In case anyone is interested in the 1S1 Sampling Unit, I uploaded some pictures of a risetime test:
/g/TekScopes/album?id=261760The time/cm is set to 0.1 nanoseconds/cm and the scope displays a risetime of approximately 300 picoseconds, which is actually faster than the recommended 330-350 picoseconds that the manual suggests. (I should probably do all the calibration steps and adjust snap-off current for 340 picoseconds.) Anyway, I think a ~300 picosecond risetime is pretty good for a 55+ year old vacuum tube oscilloscope.
I am driving the input with a USB-powered "fast risetime pulse generator" from Leo Bodnar Electronics () connected through an SMA to GR-874 adapter. The actual risetime is 29 picoseconds.
I was worried when I got the 1S1 from eBay that it would have lots of problems, e.g., maybe the tunnel diodes would have drifted and stopped working, but this unit performs perfectly, as far as I can tell. I replaced the EMT, EMC, and PTM capacitors but that's it. It takes a long time to warm up and when it is cold the trace is noisy and drifts vertically all over the place, but once everything warms up it is quite stable. I thought sampling would compromise the quality of the trace but it really doesn't. Sure, the trace is made up of dots, but Tektronix made such good CRTs, those dots are finer than the pixels of a modern LCD screen (or so it seems to me).
