet it wont cause your copter to climb.

Put a close up high def pic of your board to first check whether it has an onboard compass. Also I did PM you the details on how to enable onboard compass by soldering a pad on the fc. You should check that too.

And for your confirmation check the dialog from MP after radio calibration sequence(pic attached along with the values i have been using). I believe that should lay rest to your doubts about throttle center issue.


Title: Re: quadcopter takeoff problem
Post by: phanivyas on December 18, 2014, 07:22:55 PM
check the props dude...i have faced similar problem initially, it was due to prop mismatch..

"props -  10x4.5R" ??? are you using 2 ccw & 2 cw props? correct?


Title: Re: quadcopter takeoff problem
Post by: Swapnil on December 18, 2014, 07:27:16 PM
Vishakh,
I know the problem isn't with not using the compass. But, for the board to arm, I have to disable the pre-arm checks. (Disabling the compass individually doesn't work in 3.2. It supposedly works in 2.9 but I couldn't find a link to that firmware)
I wanted to use the compass because then I could disable the gps check only and see if there's other problems while arming (say, with the barometer).

I'll set the mid throttle like you said and report back.

The APM Planner 2 crashes now when I try to re-calibrate the radio. Will try with Mission planner now.


Title: Re: quadcopter takeoff problem
Post by: Swapnil on December 18, 2014, 07:31:48 PM
...

And for your confirmation check the dialog from MP after radio calibration sequence(pic attached along with the values i have been using). I believe that should lay rest to your doubts about throttle center issue.

Mission planner doesn't show centre values after calibration. APM Planner (which I am using) does. I'm getting proper min. and max. values. with both APM Planner and MP. The problem is with the throttle centre value.

I know the APM has different centre values in different scenarios. But, this is the radio (rx) input value we are talking about. The calibration output should correspond to the radio's values.


Title: Re: quadcopter takeoff problem
Post by: sooraj.palakkad on December 18, 2014, 07:42:07 PM
Swapnil , check this link for manual mid throttle setting :-
http://copter.ardupilot.com/wiki/ac_throttlemid/ (http://copter.ardupilot.com/wiki/ac_throttlemid/)


Title: Re: quadcopter takeoff problem
Post by: sooraj.palakkad on December 18, 2014, 07:43:50 PM
Have you checked with any other channels in throttles place  -  I mean calibrate with any other transmitter channel in throttle's place ???


Title: Re: quadcopter takeoff problem
Post by: SideWinder on December 18, 2014, 07:50:28 PM
So try disabling compass in advanced setting and then try to just skip the gps check in pre arm checks. If it works then that should be great. Also start logging all data and try log analysis. Trust me, its really helpful. Try to graph barometer altitude and compare that throttle input. Also I hope you have covered the barometer with some foam or soft sponge preferably to avoid light and prop wash.

And I dont use APM Planner much. Mission Planner is largely used by community than AP I guess. Altho both work similarly, but there are some more features in MP since 3DR is helping that project too.

And no its not necessary for radio and calibrated values to correspond. Thats how it is and I believe no 2 radios would have the same end points.   Because even tho when trims on my radio are centered, after calibration it wont be on MP,  so I would then match it to what MP shows as 1500 and change my trims on radio. I would usually center the roll/rudder/elevator. And it has worked till now for me.

And center values aint shown on MP too, but if your radio is switched on and quad connected to MP, then if you go to the radio calibration page you can see the stick values which MP/autopilot sees.

Anyways you got to follow the development thread for AP, there you can understand how does the code think, because its all programmed. Also you can ask the questions there for a clear answer. What I am telling you is what I read/know by reading through many posts and threads.



Title: Re: quadcopter takeoff problem
Post by: Swapnil on December 18, 2014, 08:58:03 PM
Okay, I did the complete procedure (flashing, accelerometer calibration and radio calibration) with Mission Planner this time. The THR_MID value was set to 430.
I was able to hover it around my living room at about 10 cm height with 35-40 % throttle. However, it was moving all around the place. Maybe that was because I was flying it inside. Will try outside tomorrow morning.

Thanks so much Vishakh (SideWinder)! +1 to you!

I was being stubborn about the throttle centre value because I saw my quad fall instantly after moving the throttle stick down just one notch. It was also because of that wretched APM Planner 2. It kept crashing and hanging every now and then. I completely uninstalled it.

Should I decrease the THR_MID value more to hover at 50%? Is 430 the least value for proper hover?
Would adding payload be better than setting a lower THR_MID value?


Title: Re: quadcopter takeoff problem
Post by: SideWinder on December 18, 2014, 09:28:27 PM
At 10cm height! You cant expect much stability. Try it tomorrow, check CG and do a auto tune . Anyhow without gps/mag APM is somewhat similar to multiwii or kk2. Alt hold should work well tho. Do remember that when you move in forward or backward direction at some speed, your altitude may drop due to pressure bubble created during flight. So compensate for altitude in alt hold/loiter or position hold etc. Try stabilize and alt hold tomorrow. Ideally if winds are stable, your quad should be stable and be able to hold altitude in stabilize mode to a very little extent i.e it shouldnt move around too much @ mid throttle.

No 430 isnt the least value, that was just my guess since autopilots estimate was 409.So you can reduce the value if you want(try 420 then 415-410). Also checking in log, throttle input vs throttle output would help you more to ascertain exact mid throttle. Sooraj had posted the link earlier.

Add payload only if you want to and is important such as fpv/gimbal/telemetry/led etc. Whats the point of adding dead load as it would also decrease your flight time too. Thats why we have that parameter.


Title: Re: quadcopter takeoff problem
Post by: phanivyas on December 18, 2014, 09:44:24 PM
Determining THR_MID with the help of log is the best option. Also my advice is to tune your quad manually. Do not perform auto tune with AC 3.2 firmware. I have done that last week and ended up with an unstable quad.


Title: Re: quadcopter takeoff problem
Post by: SideWinder on December 18, 2014, 09:50:16 PM
Well I used autotune and it has been alright, not too sensitive. I see that some users have reported the sensitive quad in 3.2 after autotune, but havent faced the issue my self. Moreover remedy for the latest autotune issue is scaling down the pids by 25% till the time we have definite answers.

Anyhow you have the option to check new pid's after autotune in flight and can choose to save it  or revert to the old values before landing.


Title: Re: quadcopter takeoff problem
Post by: phanivyas on December 18, 2014, 10:00:51 PM
After auto tune my quad's PIDs were lowered to such an extent that it couldn't able to resist slightest wind. For example Rate_P for roll & pitch was adjusted to 0.04 which is too low. The default params are OK for flying, little tuning of Stab P, Rate P & I is required which can be done manually. And yes there is the option to revert the auto tune values.


Title: Re: quadcopter takeoff problem
Post by: anis jayaram on December 19, 2014, 12:06:32 PM
RX/TX calibration


Title: Re: quadcopter takeoff problem
Post by: anis jayaram on December 19, 2014, 12:09:19 PM
MP


Title: Re: quadcopter takeoff problem
Post by: sooraj.palakkad on December 19, 2014, 12:17:03 PM
@anis, did you checked the motor connections and props ?


Title: Re: quadcopter takeoff problem
Post by: phanivyas on December 19, 2014, 12:36:47 PM
As mentioned above check your motor direction and props.