I Saved an Electron Microscope from the Trash [video]
Comments
mmastrac
dclowd9901
It's funny too how they exist in the same kinds of spaces they used to. Chat rooms, forums and other areas no one ever talks about anymore with the Twitters, TikToks and so forth dominating most social media attention.
ThrowawayTestr
In this case, the author found a discord server where he was lucky to find two other people repairing the same scope.
dclowd9901
Yep, I intentionally mentioned chat for this reason, but I find that open live chat channels wherever they exist tend to be awesome founts of social interaction and vibrant, rich interest-cultivating.
roger_
Wonder if there'll be a day when a hobbyist can buy a < $1k SEM?
Other equipment like spectrum analyzers, VNAs, scopes, etc. are now way more accessible, but they seem way less complex.
sahmeepee
You probably need a sputter coating machine too for most applications, and, at least in some cases, tricky consumables like liquid nitrogen.
I'm with you in wishing that they were more accessible, though. So many people want to do skydives, bungee jumps etc. that have never experienced the thrill of exploring the anatomy of tiny insects in incredible detail. They don't know what they're missing! That's before you even get onto cool stuff like WDS/WDX which I remember thinking was like science fiction when I first saw it.
nullc
> tricky consumables like liquid nitrogen
LN2 costs like $1/liter-- it's cheap.
Aurornis
The LN2 itself isn’t the problem.
It’s the appropriate container, handling equipment, PPE, and safety technique.
Unfortunately, as some of these things get cheaper you start attracting a different level of hobbyist who are less interested in learning proper handling technique and safety precautions. Not all of them, but enough that it becomes a problem.
I’ve seen this play out across a couple of my hobbies in the past decade. When things were hard to access, the people who put in the work to get there had an appreciation for doing it properly and being respectful of the environment and community. Once it becomes cheap enough you get a lot of weekend warriors trying to run through the process as fast as possible while seeing how many corners they can cut. Things get ugly.
nullc
Fortunately dewers have become a lot less expensive too. I'm not sure that we really need to be that concerned with the safety of LN2 while cars and ladders still exist.
For a SEM where LN2 is only going to get used in a cold trap or for flash freezing a sample, I think the HV risk dwarfs the LN2. ... or maybe the risk of a vacuum mishap causing a fire in the diffusion pump.
(or, if it's using a TMP instead of a diffusion pump-- a vacuum mishap causing a rapid unplanned disassembly)
II2II
I was mildly surprised when my university supplies office offered to sell some liquid nitrogen to me, no questions asked. Which was disconcerting for few reasons: I asked for dry ice, I lacked a dewar, and I would have to transport it on a subway. All they knew was that I was collecting supplies for a science demo. Granted, a demo using liquid nitrogen would be much more fun. It simply wasn't an option on short notice.
mikewarot
If you want to image things that small without the need for high vacuum systems, etc. you should consider an Atomic Force Microscope[1] which uses physical contact and a piezoelectric scanning arrangement, or Scanning Tunneling Microscopes[2] which send a very small tunneling current across the gap from a needle to the target.
Those are more likely to be reached by the DIY crowd.
[1] https://en.wikipedia.org/wiki/Atomic_force_microscopy
[2] https://en.wikipedia.org/wiki/Scanning_tunneling_microscope
georgeburdell
I don’t think an SEM is that much more complex, it’s just necessarily big due to the lack of miniaturization in the supporting components (lens system, transformer, vacuum chamber). When I was a grad student, there was an old SEM from some defunct company that was being maintained by undergrad assistants (with the equivalent of duct tape and bubblegum)
II2II
I'm not going to say it is impossible, but the thing to keep in mind is electronics test equipment benefited from incredible improvements in and the mass adoption of most of the components that go into them. While SEMs can benefit from that, I don't foresee it happening for some of the vital components.
richk449
Is there an effort to get this underway? I’m no expert, but I would guess that the two expensive parts are the electron source and the vacuum pump. UHV chambers seem solvable, and detectors are probably cheaply obtainable. High voltage power supplies can probably also be cheap.
ChrisMarshallNY
But they need to be calibrated, which is a real bitch.
I used to work for a defense contractor, with millions of dollars' worth of that kind of gear, and keeping it calibrated was an expensive pain.
wkat4242
Often that's not necessary for hobby use. The calibration is more about certifying the stuff that you measure with it.
For example if the magnification on the scope says 1000x but is actually 1050x, the item you manufacture and measure would have wrong dimensions that might cause issues during its use. This is why you calibrate them, so you know that something that's meant to be 1000 micrometer is actually 1000 (+ or - a certain tolerance that it's certified to). This is also needed for the certification of the products you manufacture.
However if you're just a hobbyist and want pretty pictures or you don't care too much about the accuracy, then it's not necessary.
I worked in a place with an electronics test lab and even stuff like multimeters would need to be certified yearly. This cost a lot (a significant fraction of the purchase price!) so often it would only need to be done for a few years before the meters would get replaced entirely. They'd be useless for the lab then, however they were still fully functional for home use as my need for accuracy isn't that high. Normally I use cheap chinese meters but having a real quality fluke is a big step up even if it's not calibrated officially. Unfortunately most of the stuff was scooped up by people a bit higher in the food chain than me :)
ChrisMarshallNY
I suspect some of the newer, "pure digital" stuff barely needs any actual work, and most "calibrations" are really just a verification/expensive sticker.
Back then (mid-1980s), even the digital stuff had a lot of analog circuitry, and some of these huge high-bandwidth spectrum analyzers (for example) would have hundreds of micro-pots. Calibrations could take days.
thrill
Once hacker-level-hobbyists get involved they'll probably figure out how to calibrate it with a Raspberry Pi and a 3D printer.
MisterTea
The high vacuum pumps to get to the required vacuum levels are prohibitively expensive. Even a small turbo pumping setup is going to run you north of $10k USD. That is one area I want to see come down in price - vacuum generation.
breck
[dead]
My favourite part of these videos of people rescuing obscure tech is when they talk about the community of other people also rescuing the same obscure tech and team up with them. The internet might be a bit of a cesspool now but these little nuggets that remind me of the early promise of a connected world warm my heart.