Multi-Stage Ion Thruster Holds Exciting Promise

Anyone who’s looked into high-voltage experiments is likely familiar with ion lifters — spindly contraptions made of wire and aluminum foil that are able to float above the workbench on a column of ionized air. It’s an impressive trick that’s been around since the 1950s, but the concept has yet to show any practical application as the thrust generated isn’t nearly enough to lift a more substantial vehicle.

It’s a bit early to suggest that [Jay Bowles] of Plasma Channel has finally found the solution to this fundamental shortcoming of electrostatic propulsion, but his recently completed multi-stage ion thruster certainly represents something of a generational leap for the technology. By combining multiple pairs of electrodes and experimentally determining the optimal values for their spacing and operational voltage, he’s been able to achieve a sustained exhaust velocity of 2.3 meters per second.

Dry ice was used to visualize airflow through the thruster.

While most ion thrusters are lucky to get a piece of paper fluttering for their trouble, [Jay] demonstrates his creation blowing out candles at a distance of a meter or more. But perhaps the most impressive quality of this build is the sound — unlike most of the experimental ion thrusters we’ve seen, the air flowing through this contraption actually makes an audible roaring sound. When the 45 kilovolt supply voltage kicks in it sounds like a hair drier, except here there’s no moving parts involved.

In addition to providing graphs that show how air velocity was impacted by input voltage and the number and spacing of the electrode pairs, [Jay] also pops the thruster on a scale to show that there is indeed a measurable thrust being produced. Admittedly the 22 grams of thrust being generated isn’t much compared to the contraption’s own mass of 490 grams, but in the world of electrostatic propulsion, those are pretty impressive numbers.

[Jay] says he has some improvements in mind that he believes will significantly improve the device’s performance as he works towards his ultimate goal of actually flying an ion-propelled aircraft. We saw MIT do it back in 2018, and it would be great to see an individual experimenter pull off a similar feat. Obviously, there’s still a long way to go before this thing takes to the skies, but if anyone can pull it off, it’s [Jay Bowles].

50 thoughts on “Multi-Stage Ion Thruster Holds Exciting Promise

  1. Nice prototype build.

    One obvious change would be to use aerofoil section negative electrodes in order to reduce turbulence on the outflow side (= drag). Likewise, the support frames could be profiled to reduce drag around the periphery of the thruster.

    Would having different voltages on the stages be more efficient? More stages?

    Could you reduce the weight (use thinner support frames, webwork frames, aluminium instead of brass tubes for example)

    Would using a magnetic coil to compress the outflow help?

    I look forward to seeing the next level of development.

    1. More than just using aerofoils as electrodes, I was thinking just the trailing tip of the aerofoil should be electrified. This way the electric field lines would also wrap around the aerofoil and aid in moving the air in the desired fashion.

  2. The round electrodes in the interior will tend to scatter air. Fins would be much better. Putting plastic wrap around the entire column should prevent “leakage” through the sides. If you can afford a small magnetic field, you could perhaps channel the airflow better.

      1. By adding a coil on the tail end to use as a a magnetic nosel you could make a more controlled output. Coil would have to be powered with a high frequency square wave to keep the inductive reactance up to sustain magnetic field.

  3. After logging in, I see that Alysson Rowan said much the same thing as I did, apologies for redundancy. I can only add one more thought. One could get really fancy and add an RF source to “bunch” the charged particles, basically build a variation on a Wakefield accelerator. I don’t know if it can actually be done but it sounds fun. Imagine relativistically accelerated air molecules. :-)

    1. Redundancy is good – it shows that we are on the same wavelength.

      As to an RF modulation … be still my beating heart.

      In this case, the separation between stages would need to be adjusted to get the pulses in phase across all of the stages.

      1. Wouldn’t putting the stages in parallel (vs series) get you more thrust?

        You get momentum (v) but pay kinetic energy (v^2).
        Same as a propeller, see also ‘ideal propeller theory’ (which solves for same velocity of outflow for all sections).

        1. Main difference with series config is a higher velocity. For low speed operation like hovering, parallel should be better. Conventional wisdom is that to maintain thrust during fast flight, you need the air to be going significantly faster than the craft.

          This is where old conventions break down – propellers need to vary their pitch in order to have a screw pitch faster than the incoming air. In this case maybe the ionic wind doesn’t care about the incoming velocity – any molecule that gets ionised will receive an additive electrostatic force, speed up further and encourage neighbours to speed up.

    2. With an RF source there is not so much controll if positive or negative ions are produced, but there is way, for example it is possible to shot electrons from a vacuum chamber into air through a very thin vacuum seal. The molecule s absorb them and they can be accelerated by outside high voltage electrodes homogeneous field. One problem with the asynchronous, inhomogeneous field is that you get secondary ionisation from some ions that are accelerated way to much, basically just heating up the air and the circuits.

