What I currently do is, apparently, fix broken scopes. I've bootstrapped myself up to a point were I have enough reliable equipment that I can triangulate on almost any measurement (with the exception of observing bandwidth rolloff, but that will change in a couple weeks when my TinaSA arrives). I'm going through the calibration process on my 475A using cobbled together equipment. I have a cheap old signal generator, and I calibrate it's output amplitude using a Tek DMM916 and a 2236. The signal generator isn't really good above about 1 MHz, so that's a significant limitation for the 475A calibration process, but I think I'm doing pretty well emulating what the PG506 and SG502 would do at lower frequencies. I'm also working on a DIY calibration source that will produce square waves at several frequencies with amplitudes ranging from 5 mV to 50 V, which would make some of the work I've been doing with with the signal generator and the 2236 a little easier.
Oddly, I feel that the multiple instruments give me greater confidence in what any one of them tells me. I've got three relatively nice DMMs that I've been using the make comparative measurements, as well as a programmable power supply, and the cheap signal generator. Combine those with some TCXOs and I feel as if I can actually make some pretty good measurements of voltage, resistance, and frequency. From there I should be able to bootstrap myself into at least a couple decimal places of accuracy. For the kind of things I'm doing, I'm not sure I care about anything more than that.
What I want to be doing in the short term is investigating and documenting the display protocol for an old laptop that has a gas plasma display, so that I can then replace the innards of the laptop with a Raspberry Pi. Also, I'm working on building a 6809 microcomputer, and a homemade minicomputer using 74181 ALU slices. I've also got a couple other old pieces of equipment of my father's and grandfather's that I'd like to spend some time fixing up.
It's all just hobby stuff that has interested me ever since college, and the pandemic has given me enough extra hours in the week to actually pursue it. I suspect that I don't need much more than a 200 MHz scope to do most of this work, though I'm sure a logic analyzer would be a nice addition. At some point I expect to buy a modern Chinese DSO (probably a Siglent 1204xe), but the old (70s and 80s era) analog scopes are what really interest me, both because of my history with them, and because it feels "closer to the metal" than a DSO. I know too much about the limits of discretely sampled signals to really trust a DSO, and maybe not enough about analog amplifiers to know where my trust in analog scopes should stop.
-- Jeff Dutky