It's not the battery that causes the errors as such, it has more to do with the Lipo leads bending out of the battery. The bend in the Lipo wires is what causes magnetic errors.
Try it for yourself have the battery hang straight down you will see your compass deviation move :-)
Guys , I have done several measurements about magnetic field that can affect the mag on a multi rotor with a specific instrument. The result is that there are several objects which can perturb the compass, for example battery alarms buzzers or simply magnetized metal bolts but as rule of thumb since the effect is, if I'm not wrong, the cube of the distance the best idea is to take the mag away , even two centimeters can make a big difference. So IMO, forget shields, magic batteries etc, just keep the mag away.
@Shaun: Yes, changing batteries can cause the need to recalibrate the mag or place it further away or do extra copper/alu shielding. Too bad I don't have the link anymore but there was a video on vimeo showing different sources of mag influences. If that TBE is persistent with all lipos (not in your case) there is a chance that the mag declination is off. The bad thing about naza is that you can not type in the mag declination in the gui. It has some code that uses the world mag. model http://www.ngdc.noaa.gov/geomag/WMM/DoDWMM.shtml. And models are simplifications of reality that fit most the time but can be several degrees off in you flying place - so you will have to rotate the mag a few degrees (mechanical compensation of the declination error) to counter act the tbe. I think there is a picture in the manual for that procedure.
@Patrick: Thank you very much and best wishes for the new year! (BTW I was too stupid with my last post to connect your video with the topic so: Absolutely stunning video! And a slam for the guys always thinking of pilot error..so +1 for Olivers idea!) It is very interesting to know that you use MK Fc. Well I have just SEEN one in action since it's too pricey for my hobby needs... but I think since they are pioneers in the copterfield they have surely spent some respectable amount of time on ironing out bugs. And perhaps they reached a point of reliability over the versions where other copter projects still want to go. @Luc Maximilien: Your DIY Gimbal with zoom rocks! @Søren Kuula: Thanks for the codelink. Well since the variables are in ger it's quiet easy to read for me the html display drives havok on the formating - and I just took a quick glance at fc.c and didn't load that up into "CodeBlocks". Im OK with their coding style because I don't care since it is very nice to put their flightcode on display and you have at least the chance to track down an issue If you have one - Dji's code is always closed up and you will never know what makes it tick - or not. Putting a virtual rc transmitter with nick/roll into the hands of the gps is also ok - since it all boils down to generating tiltangles - no matter what the gps part puts out "stickmovements" or angles directly. Just my point of view.
and hear what he might have to say about code style (if it has bugs, are they easy to find?).
Anyway MK is was very GPS glitch proof the last time I looked, because the GPS only had the function of controlling a virtual pitch/roll stick with limited range.
Patrick, I have been the first MK partner shop in my country, and I have spent thousands of hours of work on MK platform. MK have always had a quite conservative approach on hardware and the last FC board is a confirmation. It is still my suggestion when a customer want something reliable with the recommendation to do not to install the last version of software as soon it comes out, almost two years ago , when Mk introduce some inertial navigation that required a more precise compass calibration a lot of Mk went down because of underestimation of compass calibration procedure then an update version was released.
I hope that Freefly will match your expectations. Good that you leaved S800, IMHO , the most inefficient multi rotor ever.
I have approach only recently "serious" video shooting setting up an Hexa with Mk board and a self made brushless gimbal with a Sony CX730, chosen because of zoom capabilities as you can see at the end of this test https://vimeo.com/74928653 . The goal is to develop a video shooting multi rotor based on APM with all the needed features and sell it.
For your needs I think that Gps features are not essentials while a very good altitude hold and probably an optical flow stabilization are much more important.
About the bullet proof, IMHO , if we want that multi rotors will be accepted as working devices by air regulators authorities we must absolutely increase the safety level, a second , independent, hopefully "bullet proof" flight control that comes when things going bad should be a priority.
Comments
just watched the show the waterfall shot makes up for the crash nice work ;-)
Try it for yourself have the battery hang straight down you will see your compass deviation move :-)
I am glad that the videos helped! They are old but somehow eye opening since we can not smell magnetic fields (yet...)
