Unconventional propulsion

exploring the different means of providing thrust

Similar projects worth following
This is a series of experiments for exploring propulsion using unconventional means. Efficiency, speed and practicality isn't on the top of my list.

The theme is of this project is to think outside of the box and project can be built with minimum tools.

This project is aimed at the "Wheels, Wings, and Walkers (build something that moves)" part of the contest.

The theme is of this project is to think outside of the box. It is an attempt to try to play with alternative means of propulsion which tends not to be efficient or practical. My entry is more in spirit with the silly cars in Wacky Races cartoon.

The hurdle I face with this contest is that I do not having the access to (nor the budget for) the usual fancy 3D printing, CNC, Laser cutter type of mechanical fabrication.

Motion without movable parts

I'll look at the different type of propulsion such as Magnetohydrodynamic drive, Ion wind. Both of these relies on the fluid media (water, air) for propulsion.

Magnetohydrodynamic drive:

Ion wind:

Interesting Links on ion wind:

An Investigation of Ionic Wind Propulsion - NASA Technical Reports

Expired patents on ion wind:

Electric wind generator - US4210847 (1978) Ion wind generator - US 3638058 A (1970)

Steam:  (not my project)

Single Servo

A proof of concept to show that one servo can achieve more than you think.

Row boat


CC BY 4.0 for hardware, GPL 3.0 for source code

View all 8 components

  • Single servo for both propulsion and steering - Part 4

    K.C. Lee07/23/2017 at 23:48 0 comments

    I got the IR remote code and servo code working together.  The row boat can now be controlled from my TV remote.   I have added left turn and right turn to the code.  

    I have recorded a new video of the boat in action for the contest entry:

    Source code have been pushed to github:

  • IR Remote Decoding

    K.C. Lee07/23/2017 at 15:20 0 comments

    I ported the IR Remote code from my Automatic audio source switching project.  Straightly speaking this has nothing to do with propulsion, but it makes life a bit easy for running a demo without a tether.

    I have to make a few minor modifications as STM8S003 (8-bit) has less resources than STM32F003 (ARM M0).

    • I have to share timer 2 for both servo and IR remote.  Servo PWM is set for 50ms timer reload (0x9c40).  Measuring input pulse width is slightly more work as the code has to keep track of whether or not a timer reload has occurred.
    • STM8S003 input capture cannot be triggered on both edges, so the code has to toggle between rising and falling edge.
    • The usual I/O register, IRQ difference.

    The rest of the IR  code seems to work as is. 

    I am using my TV remote with NEC2 IR protocol.    Each byte sent is followed by its 1's complement for error checking.  NEC2 extends the 8-bit address code to 16-bit by sneaking it into the 2nd byte.  Byte 3 is 1's complement of command byte in byte 2.

    The following shows the raw IR code captured from the debugger.  The device code for my Insignia (Bestbuy house brand) TV is 0x8615 and the command code is 0x15. 

    IR Commands for the menu navigation buttons:  Left = 22, Right = 21, Up = 66, Down = 67, Select = 24.

    My logic analyzer can only decode NEC1 IR protocol, so it complains about address error and stopped processing.  :(

    I added a 7 segments LED for visual feedback for the IR commands.  F= forward, L = Left, r = Right and Off = Stop.

     IR receiver is at lower left hand corner.

    The dog almost eat my homework.  STVD delete 1 of the source file, failed to update 2 files before it crashed. It did managed to write the deleted file to a temp file.    :( 

  • Single servo for both propulsion and steering - Part 3

    K.C. Lee07/13/2017 at 18:37 3 comments

    I made a "C" shaped cutout on a foam board. I screwed in a standoff with a pan head screw on one end and glue it to a styrofoam tray. I used some White Glue. (It wasn't strong enough, so I added in some hot glue to the side afterwards.)

    The "Tower 9g microservo" is mounted on a small piece of clear plastic (from a dollar store picture frame).

    The servo arm is attached to the oak with the wire from a couple of paper clips.

    I used the insert from a "wire spicing" connector to extend the length of the paper clip.

    The servo is powered from a USB battery pack. The inertia from heavy battery pack helps to stabilize the erratic movement of the boat.

    Demo potato shot of it moving forward awkwardly in the bath tub.

    There are no fins or steering. The inertia of battery is the only counter balance for the reactive force from the servo. I hope to be able to steer and possibly move backwards by modifying the speed of the servo steps.

  • Single servo for both propulsion and steering - Part 2

    K.C. Lee07/13/2017 at 17:44 0 comments

    I have written a table driven servo driver for my STM8 breakout board.

