navX MXP Startup Issue SPI/Labview

[Eric Francis]

I am using the latest navX firmware and roborio firmware. I have implemented the suggested labview initialization in robot main for an SPI port connection.

I primarily use the Yaw information in our design for drive stabilization. About 25% of the time after a powerup, I get no live data from the navX. Rebooting the robot code (i.e. software reset) resolves the communications issue. The navX is set at 50Hz update and I read the data in periodic task. I never loose communication once it starts. Performance is fine. It is plugged directly into the expansion connector. This issue is on both of our robots so it is not a pcb issue. It is also not a power supply issue as it is not a function of our battery voltage. We have a case protecting the top of the board so viewing lights on the pcb iss difficult.

How can I diagnose this problem?

Thanks,
Eric

[admin]

Hi Eric,

Can you please try using I2C instead and let us know your results?

Thanks,

-scott

[Eric Francis]

Thanks for the reply. I will test it this evening and report my results.

Eric

[Tim Easterling]

Hi Eric,

Are you using the same navX on both robots or do you have two different navX-MXP's on each robot?

Tim

[Eric Francis]

We have 2 roborios with separate navx boards. Tonight was a busy night with our drive team but I was 10 for 10 after changing from SPI to i2C. I will continue to monitor it.

[Eric Francis]

Changing from SPI to i2c fixed the issue. Running fine for a week.

[18wi...@abschools.org]

I've had a similar bug - when communicating over SPI the NavX fails to initialize correctly (all lights go off after powerup). I resolved it by putting in a 10 sec delay between when power gets to the NavX and when I open the SPI bus. I don't know what the minimum delay needed is, but I know 2 sec is not long enough.