Scope output changes shape when changing caps, and Issues with convergence

[bkim334]

In my circuit attached, I'm looking to get an output where the width is around 30-50 ns. When I run the program with 10nF caps, I get a width of around 100us. Trying to make this more narrow, I lowered the cap to 1nF, but when I run it, the shape of the output is completely different. I was wondering if anyone knew if it could be the amps that I'm using or what? I tried to change them to high speed ones, but it didn't output anything. 

Another issue I'm having is that when I set the max time step the settings for Interactive Simulation, if I set it to anything lower than 1e-6, it will throw a simulation error and say:

Transient time point calculation did not converge
Simulation canceled

I feel like there is something small that I'm messing up and any help would be appreciated!

Edit: I didn't realize that the Interactive Analysis was just a transient analysis, which would be messed up by the step voltage I have in the very begging because it is time based. Does anyone have any clue which kind of analysis I would use to look at the output at VoutSum in response to the step function? Thanks!

[niarena]

Hi bkim 

I recommend you to make one post per question so you get better responses.

About the caps, they may be having a frequency response to that point that filters the signal, I would run the simulation with a cap that works and incrementally or decrementally change the value to see where is it that is changing, my intuition tells me that the signal is been filtered.

About the step response, the transient analysis may be work, just need to configure it, please read the section 2 of this tutorial, they used it for a step response of the circuit http://www.ni.com/tutorial/12774/en/

[bkim334]

Thanks for the response! When I run it with a 1nF cap, it works, but once I lower the value to 100pF, the output changes.  Any idea about what I could change to fix that issue?

[niarena]

It could be the circuit's response, maybe the input/output impedances are getting mixed up, a buffer or coupler could help if that is the case, but I would double check if that is actually expected by the circuit.