    ServoStep.Table point to a table that contains the From, To values as well as the number of Steps.

    const ServoTable_t Forward_Tbl[]=
    	{ TIM2_SERVO_SCALE(0),TIM2_SERVO_SCALE(30),5 },
    	{ TIM2_SERVO_SCALE(30),TIM2_SERVO_SCALE(90),1 },
    	{ TIM2_SERVO_SCALE(90),TIM2_SERVO_SCALE(90),5 },
    	{ TIM2_SERVO_SCALE(90),TIM2_SERVO_SCALE(-30),20},
    	{ TIM2_SERVO_SCALE(-30),TIM2_SERVO_SCALE(-90),1},
    	{ TIM2_SERVO_SCALE(-90),TIM2_SERVO_SCALE(-90),5},
    	{ TIM2_SERVO_SCALE(-90),TIM2_SERVO_SCALE(0),15},	

    The forward thrust is produced when the servo moves rapidly between (30° to 90°) and (-30° to -90°). The servo moves slowly between these end points to minimize the unwanted thrust.

    Once the Timer and interrupt is initialized, channel 2 of timer2 (pin 20) is used to generate a PWM waveform to drive a servo. The period is set at 20ms (50Hz) and the waveform is 1ms to 2ms.

    The following IRQ gets called at the end of each cycle and the servo is incremented/decremented. If it reach the maximum steps, then the next entry of the table is executed. If it reaches the end, it wraps around to the beginning of the table.

    // Call everytime TIMer2 overflows (50Hz)
    @far @interrupt void TIM2_Update_IRQ(void)
      TIM2->SR1 &= ~TIM2_SR1_UIF;	    // Clear update IRQ
      if(!ServoStep.Table)		    // turn off servo
        ServoStep.PWM = SERVO_OFF;
        ServoStep.Index = 0;
        if(!ServoStep.Steps)	    // Initialize PWM parm
          if(ServoStep.Table[ServoStep.Index].From == END_TABLE)
    	ServoStep.Index = 0;
          ServoStep.PWM = ServoStep.Table[ServoStep.Index].From;
          ServoStep.Delta =(ServoStep.Table[ServoStep.Index].To -
          ServoStep.Steps = ServoStep.Table[ServoStep.Index].Steps;
         else                            // Step the servo
           ServoStep.PWM += ServoStep.Delta;
       if ((ServoStep.PWM >= SERVO_MIN) &&	(ServoStep.PWM <= SERVO_MAX))
       { // Set PWM
         TIM2->CCR2H = ServoStep.PWM >>8;
         TIM2->CCR2L = ServoStep.PWM;

  • Single servo for both propulsion and steering - Part 1

    K.C. Lee05/18/2017 at 13:15 3 comments

    Most of the bots etc employs multiple servos. It is possible to do more with a lot less?
    This is a Sampai that uses a single long sculling oar called a yuloh for both propulsion and steering. The operator row the oar back and forth similar to a fish's tail.


    When the oar pushes against the water, it causes an opposite reaction force. This force can be resolved into a vertical (Fy) and horizontal (Fx) vector components along the axis of the boat. Fx moves the boat forward while Fy tries to move it sideways.

    The sideways force (Fy) cancels out over time when the oar swings symmetrically side to side.

    What happens when the Oar is offset to one side?
    The vertical component no longer cancels out (Fy' - Fy >0), and so the boat turns right.

    But wait, there is more!
    When the angle ɵ is small, most of the force is in the side way direction. By slowing down the swing during small ɵ, this thrust can be minimized.

    So what happens if the forward motion is faster than the return?
    There will be more force produced in the forward pass. (In theory) the boat can be made to move backwards.

    Could all of this be duplicated using a single servo using some clever programming?

    Links: How to make a model

  • Magnetohydrodynamic drive - part 2

    K.C. Lee05/11/2017 at 22:25 6 comments

    The problem I had previously was the magnet I used has the wrong pattern. The Radially IN / Radially Out configuration configuration used in permanent DC motors are the right type as you want the flux to go vertically the center while you apply current in the horizontal direction.

    I toke apart a DC motor. The magnet are glued to the motor body and there was a piece of V shaped spring to hold it in place. They are the weaker ceramic type.

    Since the magnets are weak, I forgo the tube so that they can be closer together. I glued 2 pieces of aluminium foils onto the sides.

    I connect 130V DC across the foil. The current is about 0.12A. This time there is a weak stream of water and bubbles exiting the top side.

    There are rare earth magnets that I could order to improve the amount of thrust, but I have decided this is a dead end and not to spend money. So instead I added some soy sauce for contrast and a couple of teaspoon of salt to increase the conductivity i.e. stronger currents to make up for the weaker magnetic field.

    The current rises rapidly to close to 2A and generate a lot of bubbles. The steam is a bit better, but most of the 260W of power ended up in heat and electrolysis.

