Metrology
No, not meteorology, metrology. Meteorology is the study of weather (oddly, not meteors), metrology is the study of measurement.I don't know why but I love measuring things and the science of measurement.Why? Honestly I couldn't tell you.I do however.Some things are practical. My caliper that can measure down to 0.0005" (accurate or precise? I'm not sure, it's an electric one, but it seems to hold its own). I have my micrometer that will nicely go down to 0.00001" and since it's all mechanical I trust it a bit more. My scale can measure something down to 0.05 grains or 1/140000 of a pound. These all seem esoteric, but they are all used for reloading ammunition. These are practical and applicable to something I do.Speaking of scales I recently upgraded the "me" scale for the bathroom.Then there are electrical things. I have a cheap little multimeter that measures a variety of things. Dave Jones from EEV Blog would lump this into the "One-Hung-Low" brand of meters. I really want to get into some more electronic stuff and such so I think I'll be upgrading that to a good Fluke at some point. ;-)This goes down the rabbit hole of calibration.So, how do you cal a multimeter? A good calibrator like the Fluke 5720A. (Ok, that's freaking overkill, but if you're going to go down that hole, go with gusto!) A reasonable $60K or so later you have yourself a source for precision calibration of your stuff.But how do you cal that?No prob. In this particular case all you need are a 10V, 1 Ohm and 10K Ohm references. The rest of the measures are derived from those.How do you cal those?The resistors are relatively easy... They are passive. You can get them at great cost and have them measured by exponentially more expensive gear. (As long as you know the actual you're good to go)The voltage on the other hand is harder.Well, you can get a 10V standard from Fluke as well... The Fluke 734A would do nicely. This is a beast that you're not allowed to even turn off. You calibrate it to 10V and leave it plugged in or running off it's battery.How do you cal that?You get another one that's already cal-ed.?That one?!You build yourself a Josephson Junction voltage reference. Given a known frequency you can get a known voltage.How do you get a known frequency?Well, that's governed by time.How do you measure that?The vibration of a Cesium atom.
- = -
Given a Cesium atom, you can cal a multimeter. Well, with a crapload of stuff in between.
- = -
I wouldn't mind working at NIST. I think I'd have fun. (I have a strange notion of fun in case you didn't already notice)