I have used the TXB0108 translator in several designs without problems--until my latest project. It is translating between 5V and 3.3 V. There is a 5V, 40 MHz crystal oscillator which is translated down to 3.3 V to drive the clock pin of an Altera EPM240T CPLD. Most of the time this works fine. However, sometimes the sytem voltage can drop from 5V to about 2V, and the 3.3V drops to zero. (The system is idle at this time.) It turns out that the oscillator continues to run, putting out a 40 MHz clock with about 2 Vpp swing. When the 5V comes back up (bringing the 3.3V with it), sometimes the output of the TXB0108 goes to an intermediate voltage level (about 1.5V) and without the clock on it, even though the clock is present on the 5V side! It can stay stuck in this state for an indefinite period of time, though it can be easily made to snap out of it by tapping the lead with a scope probe, or even touching it with a finger. Some chips seem to be more susceptible to this issue than others. I tried some resistive loading on the 3.3V side, and that seems to help, although it is not a guaranteed solution. Any ideas?
-
Ask a related question
What is a related question?A related question is a question created from another question. When the related question is created, it will be automatically linked to the original question.
