TDC Functionality?

[VA7BKK]

The 2 GHz spec caught my attention and I was wondering whether this project could be adapted to measure time intervals with sub-nanosecond precision. I don't necessarily need a crazy bit depth, just time stamps of precise and accurate rise and fall times that may be delayed by up to 10 seconds or so from the start signal to the stop signal. So it would need a clock that doesn't drift over time such as a TCXO.

Also, any option to reduce the price at the cost of more limited functionality could be advantageous for wider adoption.

[drandyhaas]

Sure, I guess you could. The effective bits of the adc measure the rising (or falling) edge very accurately, at about 1-10ps resolution, depending on the shape of the edge, since it's assumed to be either linear (no upsampling) or pure sine (with upsampling). If you knew what the shape of the edge was going to be, and it wasn't one of those two, you could use correct the time for that shape in software instead.
A txco still has 1e-10 drift, so 1 ns per 10s, but an ocxo could get you down to 10 ps in 10 s. You would feed this in via the external clock sma in from an external 10 MHz source.
Overall, this device is not optimized for cost for TDC. But it could be appropriate if you also needed detailed recording of the analog signal shape surrounding each edge, up to 6 us before and after the edge.