This is the new folding frame Project I am working on.
Can anyone tell me if it´s possible to control the eight counter rotating motors with ACM.
You need to be a member of diydrones to add comments!
This is the new folding frame Project I am working on.
Can anyone tell me if it´s possible to control the eight counter rotating motors with ACM.
You need to be a member of diydrones to add comments!
Comments
@Denni
Link for that Flyboard from Obor?
Denny, if possible could you share those calculations please.
Jani
The MK I2c is definitely not safe, BL- FETS can just burst into flames without any logical reason, even when the motors are not running. As for the Power distribution board, it is a joke. Also with okto the camera outputs don't work properly, everything just gets overloaded. I found that the Flyboard from Obor was a step in the right direction as far as getting the reliable Plush ESC's to work. Avoid I2c at all costs. IMHO Holger and Ingo make nice electronic ideas but mechanical engineers they are not.
6 fast PWM outputs is ok from APM, 8 is not. So you can fly Hexa. I fly my hexa protos every week.
This is current proto frame. Landing gears will be changed on it and then it starts to be ready to be made as a kit. But it will take a while untill kit is widely available.
There are also higher landing gears coming in few weeks for all frames.
Jani,
Can you elaborate on "Sure you can use all 6 PWM outputs but it means lower resolution"? Does this mean that the ardupirate hexa runs at least 2 channels on a lower resolution PWM?
I am about finished designing CF hexa frame in solidworks... maybe I should switch to quad frame for now?
I am actually very interested in having 12 PWM high resolution outputs to control 6 ESC's and 6 individual servos for variable pitch props I'm building ... does the current APM hardware have headroom/hardware to do this with some "tweaks", or should I start designing my own board?
No it's not about that, we have been using modified Turnigy plush ESC's too but still getting random freezeups when there are more devices connected to Atmel I2C bus. If you go to google and just look "atmel i2c freeze" you will get more than 200k posts about this issue.
This is one reason why we have been re-writing most of our library codes on Arducopter project.
Jani
I got around the I2c problem with the converter from Flash RC called Flyboard. You need Turnigy Plush controllers 400hz or you have big latency problems and then it's unstable.
@I.S. There are still additional PWM outputs in 1280/2560 that can be converted to be as FastPWM pins but that needs extra soldering. We have discussed this issue already in our developmeng group and most probably I will build one test units for 8 engines on somepoint.
It's internal Timers on main CPU that prohibits OUT5-6 to be used as FastPWM outputs. A bit like on old ArduPIlot you could only conrtrol 3 servos even thou there are 6 PWM outputs on Atmega 328. But there are not enough Timer hardware inside CPU to maintain fast and accurate PWM pulses to all PWM outputs.
On Atmega there are 2 x 8 bit Timer and 1 x 16 bit Timer. One of the Timer has to be left untouched due it's internally used by CPU and you don't want to mess with that so 3 x 8 bit Timers are left for PWM. You can "split" 16 bit Timer to be as 2 x 8 bit Timer.
Sure you can use all 8 PWM outputs but it means lower resolution which is not good for us.
On 1280 we have more Timers that van be used but we also have more hardware design issues with that too that prevents OUT5-6 to be used as FastPWM.
FastPWM is special way to handle updates to Timers without using any CPU clock pulses. To put it simple way, we setup some of the Timers to be in special mode and write values directly to their internal registers. Timer hardware uses this registers to create PWM outputs without help of main CPU. So PWM change is immediate.
-
2
-
3
-
4
-
5
-
6
of 7 Next