@Jb Thank you very much for your video!
Btw meanwhile I found the videos I meant...
http://vimeo.com/29384989
and (more interesting part2):
http://vimeo.com/29519728
@Shaun: Yes, changing batteries can cause the need to recalibrate the mag or place it further away or do extra copper/alu shielding. Too bad I don't have the link anymore but there was a video on vimeo showing different sources of mag influences. If that TBE is persistent with all lipos (not in your case) there is a chance that the mag declination is off. The bad thing about naza is that you can not type in the mag declination in the gui. It has some code that uses the world mag. model http://www.ngdc.noaa.gov/geomag/WMM/DoDWMM.shtml. And models are simplifications of reality that fit most the time but can be several degrees off in you flying place - so you will have to rotate the mag a few degrees (mechanical compensation of the declination error) to counter act the tbe. I think there is a picture in the manual for that procedure.
Cheers
Rob
@Patrick: Thank you very much and best wishes for the new year! (BTW I was too stupid with my last post to connect your video with the topic so: Absolutely stunning video! And a slam for the guys always thinking of pilot error..so +1 for Olivers idea!) It is very interesting to know that you use MK Fc. Well I have just SEEN one in action since it's too pricey for my hobby needs... but I think since they are pioneers in the copterfield they have surely spent some respectable amount of time on ironing out bugs. And perhaps they reached a point of reliability over the versions where other copter projects still want to go. @Luc Maximilien: Your DIY Gimbal with zoom rocks! @Søren Kuula: Thanks for the codelink. Well since the variables are in ger it's quiet easy to read for me the html display drives havok on the formating - and I just took a quick glance at fc.c and didn't load that up into "CodeBlocks". Im OK with their coding style because I don't care since it is very nice to put their flightcode on display and you have at least the chance to track down an issue If you have one - Dji's code is always closed up and you will never know what makes it tick - or not. Putting a virtual rc transmitter with nick/roll into the hands of the gps is also ok - since it all boils down to generating tiltangles - no matter what the gps part puts out "stickmovements" or angles directly. Just my point of view.
Cheers Rob
Hi,
Seriously? Try to have a person knowledgeable about programming to look at
http://www.mikrokopter.de/mikrosvn/FlightCtrl/tags/V2.02b
and hear what he might have to say about code style (if it has bugs, are they easy to find?).
Anyway MK is was very GPS glitch proof the last time I looked, because the GPS only had the function of controlling a virtual pitch/roll stick with limited range.
Regards
Soren
Patrick, I have been the first MK partner shop in my country, and I have spent thousands of hours of work on MK platform. MK have always had a quite conservative approach on hardware and the last FC board is a confirmation. It is still my suggestion when a customer want something reliable with the recommendation to do not to install the last version of software as soon it comes out, almost two years ago , when Mk introduce some inertial navigation that required a more precise compass calibration a lot of Mk went down because of underestimation of compass calibration procedure then an update version was released.
I hope that Freefly will match your expectations. Good that you leaved S800, IMHO , the most inefficient multi rotor ever.
I have approach only recently "serious" video shooting setting up an Hexa with Mk board and a self made brushless gimbal with a Sony CX730, chosen because of zoom capabilities as you can see at the end of this test https://vimeo.com/74928653 . The goal is to develop a video shooting multi rotor based on APM with all the needed features and sell it.
For your needs I think that Gps features are not essentials while a very good altitude hold and probably an optical flow stabilization are much more important.
About the bullet proof, IMHO , if we want that multi rotors will be accepted as working devices by air regulators authorities we must absolutely increase the safety level, a second , independent, hopefully "bullet proof" flight control that comes when things going bad should be a priority.
@ Gary Mortimer
I had a really bad experience with wooden Xoar props.
One of them was completely wrong , I do not mean out of balance, but really wrong in the CNC machining and gave incredible vibrations.
Perhaps I got the one on a thousand but since I double check every prop made without a mold.
-
1
-
2
-
3
of 3 Next