Balancing during the assembly

Firmware/software/electronics/mechanics
Post Reply
DesTinY
Member
Posts: 97
Joined: Sat Feb 09, 2013 5:18 pm
Location: Bünde, Germany

Balancing during the assembly

Post by DesTinY » Mon Apr 22, 2013 11:38 am

Hi,
does anyone have experience with propeller balancing and/or weight balancing generally?

I can imagine that slight variations in the propeller production can cause to vibrations.
Maybe that could checked during the assembly before the Motor are glued on the CF. If I mount just on propeller at a time on one of the motor and hold this motor with the (unbalanced Propeller) with my fingers maybe it is possible to feel vibrations at running!? Maybe it is possible to write an self test Program (after assembly) to check vibration of each Motor/Propeller with the G-Sensor of the CF :lol: .
Furthermore I thing it is important where and how to put the LiPo on the CF. The power consumption of the Motors are probably very different at hovering if the CF is weight unbalanced.



tobias
Bitcraze
Posts: 1916
Joined: Mon Jan 28, 2013 7:17 pm
Location: Sweden

Re: Balancing during the assembly

Post by tobias » Tue Apr 23, 2013 9:39 pm

Hi DesTiny,

You are correct on every point you mentioned. We have written, and will write some more instructions, on how to maximize the performance: http://wiki.bitcraze.se/projects:crazyf ... erformance.
Getting balanced propellers is a key factor to get good performance but yet we have no good solution other then testing each propeller and sorting them. If you find a good way to balance them please share!
We have also discussed making a test that could measure the vibration of each motor/propeller. Shouldn't be to difficult and I hope you beat us to it :-)

SU88RZ
Beginner
Posts: 7
Joined: Thu Apr 25, 2013 7:59 am

Re: Balancing during the assembly

Post by SU88RZ » Thu Apr 25, 2013 11:45 am

I haven't done this with props before, but for very small spinning tops we would support them by their axis and they would droop to one side. A dot of fingernail polish, on the top, was enough to bring them back into balance. and if it was wrong it wasn't to hard to scrap it off.

using the motion sensor to see if a prop in out of whack is a great idea. You could test which motor was wrong and you would only need to remove and fix that prop then.

As i dont have mine yet someone else will have to test of this works.

DesTinY
Member
Posts: 97
Joined: Sat Feb 09, 2013 5:18 pm
Location: Bünde, Germany

Re: Balancing during the assembly

Post by DesTinY » Thu Apr 25, 2013 8:22 pm

I have learned a lot about balancing and found some good videos for larger propeller balancing.
http://youtu.be/OXuNnYQO2s4
http://youtu.be/N02NHYrpA0g
But nothing comparable to the CF propeller size.
At this point i have to wait and will try some things if i got my CF and do a video of balancing after success.
To find the perfect center of mass(CoM) of the CF i have a idea that i have to proof, too. (Cords between the Motor and the probeller (first: M1+M3; next: M2+M4) and then balance X and Y axis of the CF on a flat table) we'll see ;)

The self test motor vibration measuring program... sounds like my first CF project :lol: First step manually than automatically after an (recognized) crash :D

To be continued... :D

SuperDave42
Beginner
Posts: 23
Joined: Tue Feb 05, 2013 8:30 pm
Location: California, a
Contact:

Re: Balancing during the assembly

Post by SuperDave42 » Fri May 10, 2013 8:02 pm

I have a couple very small MPU6050 based IMU's from ebay,

I was thinking I could hot glue the spare crazyflie motor to it and use the Multi Wii software GUI to monitor the vibrations and help to balance the props, I'll plug the IMU into a Femtoduino Arduino microcontroler tied to the IMU via I2C

small portable prop balancer. probably hook up a BT dongle to it and use the GUI on the Android Just a thought
http://www.ebay.com/itm/US-10DOF-IMU-MW ... 460ec20ff0
https://www.tindie.com/products/femtodu ... -with-usb/

https://play.google.com/store/apps/deta ... aWNvbmYiXQ..

https://play.google.com/store/apps/deta ... dGl3aWkiXQ..



Dave
Superluminal carbon fiber flying robots.

marcus
Bitcraze
Posts: 599
Joined: Mon Jan 28, 2013 7:02 pm
Location: Sweden
Contact:

Re: Balancing during the assembly

Post by marcus » Sat May 11, 2013 10:12 am

Good idea, the MultiWii android clients look very nice.

We have been thinking about using the Crazyflie itself to do the balancing, but have never gotten around to doing it. The firmware could be modified to run one motor at a time while you are holding the Crazyflie down and then the sensor output could be analyzed.

/Marcus

DesTinY
Member
Posts: 97
Joined: Sat Feb 09, 2013 5:18 pm
Location: Bünde, Germany

Re: Balancing during the assembly

Post by DesTinY » Sat May 11, 2013 2:44 pm

Hi Crazy pilots,

i want to post some pictures of my balancing experience as a preview! (after 24h with a CF :D )
I will make a video of it in the next days.

Manually find the perfect COM:
Image
Image

Manually propeller balancing:
Image

I had tried to find good balanced propeller with the battery method but without success. It is hard to feel any difference between the propeller.
I find the right place in the CF FW to put in a another selt test function for the propeller balancing mechanism, but a working Plotter in the CF Client would be very helpful to compare the curves.

marcus
Bitcraze
Posts: 599
Joined: Mon Jan 28, 2013 7:02 pm
Location: Sweden
Contact:

Re: Balancing during the assembly

Post by marcus » Sat May 11, 2013 5:47 pm

Nice guide! Improved plot functionality is planned, we are working on it but it's not ready yet.

/Marcus

SuperDave42
Beginner
Posts: 23
Joined: Tue Feb 05, 2013 8:30 pm
Location: California, a
Contact:

Re: Balancing during the assembly

Post by SuperDave42 » Sat May 11, 2013 11:46 pm

Nice ! thanks for sharing that.

DesTinY wrote:Hi Crazy pilots,

i want to post some pictures of my balancing experience as a preview! (after 24h with a CF :D )
I will make a video of it in the next days.

Manually find the perfect COM:
Image
Image

Manually propeller balancing:
Image

I had tried to find good balanced propeller with the battery method but without success. It is hard to feel any difference between the propeller.
I find the right place in the CF FW to put in a another selt test function for the propeller balancing mechanism, but a working Plotter in the CF Client would be very helpful to compare the curves.
Superluminal carbon fiber flying robots.

DesTinY
Member
Posts: 97
Joined: Sat Feb 09, 2013 5:18 pm
Location: Bünde, Germany

Re: Balancing during the assembly

Post by DesTinY » Mon May 13, 2013 2:44 am

As I promised...

here is a nice little upgrade of the CF FW to improve the performance of the CF very much!!!
With this propeller balancing self test (at start up), it is possible to find the perfect ONE. :lol:

Bad propeller balancing:
Image

Good propeller balancing:
Image

Without propellers:
Image

By the way my Propeller 2 is very close to the optimum of ~0.6 G (compare to Without propellers) in this configuration.

Bitbucket Repo Link (balancing self test still in work!)
RFC (request for comments)

Post Reply