This question was also asked in the Consumer & Computing forum at:
TUSB4041IPAPRQ1 current draw of 450mA.
We are seeing high current draw of 320mA on the 1.1V pins of the TUSB4041PAPRQ1 on a customer's board AND it can be replicated on the TUSB4041PAPEVM board when the TUSB4041PAPRQ1 is configured for SMBUS mode (external register configuration performed by software). It does not occur when the TUSB4041 is configured for i2C slave mode and is configured by I2C EEPROM.
We are using a 200mA LDO supply for the 1.1V rail, and it is going into over-current protection mode oscillations which persists until bit 0 in SMBus register 0xF8 is cleared by writing a 1 to it. After doing this (when the SMB registers can be accessed - it takes several tries to catch the TUSB4041 in a receptive SMBus/I2C bus slave state), the upstream port is enabled and the USB hub operates normally and the current on the 1.1V supply drops to about 50mA.
On the TUSB4041PAPEVM board, the current on the 1.1V supply is a consistent 320mA (about 400mA when includeing the 3.3V supply pins) when the TUSB4041 is powered up with:
- SMBUSz pulled low with 1K resistor
- SMBDAT pulled to 3.3V with 1K resistor
- SMBCLK pulled to 3.3V with 1K resistor
The TUSB4041 datasheet says, “The external interface is enabled when both the SCL/SMBCLK and SDA/SMBDAT pins are pulled up to 3.3 V at the deassertion of reset. The mode, I2C master or SMBus slave, is determined by the state of SMBUSz pin at reset.”
Table 7.7 Hub Input Supply Current in the TUSB4041 datasheet says the current on the 1.1V supply is 86 mA for “2.0 host / 4 HS devices”. So the customer did correctly select the 200mA LDO supply for this design.
There is no chip errata document for the TUSB4041.
Is this perhaps a silicon bug?
