Jose said:
"Not many people seem to have made DIY NLTL, let alone samplers based on it."
This is probably true, because NLTLs usually have lots of stages, so can get very big and complicated before you get remarkable compression. For simplicity and effectiveness, a single SRD/varactor of the right type, in the right circuit environment, is pretty hard to beat for straight up impulse generation or frequency multiplication. A whole bunch of them cascaded can form a NLTL, with its edge-enhancement and wide BW. As always, it depends on the particulars of the application.
I don't think there's a need for a "modern" replacement for the SRD/varactor. It's a common RF part - not at all an obsolete technology. Nowadays there are alternatives for many applications, by using fast active devices capable of appropriate edge speeds.
Ed
