Replies to This Discussion

Permalink Reply by Gustav Kuhn on February 8, 2013 at 1:13am

Hugues,

I've been in RC since 1980, electrics since they first came out, and Arducopter since about three years back.

And I say calibrate your ESC's one by one using the correct battery, and straight out of the throttle port on your RX.

Once you KNOW the ESC's are working, you can start to fault find the rest of the system.

And sometimes you have to do the APM upload/config/radio calibration a few times before it works properly.

Yes, I do electronic repairs/fault finding for a local RC supplier/ Drone supplier :-)

Permalink Reply by Hugues on February 8, 2013 at 2:25am

I will try your suggestion to calibrate the ESCs one by one (i did not try that yet), howeverI am pretty sure they are correctly calibrated. I can for example see in the mission planner the display of the following parameters : CH3 IN and CH3 OUT.

CH3 IN varies correctly in function of the position of my throttle stick (low PWM value 989- High PWM value 2013)

but CH3 OUT is either PWM 989 when stick full down, or PWM 1700 whatever the stick position. This means that the APM does not translate/forward correctly the CH3 value from the receiver to the motors control wire. It is as if the APM would output on CH3 OUT a fixed value of 1700 PWM...

It is also as if after I unplug the battery at step 6 of the end of ESC calibration, the APM does not "memorize" the parameters. Because as long as I stay on step 6 without unplugging the battery, everything works perfect and in function of the throttle's stick position.

Hope I do not have a defect APM board...

DeveloperPermalink Reply by Randy on February 8, 2013 at 5:14am

Hugues,

     Of course you're testing in stabilize mode right?

     If you can see those values in the mission planner then it means that arducopter is trying to output those values.

     Not that is should matter but you have a very wide throttle range (989 ~ 2013).  In fact, ESCs only have 800 positions (i.e. something like 1100 ~ 1900) so there's no advantage to having a wider range than that.

     Maybe you can make video?

Permalink Reply by Hugues on February 8, 2013 at 6:21am

Yes indeed I am in stabilize mode (i see I have the checkbox "simple" ticked next to the flight mode in the mission planner, does that matter ?).

I will make a video of my problem. I indeed see those PWM values in the mission planner. I am using as Tx a classical Turnigy 9x with ERX firmware, so nothing special in principle.

 

Permalink Reply by Hugues on February 8, 2013 at 1:53pm

heelo Randy, can you take a look at my last post in this thread with the pictures I posted. I even tried tonight with 2.8.1, same behaviour. I do not understand why the APM outputs constant PWM values to the four motor control lines whatever the throttle input value. This drives me nuts !

Permalink Reply by Hugues on February 8, 2013 at 11:00am

So I have done the following :

-I have calibrated the ESCs manually , one by one , succesfully: with direct connection between ESC and receiver, I get a normal throttle control (input=output PWM).

- I have reconnected all 4 ESCs to the quadcopter, redone radio calibration. I can arm without problem and I am in stabilize mode.

But still same problem. I illustrate with pictures:

First picture: motors are armed, throttle stick full down : ch3 in->PWM =989, Ch3 out->PWM=989

 

On the second picture I push the throttle stick at about 50% : ch3in->PWM=1633, ch3out->PWM=1119

On the third picture : I push throttle stick at maximum : ch3in->PWM=2012; ch3out->PWM=1267

In addition I notice that leaving the throttle at same position, the ch3out varies between 1200 and 1400 in a kind of random way (ch3in stays constant).

 

Is this normal or do I have a problem with the arducopter code/APM ?

Permalink Reply by Jeremy Zawodny on February 8, 2013 at 3:30pm

Thanks for the screenshots--I didn't know you could see the output values in mission planner like that.  That will definitely help me figure out if the bug I have is the same as you.

DeveloperPermalink Reply by Randy on February 8, 2013 at 7:38pm

Hughes,

    What you're showing above looks pretty normal.  You know a quadcopter adjusts it's motor speed hundreds of times per second in order to keep balanced.  if you plot a graph of the motors they're constantly increasing decreasing in all kinds of patterns.  Checkout the images near the bottom of this wiki page to map the motor numbers to the physical locations.

    So for example when motors 2 and 4 are high it's the back motors that are providing more power probably because the copter is pitched back.

     Have you tried flying?  Could you provide a video of the copter itself when you try to fly it?

Permalink Reply by Hugues on February 9, 2013 at 12:44pm

No i never tried to fly yet because i am not sure it is working. I understand you say it looks pretty normal but why can't i get a low throttle on the motors with a low stick? PWM out jumps immediately to high values instead of " copying" the minimum transmitter value of 989.
Would i dare try to fly this thing with no progressive thrust control?

DeveloperPermalink Reply by Randy on February 9, 2013 at 10:00pm

Hugues,

     yes, I think you should buy some more props, motors and arms and be mentally prepared to break it and spend hours fixing it back up.

     ...then first try taking the props off and see how the motors react and make sure they're not going to full power as soon as you give it throttle.  You can also hold it in your hand (with props off of course) and try moving it around a bit and you'll be able to hear a bit the motor reacting. i.e. try leaning it on it's side...try rotating it horizontally, etc.

     ..then find a place where you're nice and safe (i usually hide behind a door the first time i try out a new copter) and where you're not going to damage anything too valuable and then give it a try.  Start with just short bursts of throttle until you're comfortable that it's under your control.

     ...at this stage keep it in stabilize mode at all times..don't try for alt-hold or anything yet.

Permalink Reply by Tim Green on February 10, 2013 at 8:20am

You can also tether the copter to the ground. I use a big plywood board with a hook for the string in the center. I affix the copter, then stretch out the string in all directions to make sure copter cannot hit anything while flying. Star with about two feet of nylon twine for untested copter. To see if copter will rise off ground without flipping. Then gradually add more string until you can hover 5-6 feet off the ground without pulling the string taught.

Permalink Reply by Jeremy Zawodny on February 8, 2013 at 11:24am

I don't know if this helps, but I'm having what could be a similar problem using APM 2.5 and 2.9.1 on my DJI450 quad.  I started describing it in the comments on this posting:

http://www.diydrones.com/profiles/blogs/using-a-dji-450-flamewheel-...

The two differences are (1) I'm using ESCs that cannot be calibrated (the DJI Opto 30 model), and (2) instead of getting a fixed throttle response, I get no throttle response (which is like saying mine is fixed at 0).

It'll be interesting to see if this is really just two versions of the same problem.  I'm currently at a loss for what else to try.

• ‹ Previous
• 1
• 2
• 3
• Next ›
• Page