Hi All,

I have a situation where after upgrading to some high-end ESCs, I have one at random go into calibration mode. This happens on about 1 in 4 flight attempts. The only way to restore proper operation is to repeat ESC calibration with Mission Planner.

I have had a dialog with the manufacturer about the issue. His position is that the problem is cause by the Pixhawk EEPROM not retaining calibration data, and that it is a known issue. He has seen it with a number of customers over the years. The solution was apparently replacement of the FC.

I'm not so sure. I've only had problems with the new ESCs. The Pixhawk worked fine for a couple of years with my previous ESCs. 

On the other side, my Pixhawks are out of warranty and 3DR isn't supporting them anyway.

Has anyone else seen this kind of issue?

Thanks in advance for any info.

Views: 960

Reply to This

Replies to This Discussion

Maybe this is a case of incorrect terminology as I'm not familiar with the Pixhawk's ability to store ESC 'calibration' data.  All-at-once ESC calibration simply passes commands through the FC to the ESCs (no ESC data is stored on the FC itself), while one at a time ESC calibration bypasses the FC entirely.

http://ardupilot.org/copter/docs/esc-calibration.html

Thanks for the reply. 

That was my original understanding, but I believed the person giving me the information to be more expert than I.

My own experience with these ESCs also says that it is the ESC itself that is the problem.

I've started an experiment that I hope will correct the issue or narrow the scope a bit. I increased the throttle RC1_MIN and RC1_TRIM parameters by 10uS each from their RC calibration values. The hypothesis is that because the throttle min and throttle midpoint have the same value, and that value is the very lower limit of stick travel, there is the possibility that the throttle is one or more uS over zero when I press the safety switch. These new ESCs may be sensitive to that. I would think that there should be a little margin built in, but I don't really know.

Time will tell. I only have two flights so far with this change. I want at least 20 before I call it, either way.

In the interim, if there are any other thoughts, I would be most grateful.

pixhawk is not "storing" any calibration data, it's simply forwarding PWM in to all ESC outputs while "in calibrating mode" so ESC's are actually the once messing up the job..

I've seen other bad behavior on "high end" (just overpriced) stuff.  you are most likely better off running some known good open source on ESC like heliBL or others..

Roger that.

Thanks for your thoughts and advice.

Does Arducopter also NOT store all-at-once ESC calibration data from one battery-power-up to the next?  

I very recently did a radio calibration, compass calibration, and ESC calibration in Mission Planner for my Arducopter-equipped hex.  After completing the ESC calibration, I tested the ESC's (sans props) to see if they all would start up simultaneously (which they did) and that they would all respond to throttle changes evenly when no other joystick movements were involved (which they also did).  I then put the props back on and went outside for testing.  On the first start-up one of the motors did not start at the same time as the other five, and it did not spin up to the same RPM the others were at.  Stopping the motors, disconnecting the battery, then re-connecting it and binding again did not solve the issue - and sometimes it was a different motor that wouldn't spin up - but sometimes ALL the motors spun up perfectly!

Any ideas? 

No, ESC throttle range calibration is something an ESC is supposed to remember. Arducopter stores RC calibration to know the range.
Some ESC s, (usually beforenentioned open source) *can* also be programmed with preset pwm range, not needing, nor being able to be calibrated.

Thanks for the ESC info.  Still looking for some help vis-a-vis the motor-not-starting problems.  I've had only one other 'copter prior to this 550 hex and I don't recall having that same problem with it.  But then that 'copter had a completely different type of built-in F.C. with on-board FPV and no GPS, so I think I would be comparing apples to oranges if I compared the wo 'copters. 

Andre K. said:

No, ESC throttle range calibration is something an ESC is supposed to remember. Arducopter stores RC calibration to know the range.
Some ESC s, (usually beforenentioned open source) *can* also be programmed with preset pwm range, not needing, nor being able to be calibrated.

Thanks for the additional replies.

My experiment failed and I've started another. These ESCs can store a fixed PWM range, so I set that to match my RC calibration values, with a little margin.The aircraft flies OK this way, though take-off isn't quite as smooth. I'll update with results when this fails or I have some confidence in it.

we have the same behavior : with APM series, ReadytoFly ESC 30A work very well. With Pixhawk, it is a nightmare. Some time working, some time not. We change them for Afro series : no more problem. We spend almost 1 week to understand  - no way to resolve it.

What esc is this,give us name and link?BLHeli(or SimonK) software on them or not?

So...if it is this one 

ReadytoFly ESC 30A 

http://www.readytoflyquads.com/f-30a-fire-red-normal-esc-with-rapid...

you have SimonK on it and you need adapter to program them....u can do that only if servo connector is available(be aware it is not servo conectors wich go to RX,but separate one for programing Eprom).. 

When u calibrate them,box with "program with TX" must be on(if its not,they wont calibrate)....and than u do normal all at once calibration...after u finish calibration,this box must be of or otherwise every time u have you throttle high they will enter calibration mode...

In Mission Planer be sure its not 1-"one shot" or 2-"oneshot 125" but 0-"normal"

Also in mission planer set throtlle range from 1000 to 2000........

thank you for you advice Emin. We get this one : http://www.readytoflyquads.com/rtf-escez-30amp 

Cheap but hours of flight without any problem. Except when we used it with pixhawk....

Reply to Discussion

RSS

Groups

Season Two of the Trust Time Trial (T3) Contest 
A list of all T3 contests is here. The current round, the Vertical Horizontal one, is here

© 2018   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service