You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This is what happens to the FX2-side I2C bus after trying to address a pull TCA9534 while both DACs are enabled and supply IO power:
The TCA9534-side I2C bus is similar but oscillates at 1.8 MHz instead.
This happens because we connected two TCA9517s with their B-sides on the same bus. (We had to, since the A-side is the only one that goes below 2.7 V). The datasheet says to not do it, but doesn't mention that they will oscillate wildly...
As far as I can tell, TCA9509 (datasheet) is a pin and package compatible replacement for TCA9517 (datasheet) that should not have this problem. It is possible to use it on existing revC0 boards by swapping the shifter and changing R36-R39 to DNP.
TCA9509 datasheet explicitly calls out this problem (or lack thereof):
The output pull-down on the B-side drives a hard LOW and the input level is set at 0.3 of SMBus
or I2C-bus voltage level which enables B side to connect to any other I2C-bus devices or buffer.
The text was updated successfully, but these errors were encountered:
Fixed with the new expanders, just need to update the design and validate. Chips are on the way. ECNs page has been updated with a workaround for the current units (just removing the shifter, turns out that works).
This is what happens to the FX2-side I2C bus after trying to address a pull TCA9534 while both DACs are enabled and supply IO power:

The TCA9534-side I2C bus is similar but oscillates at 1.8 MHz instead.
This happens because we connected two TCA9517s with their B-sides on the same bus. (We had to, since the A-side is the only one that goes below 2.7 V). The datasheet says to not do it, but doesn't mention that they will oscillate wildly...
As far as I can tell, TCA9509 (datasheet) is a pin and package compatible replacement for TCA9517 (datasheet) that should not have this problem. It is possible to use it on existing revC0 boards by swapping the shifter and changing R36-R39 to DNP.
TCA9509 datasheet explicitly calls out this problem (or lack thereof):
The text was updated successfully, but these errors were encountered: