Hi, I would like to continue the thread raised last Friday about the Caravel or Caravel harness selection in a case of a mixed signal design. In our case we have a few digital input/outputs but also we would like to assign some analog ones in order to characterize some of photo diodes and our OTA (see the block diagram attached). We have some experience with caravel so we would like to use it again however there are some drawbacks related to the parasitic capacitance of the IO itself. are there more details we should consider making our decisions at this level ?

What are you trying to measure? If it's the DC characteristic, then the caps don't matter. Why are you using 2 TGATE switches (as opposed to one on top or bottom, or just an NMOS at the bottom)? In any case, to make useful suggestions, we'd have to know more abut your exact goals.

Hi @Boris Murmann thank You for replay! Surely we would like to measure I(V) characteristics using light as a parameter. Also measuring the step response would be great however here the IO parasitic could potentially affect the results. As for 2 TGATE switches I use two thinking about a test circuit that has much more diodes. I am not sure here but I think that isolating the diode structures by 2 gates should introduce less parasitics or am I wrong here ?

To reason about the impact of parasitics, it would make sense to draw out the whole measurement setup. It's really hard to tell from this cartoon.

At a first glance, we do not have facilities to characterize devices at IC level, so I suppose, that the high-frequency issues we will encounter due to the laboratory instruments (like scope probe capacitance) we have. However, we are in contact with some centers that do have such capabilities, so we would like to know the best design practices in order to balance our decisions. I think that doing the first iteration, the DC of the photo diode would be a good idea and if we can observe the dynamic behavior of those devices would be even better, but it seems way more difficult.