SN74LVC07A: as a levelshift

Hi Kevin,
The short answer is YES - SN74LVC07A can translate from 1.8V to 5V at 5MHz. I would recommend using a 330 ohm pull-up resistor and trying to limit parasitic capacitance as much as possible on the output.

The SN74LVC07A is an open-drain device. Your up-translation limit is likely not going to be caused by the device itself (that can operate beyond 100 MHz), but by the chosen pull-up resistance and the load capacitance (ie parasitic capacitance on the output trace).

For example, if you have a 10kohm pull-up and a 30pF load, The rising edges of your output will take ~4*10k*30p = 1.2us to reach 'high' levels. Typically you want your rising edge to be no more than 1/3 of your total pulse shape, so 3*1.2us = 3.6us, which translates to a bit rate of ~278 Kbps.

By reducing the pull-up resistance to 3.3kohm, this can be tripled to 833kbps. Reducing the resistance by 1/3 again will triple the result to ~2.5Mbps.

In other words - you will need to design the output to support the speed you need. At 330 ohms and 15pF (relatively common values), the device can output at 17+ Mbps, and if you relax some of the requirements and get the output capacitance really low (3 to 5pF), you can likely get to 100+ MHz.

Hi Kevin,
Short answer: it's best to stick with the recommended maximum in the datasheet.

This device can sink up to 50 mA of continuous current (see the absolute maximum ratings table).

The value you have highlighted is the recommended IOL maximum. If you look in the Electrical Characteristics table, you can find that the VOL at 4mA and 1.65V supply is 0.45V, which means the output has an impedance of 0.45V/0.004A ~= 112.5ohms at this low operating voltage.

As long as the supply is constant, the output impedance will remain relatively constant as well, so we can calculate the VOL at 15.2mA of current to be: 0.0152*112.5 = 1.71 V. Note that the large change in VOL will probably cause issues with whatever this is driving. To remain below 0.45V, keep the current below 4mA.