Tool/software:
Hi TI team,
Originally, I would like to use TCA9509 as voltage translator of 5V<-->3.3V, but when I checked the details, I found VIL of TCA9509 is 110mV, which is too low for slave 6# in below schematic.
Can you kindly recommend an option with higher VIL? For example, 0.8V or higher.
By the way, what's the VIL&VOL of LSF0102 ? Is it OK to use LSF0102 instead?
Thank you.
Kind Regards,
Bo
Hi Bo,
I am talking with the FAE that is supporting this issue.
TCA9509 does have a very low VILC requirement. For this application, I do not see a buffer that I can recommend that has lower VILC and has the SVO on A-side. This is why I recommended switching to a level shifter such as the TCA39306 or LSF0102. You won't get buffering with these devices, but you also won't have the limitation of VILC.
LSF0102 or TCA39306 are level translators. They do not buffer. Considering that there are 6 target devices on the B-side of the TCA9617B, I don't think an additional buffer is needed between slaves 1-4 and slaves 5-6.
The LSF0102 and TCA39306 are not buffers and do not re-drive the signal. There is no minimum VIL requirement. However, the VIL on one side of the translator will appear a Ron voltage drop higher on the other side of the translator. See this picture that I drew the FAE.
If you replace the TCA9509 with a level translator, you must consider the pull-up resistances on both sides of the level translator as well as the drive strength of each slave device on the I2C bus (slaves 1 -6). Also, each slave adds ~ 10pF of input cap on top of the 12.5pF (max) that the TCA39306 / LSF0102 could generate. This all plays a role in the rise-time and VOL at the B-side of the TCA9617B which needs to be <400mV in order for the TCA9617B to pass a LOW from B-side to A-side properly.
Please see this application note for more on the calculation of the pull-up resistor.
https://www.ti.com/lit/an/slva689/slva689.pdf
Please let me know if you have any more questions on this application.
Regards,
Tyler
