Building The World’s Smallest Jet Turbine By Hand

There are very few machines as complex to build as a turbojet engine. The turbine blades on a commercial airliner are grown from a single crystal of metal. The engineering tolerances are crazy, and everything spins really, really fast. All of these problems aren’t a concern for [Igor], who’s building what will probably end up being the world’s smallest turbojet engine. He’s doing it in his home shop, and a lot of the work is being done by hand. We don’t know the Russian translation for ‘hold my beer’, but [Igor] certainly does.

The design of this turbojet — as far as we can tell — is a centrifugal flow turbine, or something that’s not terribly different than the projects we’ve seen that turn the turbocharger from a diesel engine into a jet. The innovation here is using a lathe to machine the compressor stages by mounting an end mill to the headstock and the compressor blank on the cross slide, in a rotary table. It’s weird, but you really can’t argue with something that looks like it’ll work.

[Igor] has made a name for himself by creating some crazy contraptions. The most impressive, by far, is a gigantic remote controlled plane, powered by a handmade jet engine. This is an enormous fiberglass plane with a homebrew engine that spins at 90,000 RPM and doesn’t fly apart. That’s impressive by any measure.

[Igor] is posting a lot of his build process on YouTube and Instagram, including heat treating the compressor stages with a blowtorch. This is an amazing project, and even if this tiny turbine will be able to self-sustain, that’s an amazing accomplishment. You can check out a few more videos from [Igor] below.

39 thoughts on “Building The World’s Smallest Jet Turbine By Hand

      1. They exist, but the majority of manufacturers don’t use them due to their high manufacturing cost and fragility. Yield strength at operating temperature is an important part of material selection in this application.

        1. No, actually, they are common practice now under most of the major manufacturers for power plants, GE and Rolls Royce both use single crystal blades in a lot of their power plants precisely because they are strong, lighters, and generally easier to produce in bulk.

    1. I too have heard of the single crystal story. I’m reading the book “Exactly!” by Simon Winchester.

      But the “no they are not” is easy to explain. The blades are certainly not grown FROM a single crystal. The described technique is to very slowly cool a liquid so that everything that solidifies can do so into that single crystal. Start at one end and very slowly cool things towards the other end.

      1. I have no idea what point you are trying to make. The write-up said that he was milling on his lathe. And, of course, that is possible. Many lathes (especially historically) came with milling attachments, some very elaborate:
        But the machine he is using is a horizontal milling machine. In fact, I have one myself, a 4-axis CNC horizontal milling machine in fact:
        Many horizontal milling machines have an optional vertical head. Mine has and I suspect his has too, but for the job at hand, the horizontal spindle was probably more convenient.
        And, to close the loop, it is entirely possible to turn on a horizontal or vertical mill.
        This is me turning a workpiece outside the envelope of my lathe on my mill:
        And this is ToT using a 4th axis on his router as a lathe:

        So, vegetable nomenclature aside, whilst you can turn on a mill and mill on a lathe, the chap in the video was milling on a mill.

        1. Sorry, I was just joking. Was trying to say that you’re right, they’re definitely different even though they’re sometimes mistaken. It didn’t really come off though. I do appreciate the info!

  1. Im not sure this is the smallest, people have been making turbines about this size for years:

    They are not that terribly complex, some of the early ones had their blades welded to the exhaust turbine wheel. Now they are all cast from high temp alloys. And if you want to fly at public meets I think they require proof the turbine wheels were X-rayed for flaws.

  2. I expect there are two thresholds of precision here:

    1) the “it’ll work” threshold.
    2) the high efficiency (commercial) or high performance (military) threshold.

    Like in most things if you don’t need to squeeze every last drop of performance or efficiency out of a system you can give yourself a bit more slack.

  3. “The design of this turbojet — as far as we can tell — is a centrifugal flow turbine”

    No, the compressor is centrifugal. The turbine part is axial flow. Pure axial flow micro turbines have proved to be notoriously difficult to get to work. Most of the available small turbines, homebrew or commercially built, use a single stage axial turbine and single stage centrifugal compressor.

    All this guy is doing different is making his turbine multi stage. I expect the rough finish on his mostly hand-hewn blades to be a problem.

    1. I’m still totally floored that some hand-cut piece of scrap metal that was heat-treated with a friggin blowtorch can survive 90,000 RPM no problem and still manage to perform okay with useful TWR.

      1,500 turns per second. Handmade by some Russian guy in a garage. That’s nutso-bismol, and frankly rather inspiring! I wanna try… Maybe behind a sheet or five of perspex.

        1. You are allowed to use “centrifugal” here:
          the limiting factor for any material is the peripheral speed. (basically rpm x diameter, or omega . r , it’s all the same except numerically) Geometry does play a part, of course. The solid wheels of land-speed record cars are at the point where the geometry matters. But for a simple disc (and a turbine approximates that, I think, as there is as much, or more, inside any one point on the perimeter)
          There is a lot of detail here (first Google hit that looked reasonable, and I have not read all of it) bit fundamentally as things get smaller they can spin faster.
          (As an aside, because of this, and only considering bursting stress, the optimal flywheel for energy storage is long and thin, because you gain max revs as you reduce diameter faster than you lose moment of inertia. In practice it’s not so clear cut because of the risk of the whirling pencil bending and spinning apart)
          Everyday car turbos are far bigger than this device and run twice as fast for hundreds of thousands of (insert local units), and they are not _that_ special metallurgically or manufacturingly, they can afford to fit them to cars.

  4. “The World’s Smallest Jet Turbine By Hand”
    Give me a break. Click Bate. Should be titled “The World’s Smallest fan blade By Hand”
    There is no way this can be called a Turbine

    1. Ahh. You’ve spotted a Benchoffism. Protip: read the name of the potser first. If its benchoff, just turn down the sound an color saturation a bit. And take everything readable on this screen with a large grain of salt anyway. This is the internet, remember? NOT the real world

    2. That moment when you’re trying to be a smarta$$ but you can’t speak russian nor have read the article completely.

      The guy is building a very small turbojet engine. The shown thing that you call a fan blade is going to be the turbine of said turbojet engine.

  5. Does anybody knows how to make dynamic balance for this kind of small parts and High RPM for this purpose ?
    as far as I know they are very expensive, but does anyone “DIY” ?
    and how you balance this ??

    1. Look on Youtube for “turbo balancing”. You can spin up the rotating parts with compressed air while measuring vibration with an accelerometer. By correlating rotor position to the peak of the accel curve you can determine how much needs grinding off, and where. Even production turbos in quotidian shopping diesels seem to have hand balancing marks. (Though I don’t see many actual series production parts in my job)

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.