Balancing during the assembly

[tobias]

Great! I will try to find time to test it today. It looks pretty ready for integration

[zak]

Is it possible to make the graphic used in the balancing tab a bit smaller? I generally control my flie from my laptop which only has an 800 px high screen, making it impossible to get to the buttons at the bottom to calibrate and start the test.

Thanks for all your hard work on this feature!

[DesTinY]

Is it possible to make the graphic used in the balancing tab a bit smaller? I generally control my flie from my laptop which only has an 800 px high screen, making it impossible to get to the buttons at the bottom to calibrate and start the test.

Thanks for all your hard work on this feature!

Yes, it is possible to resize all balancing tab components in dependence on the window size, but have to figure out how!

[gmcintire]

I was just reading this thread and a thought occurred.

With one motor running, the frame should move back and forth with the unbalanced propeller. This is detected by the accelerometers.

What about an LED hooked to one of the header pins(you would plug it in for prop balancing) that would strobe the propeller on each revolution. It could flash when an accelerometer reaches its max(in a cycle).

Say you had it set to strobe on max X. This would show you the heavy side of the prop in the X direction. You could identify that heavy side of the prop while strobing. Do this one prop at a time. Then shut it down and and sand the heavy sides a little.

All without removing any props.

[tobias]

That is as a really great idea. I wonder how doable it is though. The propeller would have to turn pretty slow to be in the accelerometer frequency range and I'm not sure the motors can spin that slow. The generated vibration might not be big enough at that speed either. Maybe it is possible to use the aliasing effect though to be able to spin the props faster and still get an accelerometer reading.

[gmcintire]

Gee. I didn't think of that

But, I just looked up the accelerometer specs for the first time and I see this low pass filter for the accelerometers that can be programmed up to 260Hz. Well, that is 15000 RPM. I don't even have my crazyflie yet so don't know that much about it, but I'm thinking this would be a reasonable value for a tiny motor to turn slowly.

I think you know this system far better than I, Tobias. Can the motors run that slow???

[tobias]

The Crazyflie hovers at about 17000 rpm so it might not be a problem. Running them at 5000-10000 rpm would probably generate enough vibrations.

DensTinY did some great work with a balancing detector you can have a look at.

[gmcintire]

I agree. He did some 'very' excellent work!

I was actually hoping he would incorporate this LED strobe concept in his work. He certainly has the experience and skills to make it work if its possible. DensTiny??? What do you think?

[DesTinY]

Hi,
first of all thanks for the kind words. Sorry for the delay, I am currently very busy with my bachelor thesis and can just follow the posts in the moment.
But I want to complete my work for an request for integration in to the Main PC client. (BTW: Help are welcome )

The LED strobe concept is a really good idea!
You are right to detect the heavy side of the Propeller should not be the Problem, but to show/tell it to the user.

May be if we mark one side of the Propeller with a white dot and use a external LED (on GPIO) like this found on YouTube (Link).
Than it should possible to let the white dot point in a defined direction (depends on Motor and LED frequency), where we know e.g. this is the heavy side of the Propeller.

Best regards,
DesTinY

[tobias]

I think you guys are on to something. Why not make a generic balancing device which could measure any propeller