Post Reply 
 
Thread Rating:
  • 0 Votes - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Possible MPU Problem
12-15-2017, 04:40 PM (This post was last modified: 12-15-2017 04:46 PM by JLeether.)
Post: #1
Possible MPU Problem
I think I may have a bad MPU....possibly. I've read through a lot of the other posts here and it seems like most people having issues with the MPU are getting further than I am.

My B-Robot has only initialized twice...ever. It wasn't the very first or second time I powered it on, either. I tried and tried and randomly in a set of about 50 attempts, gave me the ready signal twice. It worked fine those two times and responded to all controls. 90% of the time it just sits there. In debugging all I get is:

JJROBOTS
Don't move for 10 sec...

I added a few more serial outputs and discovered it's hanging at MPU6050.ino line 667.

n = Wire.endTransmission(false); // hold the I2C-bus

If I unplug the MPU before powering up, everything else initializes, the arm moves and the motors start to slowly turn. Naturally I get MPU errors in the console when I do that since it is not present.

I'm very experienced with Arduino and stepper motors. Everything is connected properly. I have not specifically used the MPU6050 before (although I do have another robot project partially done that is utilizing one). On the MPU connections I have 5v to VCC, ground to ground, SCL to SCL, SDA to SDA. I have power lights on both the Leonardo and the MPU. The connector for the MPU isn't pressing on the top cover or putting any side load on the MPU itself. On power up, the WiFi light flashes, the power lights come on and the servo does the normal twitch. After that, nothing.

   

I have read that the standard Arduino Wire library can hang on the endTransmission function at times, too. Not sure if this has anything to do with it.

Sadly, my two successes were many, many attempts ago. It's not behaving like it's going to work anymore. If something was kinda working, but in the process of dying, it's dead.

Anyone else experienced this kind of issue?

....and that fast I got it sorted. While, yes, it was plugged in right, the connector between the MPU and the brain board had some....issues. I've got them sorted now.

Apparently my shake down "wiggle all of the connectors" didn't catch it the first time.
Find all posts by this user
Quote this message in a reply
12-16-2017, 06:56 PM
Post: #2
RE: Possible MPU Problem
Hi! Can we get more info about that problem? Were the connectors inside the plug the source of the issue? Bad contact? Or the headers on the PCB?
We are about to integrate the IMU on the Brain shield... (trying to keep the same price)
Visit this user's website Find all posts by this user
Quote this message in a reply
12-17-2017, 02:09 AM
Post: #3
RE: Possible MPU Problem
(12-16-2017 06:56 PM)JJrobots JP Wrote:  Hi! Can we get more info about that problem? Were the connectors inside the plug the source of the issue? Bad contact? Or the headers on the PCB?
We are about to integrate the IMU on the Brain shield... (trying to keep the same price)

The best I can tell, it's a broken wire in the short run between the IMU and PCB. When I bend the wires into a perfect arc so that the wires meet the connectors perfectly straight, I have no issues. Unfortunately when I do that, the wires interfere with the top cover. If I bend them into any other shape, I don't get clean data to the IMU and it never initializes. I've popped the crimp connections out of the headers and they all look like they're crimped cleanly, are making good connection at the crimp and fit tightly on the header pins. It seems odd (and even unlikely) for multistrand wire to be broken inside the insulation, but I'm definitely losing the connection there when the wires either press on the top cover or I bend them to the side so they don't touch the cover. For the moment, I have some standard Arduino jumper wires installed. Not ideal since they are longer and can cause interference, but they're doing the job for now. My plan was to make a suitable replacement. I definitely like the integrated IMU idea. That brain board is definitely a one stop shop for a two motor robot.
Find all posts by this user
Quote this message in a reply
12-20-2017, 04:13 PM
Post: #4
RE: Possible MPU Problem
Ok, thanks for the feedback.
It is... unlikely. But, yes, can happen. We are currently testing the new board with the integrated IMU. It will take some time before releasing it.
Do you want us to you a cable replacement? If so, let us know at: info@jjrobots.com
Visit this user's website Find all posts by this user
Quote this message in a reply
12-20-2017, 04:23 PM
Post: #5
RE: Possible MPU Problem
(12-20-2017 04:13 PM)JJrobots JP Wrote:  Ok, thanks for the feedback.
It is... unlikely. But, yes, can happen. We are currently testing the new board with the integrated IMU. It will take some time before releasing it.
Do you want us to you a cable replacement? If so, let us know at: info@jjrobots.com

Nah, don't go to that trouble for a stubby wiring harness. I appreciate the offer, but I can definitely fabricate something that will work. The robot works fantastic otherwise.

I'm still not convinced there's a broken wire either. There's just something unique about the route of that wire that's causing something to lose connection...somewhere. Even when I do get it to fire up with that connector, the slightest bump sends it into a frenzy. I may just try shortening it a touch so it doesn't touch the top cover, but will still route in a nice, clean arc.

We're using the B-Robot and the Sphere-o-bot for a local TV news spotlight our makerspace is doing on Arduino controlled robots. The Arduino jumpers will get the job long enough for us to get through that. I see an iBoardBot in our future. Keeping an eye on the Blimpduino and that 3D display, too. We set up a table at public events to drive membership and our summer programs. The robots are always winners a the booth.
Find all posts by this user
Quote this message in a reply
12-23-2017, 07:15 PM
Post: #6
RE: Possible MPU Problem
(12-20-2017 04:23 PM)JLeether Wrote:  
(12-20-2017 04:13 PM)JJrobots JP Wrote:  Ok, thanks for the feedback.
It is... unlikely. But, yes, can happen. We are currently testing the new board with the integrated IMU. It will take some time before releasing it.
Do you want us to you a cable replacement? If so, let us know at: info@jjrobots.com

Nah, don't go to that trouble for a stubby wiring harness. I appreciate the offer, but I can definitely fabricate something that will work. The robot works fantastic otherwise.

I'm still not convinced there's a broken wire either. There's just something unique about the route of that wire that's causing something to lose connection...somewhere. Even when I do get it to fire up with that connector, the slightest bump sends it into a frenzy. I may just try shortening it a touch so it doesn't touch the top cover, but will still route in a nice, clean arc.

We're using the B-Robot and the Sphere-o-bot for a local TV news spotlight our makerspace is doing on Arduino controlled robots. The Arduino jumpers will get the job long enough for us to get through that. I see an iBoardBot in our future. Keeping an eye on the Blimpduino and that 3D display, too. We set up a table at public events to drive membership and our summer programs. The robots are always winners a the booth.

Please have a look at your Arduino female connector strips (IO-femal headers and so). At my Leonardo board I had seen many faulty soldered joints. I had to scrape and clean every single one and then to solder again. Cold solder joints can make nasty problems.
Find all posts by this user
Quote this message in a reply
Post Reply 


Forum Jump:


User(s) browsing this thread: 1 Guest(s)