Hello,
I am trying to change my pullups on the SMBD/SMBC lines from 10K to 3.3V to 10k to 2.5V. For some reason this is causing me to not be able to communicate with the EV2400. I tried looking at the schematic for the EV2400 but couldn't determine the pull up voltage as it appears to be based on a digital pot. Any clues as to why this isn't working as I thought this whole system could be run off of 2.5V?
Thanks, Mike
Matt,
Sorry for the delay, jumping around on multiple things.
To give you some background, these SDA/SCL lines (currently pulled up to 3.3V) connect to the BQ78350-R1 and to my MCU which is on the "PACK" side of the FETS. My MCU is connected to a 3.3V rail so therefore needs the Logic HIGH to be 0.8*3.3V = 2.64V. If I were to pull up to 2.5V, I would never be able to communicate with my MCU. So what happens often, is we need to communicate with the battery when the FETS are OFF (end of discharge, permanent failure, etc.). Since the pullups are to the 3.3V rail, this rail is OFF when the FETs are OFF so we cannot communicate with the Bq78350-R1.
Now that you have the background, I'm trying to diode_or the pullup to 3.3V and 2.5V. This way when the 3.3V rail is running, the lines are pulled up to 3.3V (or a small diode drop below 3.3V), and when the 3.3V rail is not running, the lines are pulled up to 2.5V instead and at least I can still communicate with the BqStudio through the EV2400. This is where the problem came in the other day. That being said, here are some scope shots:
Powering off of 3.3V, my processor is actually in reset so the only ones communicating should be the bq78350 and the EV2400. I have these settings on the software:
Here is the SCL and SDA lines measured at the board.
Next I sent the "FET_EN" command to actually turn off the FETs. The 2.5V rail is still running, but the 3.3V rail is shut down which should mean the lines are pulled up to 2.5V. I've lost connection to the BqStudio, the greyed out chip and battery on the left hand side goes from grey to normal, back and forth.
Now there is nothing worth capturing on the oscilloscope, the lines are steady with no indication of communication. Here is an image of my circuit with the voltage I'm measuring at each point now.
So the lines are being pulled up to 2.227V, which is over the 2V min for the bq78350-R1. If the SDA/SCL lines are pulled to ground, We would have about a 2.5V / 5k = 500uA of current. The diode drop at 500uA should be less than 300mV (since the spec for the diode is 310mV @ 1mA).
What do you think?
Thanks,
Mike
