search

IEEERobotics / bot

Robot code for 2014.
BSD 2-Clause "Simplified" License
18 stars 11 forks source link

I2C Issues with Servo Cape and other devices #454

Closed SeanKetring closed 8 years ago

SeanKetring commented 8 years ago

Some bug in the I2C routine on the servo cape is causing the bus to fail. If the user inputs a read command to the servo cape it will break the communication. So to use anything in the stack, do not run I2C detect and do not attempt to read from the Servo Cape. This is a way to get around this bug, until we can figure out what it is.

AhmedSamara commented 8 years ago

When you run an i2cdetect -y -r 1, the read is extremely slow, and no devices are detected,

It looks like the issue is beign caused by the SDA being held low, but we're not sure why or how to fix it.

similair issues happening here
http://lm-sensors.lm-sensors.narkive.com/FFQ2T3ZL/i2cdetect-running-slow-but-not-bus-inactive-slow-not-sure-why

AhmedSamara commented 8 years ago

According to the TI I2C troubleshooting guide:

Adding more I 2C and SMBus devices on the bus may exceed the 400-pF limitation. I 2C multiplexors, I2Cswitches, and I2C buffers and repeaters can isolate slave devices that are not currently needed to reducethe overall system loading, and then meet the maximum load capacitance specification.

We do have a lot plugged into the bus right now, maybe this is causing problems?

AhmedSamara commented 8 years ago

Never-mind, it's not a hardware issue.

Removing the I2C module from the PIC on the servo cape fixed the problem. It's definitely the servo cape holding the line low for some reason.

@jasteve4 just notified us that there's a bug in some PIC controllers with I2C buses that force themselves low, so there's that to look into.

Still no idea why that's going wrong.

Removing it and putting it back in fixed it.

SeanKetring commented 8 years ago

PIC was found to have silicon defects, switched to a new processor