TXB0304: driver strength

Feng Wu

Part Number: TXB0304
Other Parts Discussed in Thread: SN74LVC125A, SN74AUC34

In customer use, they use a 3.3V for both VCCB and pull-up voltage. But if they use 4.7k pull-up resistor, the Voh just 0.9V while they use 1k pull-up resistor, the Voh=3.3V. So it looks the souring current of B I/O pin is limited. I want to check what's the maximum sourcing current of the B I/O pin, I'm not clearly with the description of datasheet.

over 5 years ago

0 Clemens Ladisch over 5 years ago

Guru 317480 points

The sinking current is specified as the test condition for V_OLx; it's 20 µA.

0 Chad Crosby over 5 years ago in reply to Clemens Ladisch

TI__Expert 6680 points

Hi Feng,

Check out section 8.4.2 of the datasheet. It goes into detail about pullup/pulldown resistor sizing.

Thanks!

Chad Crosby

0 Feng Wu over 5 years ago in reply to Chad Crosby

TI__Prodigy 300 points

Thanks Crosby, one more question, in datasheet I see For proper operation, the device driving the data I/Os of the TXB0304 must have drive strength of at least ±3 mA. If the input driver strength is not enough, could we add external pull-up to supply 3mA input current?

0 Clemens Ladisch over 5 years ago in reply to Feng Wu

Guru 317480 points

How would you pull that signal low? Is the driver stronger when sinking current?

For an unidirectional signal, a unidirectional or direction-controlled level shifter would be a more robust solution.

0 Feng Wu over 5 years ago in reply to Clemens Ladisch

TI__Prodigy 300 points

Thanks Ladisch, I need to clarify my question: In customer application, the direction is 3.3V to 1.8V, and they use pull up resistor in 3.3V side. the 3.3V is from DC/DC and both supply Vccb. If they use 4.7k, the VioH=0.9V while VioL=0.2V; If they use 1k, the VioH=3.3V while VioL=0.2V. So they want to check if there is any risk in the pull up resistor.

0 Clemens Ladisch over 5 years ago in reply to Feng Wu

Guru 317480 points

Section 8.4.2 of the datasheet says:

If pull-up or pull-down resistors are connected externally to the data I/Os, their values must be kept higher than 20 kΩ to ensure that they do not contend with the output drivers of the TXB0304.

Do all four channels go from 3.3 V to 1.8 V?

0 Feng Wu over 5 years ago in reply to Clemens Ladisch

TI__Prodigy 300 points

yes , all four channels go from 3.3V to 1.8V. But the issue is found in input side. before connected TXB0304, the periperal I/O 3.3V is normal, but after connected with TXB0304 I/O(pull up resistor is 4.7k), the voltage is pulled down to 0.9V, it looks like the periperal I/O sourcing current is too low to drive I/O pin to 3.3V. After changing pull up resistor to 1k, the souring current into I/O increased so the voltage of I/O pin can go to 3.3V. So is it any risk in this usage? If it is dual-direction application, you mean it maybe compete with output driver, how about single direction applications,if the 3.3V is only used for input?

0 Clemens Ladisch over 5 years ago in reply to Feng Wu

Guru 317480 points

The TXB is always bidirectional; its output drivers on both sides are always active. The 3 mA drive strength requirement is necessary to ensure that external devices can overpower the TXB outputs.

For this application (unidirectional downtranslation), I would recommend a simple buffer with overvoltage-tolerant inputs, e.g., SN74LVC125A or SN74AUC34.

Logic

Logic forum

TXB0304: driver strength