BQ40Z80: BQ40Z80 SMBUS not working

Hi,

There are a few things from your schematic that could be causing this behavior.

First, I would be interested to see what is connected to the BMS_VCC net. Could you provide a schematic of where this connection goes? Id like to see how it differs from BATT+. If the voltage here is too low, the device may brown out and not communicate.

Second, it may be wise to change the ground connection on SMBC and SMBD from BATT- to P_GND if the issue persists.

To help me further debug this, could you send logic analyzer scope captures of SMBC and SMBD when you are trying to communicate?

Thanks,

Jackson

Hi,

The BMS_VCC is connected to BATT+ through MOSFET. Currently we are bypassed MOSFET by removing and directly connecting.

The same battery voltage appears at BMS_VCC

Also the ground connection is provided to only esd diodes.

Can we remove the diodes and check if it is working

Any other reasons

Understood,

If you can send me the logic analyzer scope captures I will be better able to find a root cause of the behavior.

Thanks,

Jackson

We have tried by removing the ESD diodes D6 & D7, but not working

The logic analyser and CRO waveforms are given below.

The same code is working  while connecting the STM32 EVK & BMS EVK

The waveform in the evk is given below

Thanks for sending these over. I am going to sit down and look at these now and will get back to you by the end of the day tomorrow.

Best Regards,

Jackson

Any update on this

We are not getting any voltage on the pin PBI

Hi,

Thanks for providing the logic analyzer captures. It is clear that for some reason on your board the BQ40Z80 is not acknowledging. Also the fact that you are getting no voltage on the PBI pin tells me that the device is most likely in shutdown mode. What is the voltage you see on BAT?

Thanks,

Jackson

In BAT 12v is connected, and we are getting the same. It is the same voltage as that of BATT+

Have you seen this behavior on multiple units? It is possible that the unit in your build has taken some sort of ESD or physical damage so it would be good to rule that out if possible.

Thanks,

Jackson

We had changed the Ic and it is not working

Also we have tried in multiple boards

Hi,

Can you provide me with some more info on where the shutdown pin is connected? Also what is the voltage you are observing at this pin when trying to communicate?

Thanks,

Jackson

Currently the shutdown pin is not connected anywhere. It is left open

Hi,

I have a few things we can check,

1. Can you check the TS pin to see if it is pulsing every 250ms?

2. Can you check to see if the device is in ROM mode by sending a command to 0x16 and seeing if you get a valid result?

If it is still not working could you disconnect the battery and use a power supply to force a voltage on PACK and VCC?

Thanks and let me know your results,

Jackson

Today BMS starts communicating after connecting PACK & VCC. 

In real case we cannot connect it to VCC. I think a pulse is required at PACK to wake up the BMS.

Is it correct, whether we have to modify the schematics.

Good to hear we are at least able to get communication now! I think that confirms that somehow the gauge was in Shutdown Mode. Are you only able to communicate now while connecting PACK and VCC to a voltage source or does it still communicate after you reconnect these pins to the battery? I'm trying to get a full understanding here so we can avoid schematic changes if possible.

Best Regards,

Jackson

At the power on pack and VCC has to be connected, after that we can remove it, communication stil happening after removal also. But connection is needed to start the communication. In schematics shutdown pin not connected anywhere, it is left open. 

How the PACK+ connection is established in the current schematics if we cannot short to VCC. Any change in scheme required at PACK+

Can you clarify what exactly you are asking in that last question? I am not sure exactly what your question is.

Thanks,

Jackson

You have mentioned that voltage is required at the PACK+ at the initial booting. In our schematics the PACK+ is not connected anywhere, then how voltage appears at the boot up

Ancy,

it appears that PACK is connected to BATT_OUT. Am I misinterpreting your schematic?

Best Regards,

Jackson

But at the we didn't found voltage at BATT_OUT

Yes, you will need to supply voltage here. Think of this more as your system voltage or connection to the charger that either pushes current into or drains current out of the battery. This node will not inherently have a voltage simply from the insertion of a battery.

Best Regards,

Jackson

In EVK at the start of development we had shorted PACK+ & BATT+. But now it is working without any short given at boot up. But in our board for every turn ON we have to short PACK+ & BATT+. I understand that inorder to wake up the IC from shutdown we have to connect charger at the start up, by connecting charger at BATT_OUT, PACK+ will get the voltage and IC will be turned ON. But in our condition we have to connect to charger for every turn ON. Hope you got the point. 

Hi,

Even on the EVK you need to have voltage on PACK+ whenever booting up. This is only true when the battery is removed. If you leave the battery connected, the gauge will stay powered on from the BATT+ connection after PACK+ is removed but once a connection to BATT+ is also lost, a PACK+ connection is required to reboot the device.

Hope that clears things up,

Jackson

Charger is connected to PACK+ . So in your reply I understand that the ic will reboot once it is connected to charger. So if I remove the battery I have to connect to charger before applying load at PACK + inorder the reboot the IC. Is it correct