  4. Would a microwave exciter introduced into and directed through a guided wave (to approach from the base of the center of flow of the direct central part of a chamber) have an acceleration effect ?

  5. Honestly surprised nobody tried stacking those things before?? hope his idea’s for improvements work, while i highly doubt we’ll ever get flying cars etc. at the least this could be useful for space

    1. Yes, stages have been done for EHD pumps and laboratory thruster research. There’s a host of other problems to deal with as well but it’s good to see progress and interest. For example, like the MIT project, the batteries and rectifier or DC to DC converters are too heavy to make flight realistically plausible. MIT flight batteries exhausted in 9 seconds, for example. becomes a power to weight problem that requires another approach and at the end of the day, a jet turbine to turn it.

      1. Long before the MIT ion propelled glider, I built and patented a VTOL version that lifts its power supply for almost 2 minutes at a time. There are about 40 flight footage videos various prototypes of it on YouTube and a website. They are very easy to find. The onboard DC to DC converters that I developed work just fine for this purpose. It is surprising that everyone doesn’t know about it. Adequate verification is provided.

  6. As well as the suggestions on aerodynamics above, you can trivially improve power to weight by just pulling out one of those guide rods its all built around – 3 of them is sufficient to keep everything in the same plane, and one of them is likely strong enough on its own. Plus the blue power distribution bars could easily be dispensed of as separate parts and made part of those red rod of structural frame its all built off. So 3 long braces rather than 6.

    Really cool looking and logically constructed prototype build though – colourful in a way that is actually kinda useful. At a glance you know what each part is and how it interacts with all the surrounding ones. Also Impressive performance, I’ve never seen any demo of the concepts that gets close to that airflow, though I’d love to know how the thrust per watt changes when adding coils – which I didn’t noticed in there when playing with turning off coils.

  7. Ion propulsion makes sense in space where you don’t need thrust anywhere near equaling the mass of the craft, and specific impulse matters more in terms of making use of precious finite reaction mass. But is there ever going to be a case for atmospheric flight where this will be preferable over a simple electric fan? Reaction mass is free and all around you, and the fan’s probably going to be more efficient in terms of energy usage. Far more useful thrust as well. And the total mechanism will weigh a lot less. I could be wrong.
    This thing still beats a fan any day of the week in cool factor, that’s for sure.

    1. Fan and this style of Ion propulsion are both reliant on the reaction mass surrounding them, so a comparison to space flight to me rings a bit hollow. Not that I disagree with your points.

      With how many variables there potentially are I can see there being many potential situations where this type of thrust can work out better than the mechanical fan – maybe not in conditions likely to occur on Earth, but between atmospheric density and composition there are bound to be scenario where fans don’t work as well.

      Perhaps even on Earth in the thin upper atmosphere? It would need rather giant fans to find enough air to work on, that would therefor be heavy where the ion thrust could in theory be built into the entire length/width of the wings and fuselage as its shape is far less predefined, giving you access to so much more ‘reaction mass’ to work on without having to supersize the driving propeller and surrounding airframe.

      Far from convinced such things will fly sensibly on Earth, but I don’t really have enough information to do anything but guess.

    2. You are correct that efficiency of electric to kinetic energy conversion is indeed a problem. There are many problems with this, however. I was surprised to find that mechanical electric fans for RC devices are actually not that efficient. 2-3% in terms of thruster to power ratios! Just check out the posted tech specs for any consumer grade fan motor.

      Now, EHD propulsion historically is also abysmal for efficiency, often 1-2%. However, there are interesting simulations and tests that have been that show if you combine methods, and get it moving, the faster it moves, the more efficient it becomes.

      However, this leads us to the next challenge. The power supply to get it moving in the first place. They’re heavy…..

  8. I wonder if he did some optimization (experimental or theoretical) off-camera when going from the uniform spacing to staggered. There are at least 5 parameters (distances between the sections) to tweak, but the video just shows “I picked this spacing and it’s better than the previous one, so I’m done”.

  9. 45kV and crocodile clips and standard wire. I hope this guy knows what he is doing… Can’t watch the video now, so i have to ask: How much power does this supply deliver? How deadly is it?

  10. One of the challenges is to provide complete neutralization of the exhaust before it hits the next stage, because any ions that don’t hit the negative grid are going to be attracted backwards and repelled by the next positive grid.

    Ion thrusters used in space inject a negative particle beam into the exhaust to neutralize it, so this doesn’t happen.

    1. The decellerating effect had me worried. Having slept on it –

      Perhaps pulsing the cells in a suitable phase would work – the already ionised air would merely carry on into the next cell.

