You don't need to spend a fortune to see what a transmission electron microscope can show you.
In a modest workshop somewhere between curiosity and necessity, a maker known as ProjectsInFlight accomplished what commercial pricing had long made unthinkable for most: converting a scanning electron microscope into a transmission electron microscope using hand-machined aluminum and iterative ingenuity. The distinction between these two instruments is not merely technical — one reads surfaces, the other reads through, and that difference is the difference between shadow and substance. By solving a 14-millimeter clearance problem with patience and a lathe, this project quietly challenges the assumption that advanced scientific vision belongs only to those who can afford it.
- Commercial SEM-to-TEM adapters exist but carry price tags that effectively lock out independent labs and hobbyists, making the barrier to transmission imaging less scientific than financial.
- A razor-thin 14mm clearance inside the vacuum chamber turned a conceptually simple conversion into a precise engineering puzzle with almost no margin for error.
- Secondary electrons kept contaminating the signal — invisible interference that degraded every image until an iterative shield redesign finally cut off their escape routes.
- Gold nanoparticles confirmed the concept worked, and a mosquito wing imaged with striking structural clarity confirmed it worked beautifully.
- Biological cell imaging remains the horizon — demanding sample preparation techniques the current setup cannot yet handle, keeping the project alive and unfinished.
A scanning electron microscope reads surfaces; a transmission electron microscope reads through — shooting electrons straight past the sample the way X-rays pass through tissue. The difference in method produces a difference in what becomes visible. For years, converting one into the other meant buying an expensive commercial adapter. ProjectsInFlight decided the price was unjustifiable and built one instead.
The central obstacle was physical: barely 14 millimeters of clearance inside the SEM's vacuum chamber. The standard sample holder consumed most of that space, so a thinner aluminum plate was machined to replace it, freeing room for the redirecting mirror and the shield meant to block unwanted secondary electrons. A clever mechanical hack allowed the mirror angle to be adjusted without breaking vacuum, enabling tuning between tests rather than full teardowns.
First results with gold nanoparticles were promising but flawed — secondary electrons were leaking around the shield and corrupting the signal. The shield was made taller. The leakage stopped. Image quality improved sharply. A mosquito wing, contributed by nature and prepared by hand, then revealed its intricate geometry with a clarity that justified every iteration.
What remains is harder: biological cells demand specialized preparation techniques the current setup cannot yet provide. But the larger point has already landed — transmission electron imaging does not require institutional budgets. It requires a machine shop, careful thinking, and the willingness to fail inside a vacuum until something finally works.
A scanning electron microscope and a transmission electron microscope look similar enough—both fire electrons at samples, both live in vacuum chambers, both produce images that reveal the invisible. But they work in fundamentally different ways. A SEM bounces electrons off a sample's surface and catches the ones that scatter back. A TEM does something closer to what an X-ray does: it shoots electrons straight through the sample and reads what makes it through on the far side. The difference in principle produces a difference in what you can see.
For years, if you wanted to turn a SEM into a TEM, you bought an adapter. These devices exist. They work. They also cost enough money that most labs and hobbyists never seriously consider them. A maker who goes by ProjectsInFlight recently decided the price was absurd enough to justify building one from scratch.
The conversion itself is not new. Scientists have been doing STEM imaging—that's scanning transmission electron microscopy—with simple reflecting adapters for decades. What's new is the realization that the adapter doesn't need to be a precision instrument manufactured by a company with a clean room. It needs to be a piece of metal, carefully shaped, that fits inside an impossibly tight space and does one job well.
The tight space was the real problem. Inside the SEM's vacuum chamber, between the sample holder and the delicate components that surround it, there was just under 14 millimeters of clearance. Fourteen millimeters. That's the width of a pencil eraser. The standard sample holder was bulky enough that it consumed most of that space. ProjectsInFlight machined a thinner aluminum plate to replace it, which freed up room for the mirror that would redirect the primary electrons and the shield that would block the secondary ones—the electrons bouncing off the surface that you don't want to measure.
The machining work was straightforward: lathe, mill, tap threads. The real work was iteration. A hack was rigged up so the mirror angle could be adjusted even after the chamber was sealed and evacuated, which meant the adapter could be tuned without breaking vacuum and starting over. The first test sample was gold nanoparticles. The images came through, but the secondary electron shield was leaking. Electrons were sneaking around it, contaminating the signal.
The shield got taller. The leakage stopped. The images got dramatically better. Then a dead mosquito—donated by nature, prepared by the maker—had its wings imaged. The TEM pictures showed the intricate geometry of those wings with a clarity that vindicated the whole project. You could see the structure. You could see why they work.
What comes next is harder. Biological cells are the real prize for TEM imaging, but they demand preparation techniques that go beyond what the current setup can handle. That's the next frontier. For now, the point has been made: you don't need to spend a fortune to see what a transmission electron microscope can show you. You need patience, a machine shop, and the willingness to fail a few times in a vacuum.
Notable Quotes
The secondary electron shield was leaking—electrons were sneaking around it, contaminating the signal.— ProjectsInFlight's iterative testing process
The Hearth Conversation Another angle on the story
Why does it matter that someone built this adapter instead of buying one?
Because the commercial versions cost tens of thousands of dollars. If you're a hobbyist or a small lab, that's not a purchase—that's a barrier to entry that doesn't exist anymore.
But the conversion itself isn't new. Scientists have been doing this for decades.
Right, but they were doing it in institutions with budgets. The insight here is that the adapter is simple enough that a skilled maker can build it. That changes who gets to use the technology.
What was the hardest part?
Fitting everything into 14 millimeters. That's not a lot of space. You can't be sloppy. Every component has to be precisely machined and positioned, or the whole thing fails.
And the secondary electron shield—why did that need to be redesigned?
Because electrons don't travel in straight lines the way you might expect. They scatter, they bounce around. The first shield design didn't account for that. It had gaps. You had to make it taller, extend it further, to actually block what you wanted to block.
What does a TEM image show that a SEM image doesn't?
Different information. A SEM shows you surface texture and topography. A TEM shows you internal structure—what's happening inside the material. For something like a mosquito wing, you get to see the delicate architecture that makes it work.
What's the next step?
Biological cells. That's where the real power of TEM becomes obvious. But cells need preparation—they need to be sliced thin, stained, mounted carefully. That's a whole different skill set.