GY_86 MS5611, HMC5883l and mpu6050 sensor module

[tovrin]

http://www.ebay.com/itm/10DOF-IMU-MS561 ... 1e6f620a83

ok, i picked up one of these, it arrived yesterday, i am going to try to get it connected tonight. is there anything i need to know? I have read the thread in development about this sensor, i know the DMP features dont work, yet, and i dont need to use the Fsync, inta, or drdy connections, right?
which is best to use, the 3.3v or 5v? i have both on my FC.

[copterrichie]

I used the 5 volts (VCC_IN), SDA, SCL and GND.

[tovrin]

thanks Rich, was it difficult to calibrate? ive read up on calibrating the mag, calibrating the acc i assume is the same as the wii parts im replacing, what about the barometer? im going to place some wind shielding over the sensor (a piece of soft foam i think will be appropriate) even though its usually covered by my canopy

[copterrichie]

the MPU6050 does not have a Mag, just gyro and acc.

[tovrin]

ok, i misunderstood somewhere, this board has barometer, magnetometer and a mpu6050 on it, i thought the 6050 included the mag and baro.


(updated the title to reflect that, thank you)

[tovrin]

ok, apparently my ignorance increases as time goes on, its printed very plainly on my board, and in the picture above that its a gy-86, which i didn't know even existed as a chip set until i was looking through config.h and spotted it.

so i solder the wires, boot up the board with ftdi, changed sketch to reflect new board (selecting freeimu because i saw all the same components), leaving pullups enabled (not knowing they weren't necessary with this board) the board shows a green light, the GUI is bouncing all over, i2c errors going from -7000 to 7000 back and forth. so i unplug, check my soldering again, realize my mistake on the gy-86, disable pullups change to 400khz, each time booting it up and seeing crazy readings, then it quits, doesn't light up anymore, and i get nothing from GUI except same i2c error pattern.

was it that easy to kill?

[copterrichie]

If you have a good magnifying glass, look to see of the voltage regulator is cooked, if not, you maybe OK. Make sure you have it wired correctly and using the right MWC settings, these boards should not require the Internal Pullups.

[tovrin]

everything seems intact, i tried using the 3.3v and same result, nothing. if no one has a better idea by tomorrow, ill contact the vendor

[tovrin]

I plug my old wmp back in without acc, and im getting the exact same results. ive done something wrong, i have an arduino dueminalove im going to load the multwii sketch on and see what i can learn with these sensors.

what exactly does it mean when the i2c error scrolls from 30k - 0 - -30k? what would cause that?

[tovrin]

I hooked my gy-86 up to my dueminalove and it works fine. something on my paris board is malfunctioning

[copterrichie]

Hmm, that is good news actually that the sensor board is not defective or burned out. I know you will resolve the issue however, it should be a very easy connection. Just the SDA, SCL, Power and Ground.

[copterrichie]

Oh, I don't have a Paris board but maybe there are pullup resistors on that board.

[tovrin]

the paris board 3v and 5v output are no longer giving out proper voltage, the 4v5 was pushing out 5v so i connected gy-86 to it and it worked, for about :30 then the light on the sensor started blinking and the i2c errors started flashing, i unplugged it, am not sure what is causing this, i let it run on my dueminalove for minutes and it didnt do that.

the board does not have pullups, they are disabled in my code

[copterrichie]

sounds like a voltage regulator problem on the Paris board, can you bypass it?

[tovrin]

I was wrong about it not doing it on the dueminalove, just took longer, i am returning the gy-86 for a new one, going to have to find a new board now too.

[tovrin]

the vendor is giving me shit saying it was fully tested, and hes not gotten one bad response on ebay over this chip.

[copterrichie]

the vendor is giving me shit saying it was fully tested, and hes not gotten one bad response on ebay over this chip.

Where are you located?

[tovrin]

ok, Rich, i have to assume that i had some solder splatter or something causing a short, my 3.3 and 5v outputs are pushing mv now. but the one labeled 4v5 is working fine, giving me 5v solid. i managed to get it working today, everything seems fine, though apparently my old PIDs arent as useful with the new brains.


thanks everyone for the help!

[jevermeister]

Try the LPF,s The new sensors are far mor accurate than the WMP, but are very sensitive to vibrations,
I changed the LPF and did not touch the PID

Nils

[tovrin]

i will check that out tonight, pending the weather permits any flight.

[tovrin]

ok, here is my report. i changed the LPF to 188 and didnt feel much different, acc mode still not grabbing the same everytime, then i changed to 98hz, and suddenly its a new bird. acc mode locked into a small radius at a height of only about 6 foot, turned on baro and mag just to turn them on and let it hold while i tried to explain to the lady how awesome this moment is...

anyway, something interesting happened i sisnt expect and am hoping is common, and/or means something is right or wrong, once i had it perfectly level, the props started resonating, it sounded like what you hear when you run your fingers around a crystal glass, and as soon as i moved the quad it stopped until i was level again, or rather until it came to a hover.

tell me this is common, and means i achieved quadcopter karma or something!

these are gemfan 11x4.7 colored props, 750kv motors.

[jevermeister]

Hi, cool,
can you make a video of that?

This seems to be the rare occurance of a hover rpm that is equal to the resonating frequency of your frame.

It is not good by the way, control your nuts and bolts after that

[Alexinparis]

anyway, something interesting happened i sisnt expect and am hoping is common, and/or means something is right or wrong, once i had it perfectly level, the props started resonating, it sounded like what you hear when you run your fingers around a crystal glass, and as soon as i moved the quad it stopped until i was level again, or rather until it came to a hover.

tell me this is common, and means i achieved quadcopter karma or something!

This is what I noticed with a plastic frame (dji style).
This is obviously not good at all because LPF is just a “stability patch” in this case;)
I never encountered such a problem with alu arms.

[tovrin]

ill try to make a video tonight.

besides making sure everything is tight, what can i do? my props are balanced, my bell housings as well.

[jevermeister]

The resonance will lead to loosing anything, you could alter some parameters of your frame to move the resonance frequency up or down, my resonance point is near full throttle, I use PID attentuation to compensate it.

Frame parameters:
- weight
- material
- diameter
- motors etc

just add some weight or move the lipo a little off center and you will see it will have a shifted resonance

this is very intriguing I am so looking forward to your video

[tovrin]

i could not replicate it tonight, I think the rain all day might have had something to do with that, but you brought attention to a fact i hadnt really given notice too, my battery has some slop in its place, I need to secure it better.