Our design made use of the DS92LV2412 datasheet revised on OCT 2012. In some recent testing, it was identified that the notes regarding the ID[x] pin have changed. Our design current has this pin grounded, based on the note saying:
"The third pin is the ID[X] pin. This pin sets one of five possible device addresses. Three different connections are possible. The pin may be tied to ground. The pin may be pulled to VDD (1.8V, NOT VDDIO)) with a 10 kΩ resistor. Or a 10 kΩ pull up resistor (to VDD1.8V, NOT VDDIO)) and a pull down resistor of the recommended value to set other three possible addresses may be used. See Table 11 for the Ser and Table 12 for the Des. "
The new revision says:
"The third pin is the ID[X] pin. This pin sets one of four possible device addresses. As shown in Figure 33 , Table 11 and Table 12 different Resistor values could be used to set different SMBUS addresses."
OLD REV:
Alludes to 5 addresses, provides 4 in the tables referenced. Does this mean it was an error to say the pin could be grounded?
We have tried all the provided addresses without successfully being able to communicate to the deserializer, perhaps there is a 5th address that will work when that pin is grounded?
We current do not NEED communication; however, my concern is TI realized grounding that pin causes a degradation of performance, reliability, etc so they revised the datasheet accordingly. That's a pretty big oops and we need an answer immediately. If this is a non-issue, but we just can't communicate, we're OK. If it slowly degrades the chip, we'll need rework.
