It's long been an established "art" to condense nitrogen to a liquid normally via high pressure. The overall wear and tear and repair on these systems though is horrible though, ten's of bars of pressure is certainly not helping.

The barrier in low temperature refrigeration hits in about -150C. An autocascading system mowing down with R14 (CF4) and argon saturation can achieve -150C but there's very little lower temperature refrigerants past that to help out.
I've been wanting to build a nitrogen drip for a long time, and being in the autocascade field it seems there is two options to try. Mainly looking for discussion on viability here.

Of course past the subcooler another phase separator could be added, and the subcooled R14/Argon could be seperated from additional Argon charge in its gaseous phase after maximum saturation has occured. Thus a further liquid stream of argon could be attempted to be produced, resulting in temperatures in the -160 to -170C range. However with nitrogen located at -196C, there is still a rather large leap.

Assuming the system is to be made to attempt a sub 250 psi high side pressure against a 1 to 2 bar low side (15 to 30 psi), then even using another stage past Argon of Nitrogen will of course result in a temperature warmer then required.

So either nitrogen could be made under slight pressure, the final evaporator being a tube in tube design featuring a hook up to a N2 tank with regulator, or even an air compressor could be hooked in line, with liquid air being justified as close enough to liquid N2.

The other option would be to proceed from R14 to R50, introducing a flammable refrigerant is not so much a problem, as methanes density is fairly low and requires a very large amount of the stream to result in any real load capacity. It could very well take more then the possible argon stream WITH subcooler.

Any ideas?

A second compressor could be added in cascade style, condensing nitrogen and evaporating it in vacuum resulting in the -200C goal or lower, but ideally then the compressor would be very very low, would have to be screened of moisture (now you need ANOTHER system for vapor trapping) if it were to take from the open atmosphere, and your first autocascade would truly have to be able to pump a hot nitrogen stream from 40C to -170~ Centigrade. Which obviously is a large load in most situations.