IMU stops abruptly, is it dead?

Other Hardware Sensors
1

Hello,
A while after I managed to get reasonable output from my Drotek 6050 IMU board while watching output values in serial monitor it suddenly stopped displaying them and by that I mean all earlier values were gone and it was just a blank page.
This happened while running this code with Kalman filtering from TKJElectronics. Then I tried running Krodal's code, which again gave no results (unlike before). Then I tried running the I2C scanner and got told that there's a device on 0x69 (as should be). Now, when running Krodal's code I get this:

InvenSense MPU-6050
WHO_AM_I : 68, error = 0
PWR_MGMT_1 : 40, error = 0

MPU-6050
Read accel, temp and gyro, error = 2
accel x,y,z: -5374, 0, 300
temperature: 37.212 degrees Celsius
gyro x,y,z : 2050, -15552, -32255, 

MPU-6050
Read accel, temp and gyro, error = 0
accel x,y,z: 0, 0, 0
temperature: 36.506 degrees Celsius
gyro x,y,z : 0, 0, 0, 

MPU-6050
Read accel, temp and gyro, error = 0
accel x,y,z: 0, 0, 0
temperature: 36.506 degrees Celsius
gyro x,y,z : 0, 0, 0, 

MPU-6050
Read accel, temp and gyro, error = 0
accel x,y,z: 0, 0, 0
temperature: 36.506 degrees Celsius
gyro x,y,z : 0, 0, 0, 

MPU-6

As you can see it just stops mid-word and doesn't print anymore. Then if I wait long enough it shoots some zeros at me again and then stops once more. Any ideas why is this happening and if the IMU is dead?

Additional info:
The board I'm using is from Drotek
I was not using pull-up resistors for SDA/SCL to analog pins - can this be the case somehow?

Maybe anyone reading this could also recommend a good tutorial explaining how to use perpendicular axes of accelerometer alone to get more reliable pitch and roll data in dynamic conditions?

Thanks!

2

Check if there are pullups installed on your board (your supplier unfortunately doesn't supply schematics for the product). Disconnect the breakout from your Arduino and measure the resistance between Vcc and SDA/SCL with a multimeter. Post the results.

Error 2 means that the device didn't react on the I2C bus, so it's possible that your device got broken, although I cannot see a reason at the moment.

3

Both of these show 4,6k? resistance, which sounds fairly decent, since I can see two resistors labeled 472 on board.
I get the meaning of error 2, however why doesn't it indicate that error all the time then (all further print ins are error=0)?

4

I get the meaning of error 2, however why doesn't it indicate that error all the time then (all further print ins are error=0)?

That's a good question. Do you have a logic analyzer or scope to attach to see what's going on the line?

5

Unfortunately I don't have any of these, nor do I know someone who does.

6

Please post a wiring diagram of how you connected the IMU to the Arduino.

7

What is a reasonable software to draw diagrams with? The ones I found via forum search mostly have dead links or rather absurd prices for a few uses.

The way I had IMU connected is:
3.3V -> 3.3V
GND -> GND
INT -> D2
SDA -> Analog 4
SCL -> Analog 5

No extra resistors were used.

8

I would try adding another pair of pull-ups (4k7) to lower the resistance to about 2k3 which is a reasonable value for 3V3 if the connection between the module and the Arduino is more than about 20cm (8").

How do you power your setup? By USB?

What is a reasonable software to draw diagrams with? The ones I found via forum search mostly have dead links or rather absurd prices for a few uses.

Fritzing (http://fritzing.org) is a nice and easy-to-use program for such tasks and it's free.

9

Yes, it's powered by USB while troubleshooting.
I'm not entirely sure how am I supposed to add those resistors in parallel to ones on board since I most likely won't be able to solder them properly. Should I search for documentation on anything related to disabling built-in pull-up resistors or something like that is not doable? Could just use shorter cables for test purposes and see if IMU is operative with those as well.
Connecting cables are currently 1m long each, but if it somehow starts working I will most likely need to have each of them at least 2m long. This could cause an issue I suppose?

10

Yes, it's powered by USB while troubleshooting.

Check the voltage between GND and 5V during operation. What value do you get? USB is often not very stable in voltage if a bit of current is drawn.

I'm not entirely sure how am I supposed to add those resistors in parallel to ones on board since I most likely won't be able to solder them properly. Should I search for documentation on anything related to disabling built-in pull-up resistors or something like that is not doable? Could just use shorter cables for test purposes and see if IMU is operative with those as well.

You don't have to solder the resistors onto the board. Just connect them from SDA/SCL to 3V3.

Connecting cables are currently 1m long each, but if it somehow starts working I will most likely need to have each of them at least 2m long. This could cause an issue I suppose?

An I2C bus shouldn't be more than about half a meter in total length otherwise the signals are flattened too much from the cable capacity. Move the Arduino directly beside the IMU or use I2C bus extender chips (on both sides).

11

Thanks for all that help, learned a few things in the process! It seems safe to say that this IMU is done from what I've checked. Will receive another one in a week or so and check how that runs.

Tested voltages on USB. 5.04V and 3.28V for appropriate connections. Seems OK.
Switched out for shorter wires - no results. I2C won't even find it anymore. Krodal's code won't give any output just the default print ins. After reducing pull-up resistance 2 times there's no significant change either. Krodal's code gives error 2s all the time then.