This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

IO-EXPANDER-EVM: Standby Mode conditions

Part Number: IO-EXPANDER-EVM
Other Parts Discussed in Thread: TCA6424, TCA6416, TCA9535, TCA9534

Good evening, 

On my design board , which is battery powered, I need to optimize the power consumption during sleep mode ( 40uA at 3.3V) , pull-up resistors on SDA and SCL are switched off.

But I noticed that my eval board IO-EXPANDER-EVM  is withdrawing 2.5mA of current ( I/O configured as input, both pull-up resistors R9 and R14 are removed as I am using pull-up resistors which can be power-cycled)

It seems that the IO Extender used on this eval baord doesn't go to standby mode as both SDA and SCL are not pulled-up . 

I have done some research and I figured out that some IC like TCA 9534 and TCA 9535 doesn't need to have both SCL and SDA pulled-up to go to standby. can you confirm this ?

Is there any IO extender which can be configured in standby mode via software ?

many thanks for your guidance, 

damon

  • CMOS inputs always must have a valid voltage (near VCC or GND) to prevent shoot-through currents; see [FAQ] How does a slow or floating input affect a CMOS device? You must not disable the pull-up resistors unless the device is actually powered off.

    ("Standby" simply means that the I²C bus is idle. These I/O expanders do not have an internal oscillator; the state machine is triggered entirely by the SCL clock edges. Standby is not really a separate operating mode that needs to be entered or left.)

  • Hi Damon,

    On my design board , which is battery powered, I need to optimize the power consumption during sleep mode ( 40uA at 3.3V) , pull-up resistors on SDA and SCL are switched off.

    What is the reason to switching off the pull-up resistors? I2C is an open-drain protocol. The pull-up resistors in the idle state are drawing very little current due to high impedance of the open-drain FET and the high impedance of the input buffer. Switching off the pull-up resistors could create a false clock edge on SCL. I don't see a reason to switch these resistors off during "standby mode." 

    But I noticed that my eval board IO-EXPANDER-EVM  is withdrawing 2.5mA of current ( I/O configured as input, both pull-up resistors R9 and R14 are removed as I am using pull-up resistors which can be power-cycled)

    It seems that the IO Extender used on this eval baord doesn't go to standby mode as both SDA and SCL are not pulled-up . 

    Removal of R9 and R14 would cause SDA and SCL to not be pulled-up. Adding the 4.7kohm pull-up resistors should fix this problem, and not add any noticeable additional current. 

    The 2.5mA of current could be from multiple sources. TCA6416 and TCA6424 are on the board by default, measuring current through VCCP or VCCI is a combination of the two devices. Supply current is also used to power LED's on the board as well. 

    What devices do you have on the EVM? 

    I have done some research and I figured out that some IC like TCA 9534 and TCA 9535 doesn't need to have both SCL and SDA pulled-up to go to standby. can you confirm this ?

    Where are you reading this information? From the conditions in the electrical characteristics of the datasheet, the measurement of standby current is the same across devices TCA9534 and TCA9535 as well as the 6416 and 6424. 

    Is there any IO extender which can be configured in standby mode via software ?

    Our IO expanders are considered to be in "standby mode" when...

    • All p-ports are configured to be inputs
    • All inputs are biased to either VCC or GND and not left floating to prevent shoot-through current (see Clemens Comments)
    • Output current is = 0
    • I2C is idle (pulled high through pull-up resistors) SDA = SCL = VCC, fscl = 0 Hz (no clocking on the i2c bus). 

    Regards,

    Tyler