Frank'n Quad Build

[wooden]

I can't believe you ran ESC calibration with props on... Glad you're ok, but there's a reason every manual (and Configurator!) tells you to remove the props when doing ESC calibration!

[MikeD]

Glad the damage wasn't too bad.

.... it took my figure as a tax...

Hope you get your figure back soon.

[jbsmith78]

I can't believe you ran ESC calibration with props on... Glad you're ok, but there's a reason every manual (and Configurator!) tells you to remove the props when doing ESC calibration!

LOL... Yeah, the several hundred times I've done that by the books just don't count... It's the one time you are in a hurry!! LOL Oh well, no worries... Got a new DVR heading my way and Friday is supposed to be cold by no wind so expect some new FPV vids!! Waaahhhooooooo

Found a great place to fly right by the warehouse so I'm pumped!!!

[jbsmith78]

Glad the damage wasn't too bad.


Hope you get your figure back soon.

Ah, the finger tax!! LOL I CRACK ME UP!! LOL

[jbsmith78]

So new little trick - I never liked the ACCEL readings and how much noise you get in the system when the quad is just sitting still on the Z axis - my range was bounding from -8.8 to -11.0 so I:

#1 - lowered the input voltage to 2.5v from the regular 5v using the 3.3v pin on the Arduino and a 150 Resistor

#2 - added a .01 UF Cap along side the 150 R to ground

Results = now the fluctuation on the ACCEL Z Axis is from -9.70 to -9.95... pretty cool!

I'm sure it won't be really noticeable in flight since it was really rock solid before but thought that was a neat trick to quite that sensor down some; from the data sheet those sensors on the 9 degrees of freedom are supposed to be supplied with 2.4v anyway.

Pretty cool!!

I wanted to add some pictures of the output for you to see... plus I had some updated data by adding in a new capacitor. When I added in that new capacitor it dropped my range from that -9.70 to -9.95 to a solid and constant -9.8 which I count as a VICTORY!! It's a quick fix and from what I can see really makes that sensor nearly 100% noise free!!

For my tests I used 2 separate 2.1 AQ Shields, a 2 nearly new 9DOF sensors: (these are the tests I ran and I always ran them on both shields and both 9DOF so I would have duplicate results for the noise on the Axis Z Plane):

1. I powered each by the normal method from the AQ Shield = Picture labeled CONTROL below
2. I powered both from the Shield using 5v from another pin (on the level converter) = no change to noise
3. I powered both from a 3.3v source on the Shield = slight reduction (I would say about 50% less noise)
4. I then used a separate power-source from a filtered power-supply I had from Range Video; I did the above tests again using 3.3 and 5v to see if I just had dirty power from the AQ or the Arduino = no changes from the above findings aka, my power is clean on the shield.
5. Starting with the 3.3v from the AQ Shield, since that was my best results, I then added in a 150ohm resistor to drop the voltage down to 2.4v (the specs in the data sheet for those sensors, I know the board has a voltage step-down that takes 5v to 3.3v but I just wanted to see what would happen). = ACCEL Z Axis now ranges from -9.70 to -9.95 (getting cleaner!!)
6. Keeping with that set-up I then added in a .01uf Capacitor to ground in line with the 150ohm resistor and re-did everything = a solid and constant -9.8 which I count as a VICTORY!! (picture labeled VICTORY!)
7. So knowing that the 9DOF has a voltage regulator I then re-did the above tests with the resistor and capacitor but this time using the pin-out 5v that is supplied for the 9DOF on the 2.1 Shields = ACCEL Z Axis is from -9.70 to -9.95

My conclusion for this is that it's the conbination of using a lower start voltage, along with the resistor and capacitor that produces the nearly noise free range on the ACCEL in the 9DOF.

Enjoy and let me know what you think; I also tested all this away from strong electrical noise and everything; I didn't do it in a "clean room" or anything but I think it was pretty solid - oh and why did I do this; because I was insanely board one windy and cold afternoon while my girls were talking a nap... so this is what you get! LOL

(Control Picture)


(VICTORY Picture!!)

[Sicarius]

Ha nice work on the noise! (although i don't think the Z accel matters that much, from what i know it's only used as a gravity indication :P)
Not such great work on the finger.. calibrating with props on is not the greatest thing to do! I've been punished by doing things quick quick way too often! (see below for a small amount of those punishments)

[MikeD]

@jbsmith78: Was the noise only on the z axis?
I would have expected power supply noise to affect all axes.

[jbsmith78]

@jbsmith78: Was the noise only on the z axis?
I would have expected power supply noise to affect all axes.

Yeah, just on the Z; it's always been that way - on the 6DOF and then on these 2 9DOF I was using for testing.

[jbsmith78]

Just uploaded the new 3.2 software and the 3.2 config (mac OSX) and took it out for a test flight; awesome... I was using the 3.0 version before for all my fun and the 3.2 seems more stable and a lot more responsive from what I can see. Still working on getting my ground station setup with the DVR but I should have some first FPV flights to post before long. Getting a lot of FUN goggle flights in though...

TOO MUCH FUN!!!

[Trebhill]

Sup buddy...long time. Just reading about your thumb incident...wow! Glad it wasn't worse. Reminds me of the movie 'Me, Myself and Irene'..."AAAAAHHHHHHHH, where's my thumb?"

Trying to find time to get back in the game myself, been busy with school. Almsot over, graduating in May! WAHHOOOO