luisrodenas

Hello cbanta. First of all, these forums are almost abandoned, so I would suggest you look for help somewhere else, like the Invensense forums. Now some suggestions: in order to detect sound frecuencies I would use Fourier Transform in order to get the frequencies of the signal. If you knew what frequencies you are looking for or at least a range you want to detect, you could use instead some band-pass filters. First you have to capture the signal, I would do this using a microphone. Audio frecuencies go from 20 to 20000Hz (https://en.wikipedia.org/wiki/Audio_frequency) , in order to detect a frecuency there is a law somewhere that you have to measure at double the frequency you want to capture. If you are using an accelerometer, it is hard to reach high measure frequencies. Usually it is about 1kHz, but let's suppose you can get to measure at 10kHz, still you would only be able to capture audio up to 5kHz, missing all audio from 5 to 20 kHz. So that is why I would try with something different that an accelerometer. Good luck with your project and do a larger search online on that topic, I am sure there are lots of info out there.

Hello MattE, Just wanted to make sure you know there is a developers corner at the Invensense's website. http://www.invensense.com/developers/login/ At first there was very little information available about their products, but nowadays they have lots of information, code, examples for different platforms, etc. I guess that's the reason this forum is now kind of outdated, even though things here are still usable, reliable, easy to use and implement, etc... But the most updated features are at Invensense's developers corner.

Hello wambo, please attach the modified version of the sketch that is not working for you and I will take a look.

Hello Simon, I am sure you cannot disable gyro drift compensation, with new libraries MAYBE there is some option to adjust how much it compensates, but that is just a guess. Your easiest option is to implement your own filter for yaw, google "complementary filter" (easy) or "kalman filter" (advanced). Good luck.

Hello Vickylance. Please look for the MPU6050 datasheet and read it. There you will find the values needed to convert measures to degrees/second. Your assumption is partially wrong. 250º/s is more sensitive but has a NARROWER range, only +-250 º/s. 2000º/s is less sensitive but has a WIDER range.

Hello gilperon, There is no way of getting DMP angles without fifo buffer, but if I am not mistaken there was a way of just leaving angles and removing giro and accel data from buffer... Anyway, I think the easiest way is to make your arduino read that buffer and use or discard data, but read it. It takes very little time if you are using fast I2C. By the way, you should use gyro data as well to control your quadcopter. Sorry I can't be of more help. Good luck

Hi, I think you cannot call the function readBytes() from an ISR, at least you shouldnt. The reason for this is because that function includes I2C communications, that take some time and also may conflict with ISR manager for reasons I don't really know. Just use the timer interrupt to change the value of a flag, and then check that flag from main code and execute the readbytes().

Gyro returns ANGULAR VELOCITY, not angle. You could integrate this measure to get angle, but better apply some sensor fusion algorithm. By the way, I think pin INT should not be connected to 3.3V.

Check these pictures: http://www.google.com/imgres?imgurl=http%3A%2F%2Fmk90.ru%2Ffiles%2Fimg%2Fpromini_pins.png&imgrefurl=http%3A%2F%2Fmk90.ru%2Fstore%2Fru%2Fmicrocontrollers%2F320-freeduino-pro-mini.html&h=346&w=530&tbnid=KWPd379HUDUazM%3A&zoom=1&docid=0E1PS3sfMZXUjM&ei=gJSiU-LPOI6u7AamyYGYDA&tbm=isch&ved=0CCsQMygPMA8&iact=rc&uact=3&dur=4756&page=1&start=0&ndsp=18 http://www.google.com/imgres?imgurl=http%3A%2F%2Faeroquad.com%2Fattachment.php%253Fattachmentid%253D4210%2526d%253D1315917105%2526stc%253D1&imgrefurl=http%3A%2F%2Faeroquad.com%2Fshowwiki.php%3Ftitle%3DHardware%2BAssembly%2BMini%2BShield%2Bv1%2B0&h=493&w=640&tbnid=qT03lmkFYKOWDM%3A&zoom=1&docid=T02Xc-RT9lrLNM&ei=gJSiU-LPOI6u7AamyYGYDA&tbm=isch&ved=0CC4QMygSMBI&iact=rc&uact=3&dur=2774&page=2&start=18&ndsp=23

¿? The pins just above A2 and A3 are A4 and A5.

It looks like you are using an Arduino Pro Mini. At the official Arduino website (http://arduino.cc/en/Main/ArduinoBoardProMini) you can see that the I2C pins are A4 and A5, not 4 and 5 that you are using. Maybe that's the problem.

Yes. As far as you read the DMP's fifo fast enough so it does not overflow. Also, you can change fifo's frequency to fit your needs.

Hi, Trivial question, but just to confirm things: Are you using MSP430? We need more info so somebody can help you. MPU's board? MCU board? Wiring? Example you are trying to run?

Hello deck, That behaviour you describe can be found with the YAW measure, but shouldn't happen with PITCH and ROLL. Rigth now I cannot confirm you if this is happening in my MPU because I never use it for such long times. So, have you calibrated DMP with your offsets? What measure are you using (yaw, pitch or roll)? Your project sounds really interesting, is it online somewhere?

I use MPU6050's 6 axis DMP to control quadrotors. It works fine as long as the sensor is well placed in the quad. The noise depends on the FIFO's frequency (200Hz has more noise than 100Hz), but it is almost neglectable in both cases (around 0.2º, search my other posts to find more info). I have also tried raw values + Kalman, and it works also fine, but I believe DMP is enough and easier. I have tested its measures versus a Microstrain 3DM-GX2 (2000 euros a few years ago), and it is almost identical. Good luck with your proyects!