      1. I had the same thought. You could even get a push/pull effect with very precisely timed (and maybe shaped) pulses at each stage. This would probably result in higher pressure in the airflow that would need to be contained like it is in a jet engine. At some point you should get a jet effect if you can get enough air mass moving. This is a very ool project!

  11. What’s the efficiency like? How much power does it require to produce this much thrust? Is there any limits on how many can be stacked?

    Does each stage benefit from having the same voltage? My guess is that you would want increasingly staggered stages – that way uncharged air can become charged and already charged air can become further charged.

    As others have mentioned, use thins instead. The same applied to PC fans, put less in the way to increase airflow.

    Regarding a housing to prevent leakage, I would probably use a charged foil of a similar voltage to guide the air towards the center (may not work with different voltage stages). Uncharged air may hit the outer wall, become charged and move away and reduce friction. You will likely want an uncharged lip on the front.

    I would probably also create a slight nozzle of sorts to increase air pressure. That way you get more air colliding and balancing charge (which might negate the need for different voltage stages).

    Very very cool work.

    1. By enclosing the whole array inside an open-wound solenoid (to minimise weight), you could not only have a magnetic nozzle, but allow reaction mass to enter the accelerator along its whole length.

  12. While this was a fun and interesting looking experiment, the amount of thrust the device produces compared to the weight of itself and its power supply, is abysmal. It is also unfortunate that Jay on the Plasma Channel mentions the 2018 MIT ion bungy launched glider, that leveled itself for a few seconds, but doesn’t mention the much earlier patented Self-Contained Ion Powered (propelled) Aircraft or Ion Propelled Vehicle. The earlier series of devices were patented for lifting themselves and their power supplies against Earth’s gravity for several minutes vertically. The thrust to weight ratio is approximately 1000 times higher than the heavy acrylic device above can produce. There are about 40 videos of the self-contained Ion Propelled Vehicles online, as well as a website.

    1. And as the holder of the patent of the “Self-Contained Ion Powered (propelled) Aircraft” I’m sure you’re entirely unbiased and only have [Jay]’s continued happiness at your heart.

      1. I have always thought very highly of Jay and his channel. I like the man and think many of the devices he has made constitute a really excellent job/build. It’s just that I know he must be aware of the ballpark thrust to weight ratio of other recent well known ion propelled devices. Considering the historical facts, it would have been more fair to mention the original patented one that can lift its power supply vertically for minutes at a time. I wish him and you well!

  13. Oh god, I can smell the ozone through the video.
    That aside, I would try setting it up in a way that the stages can be moved while in operation using a steeper or other computer manageable device. Then just setup a feedback to let it self tune.

  14. That is so cool. Just being able to blow ‘some’ air is interesting. In this case, it looks like fun. Who cares what others have already done. Sometimes it’s ‘fun’ re-invent the wheel for its own sake. Seriously.

    Who really cares about weight to thrust at this point. To me the idea is to find ways to optimize the creation of the ions, being able to contain/direct them, and accelerate them to possibly mach speeds and beyond — ie. max propulsive force for the energy put in. Solving the weight issue is really after you know what will and will not work… Ie. It’s the last step from all the experimental data you have gathered. At least that is how I look at the ‘fun’ problem :) .

    1. Sometimes, reinventing the wheel will get you a better engineered wheel.

      On occasion, reinventing the wheel will let you see how to replace the wheel with something better

      … or different.

    2. You are very much correct in that the amount of thrust produced per energy put in (grams of lift per watt), is more important than thrust per weight. They both really matter though for anything that flies. On my Ethan Krauss YouTube channel there are about 40 flight videos of ion thrusters with onboard power, that address both issues. They have thrust per watt and thrust to weight ratios that are many times higher than any other Ion thrusters. Sorry to repeat myself, but most people seem not to realize the details. Still, Jay’s project looks really great, and he’s discovered some interesting details, as well as a good way to experiment with them.

  15. Major De Seversky was featured in Popular Mechanics in 1964, featuring an ion powered flying machine. The article showed a photo of a primitive working model. I built a balsa and wire replica for my (1972) High School physic project. Powered by a VanDeGraf Generator , It levitated and crashed. A quick search shows pictures of Seversky’s design.

  16. I think there is lots to explore on this one.

    Aluminum honeycomb could be used for the accelerators, and that could produce a laminar output. Also, it would be interesting to see how clocking the ones in the video would change the output(generate a vortex?).

    Overall, it seems a fun setup, but I agree that a closed loop optimization with motorized parts should be in the final video of the project.

  17. Is there a way to contact Jay?
    I believe that NASA already has a very, very, very large working Ion or Plasma engine.
    I would like to discuss what I witnessed flying over my car one night near Kingman AZ in 2008. It may help Jay to scale up his project.
    Thanks.

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.