    The DC motor has a few orders of magnitude higher thrust and efficiency:

    • much smaller air gap for the magnetic flux
    • copper is a much better conductor than water. So much higher currents and lower conduction loss
    • multiple turns on the copper

  • Ion wind test #1

    K.C. Lee05/11/2017 at 00:47 7 comments

    I cut 4 strip of foam boards 0.74" x 6" (18.8mm x 152mm) from the dollar store. I glued a 0.5" (12.7mm) wide strip of aluminium foil and a couple of drain wires. This time I use a UHUstic glue stick. :)

    The 4 pieces was stacked together and the drain wires twisted together forming the "ground" connection. Large sized popsicle stick I got from the craft section of a dollar store for scale.

    On each end of the stack, I use hot glue to attach a piece of card board. I drill 4 holes that lines up with the aluminium foil.

    I threaded a piece of 0.002" (0.05mm) dia of magnetic wire between the opposite side. The bendy cardboard helps to provide a bit of tension for the wire.

    The wire is attached to the "hot" side of the multiplier.

    One thing I notice is that the wire bends towards to aluminium when power is applied due to electrostatic force. I added a piece of foam in the middle to prevent the wire from bending. I added extra hot glue to electrically insulate corona spots on the foil and drain wire.

    Power off:

    Power on:

    The wind is a bit weak, but enough to lift the foil and the receipt on the left side The power supply is drawing quite a bit of power and the output voltage is below the arcing threshold. The transformer in the flash circuit is slightly warm.

    I'll need to rethink the geometry of the design to optimize for air movement, electric field etc. I might have to look at the flash circuit to see if I can tweak a bit more power out of it or to seek higher power alternative.

  • It's over 9000V!

    K.C. Lee05/08/2017 at 22:12 4 comments

    The voltage I am playing with is at a dangerous level and the arc can easily jump 0.3". Please treat high voltages with a bit of respect. The ozone and UV from the arc isn't good for your health.

    While waiting for the ionizers, I played around I did some measurements on the flash circuit in one of those disposable cameras. It turns out that the output voltage can go quite a bit higher than the 300V or so. I don't expect it to work this well.

    One way of generating a high voltage source is to use a voltage multiplier with a high voltage AC source. I bought some 0.01uF 3kV caps from China as spares after fixing my scope. I also bought some 1N4007 high voltage rectifier (1000V) as they were cheap.

    I made some "springs" out of the trimmed leads. The caps are divided into two groups. They are laid out side by side and solder just enough to hold them together.

    A voltage multiplier looks like this. There are two rows of capacitors that are interconnected with diodes. 10 caps and 10 diodes makes a 5X multiplier.

    The springs are used to hold and connect the leads together in this free form multiplier. The second row of capacitor goes on top in the opposite direction.

    I guess I should have connected the bare wire to the lower left terminal instead of top.

    I stripped off some of the parts (large capacitor, flash tube) for safety. I also desoldered one of the leads of the rectifier to disconnect the rest of the high voltage circuits.

    I connect the multiplier to the AC output. The resistor in mine is 82R instead of 220R. The following shows a 5X multiplier. The multiplier can be extended to a certain point by just adding more diodes and caps.

    I swapped the header I had previously as the spacing was a bit too close so I could not control where it sparks. I bent the pins to form a spark gap of about 0.1". I also connected the bare wire to the right terminal this time.

    I hooked up my bench supply and started getting sparks at around 1.2V. Each arc discharges the capacitor and the charge has to build up again. At around 1.5V, the current was 0.4A.

    So according to this chart from here, the output is a bit below 5kV. My guess is around 3-ish. This means that the (loaded) AC output from the flash circuit is around 600V.

    With the additional block, I now have a 10X multiplier. I have to change the spark gap a couple of times to now 0.3" (7.62mm). :)

    It starts arcing at around 0.9V, so the output is round 9000V at this point.

    This shows the corona discharge just below the threshold for arcing.

    My old "Sparky" has a much beefier inverter (long exposure). The 10X multiplier is sitting on bubble wrap for insulation. May be it was higher voltage.

  • TIE boat - parts ordered

    K.C. Lee05/04/2017 at 16:00 0 comments

    I have previously built a lifter with a high voltage inverter and a 20X voltage multiplier. One side is attached to very fine "corona" wire when the other to a collector made of a piece of aluminium foil. Due to the geometry, the fine wire produces a very high electric field that ionizes air molecules. The charged molecules is attracted and accelerated to the opposite charge on the foil. They collides with neutral air molecules along the way transferring some of their momentum. This creates an ion wind and can lift up a small craft that weights a few grams.

    The idea is to use ion engine for propulsion on a boat as weight is less of a concern. A TIE (Twin Ion Engine) configuration can be used to steer the craft.

    I have some idea of building a stack of aluminium foil coated popsicle sticks or foam boards similar to mulitplane. The gap between the wings allows the wind to pass through. The corona wires are split between left and right side and powered by separate ionizors. The 12V supply to them can be switched easily with a MOSFET or a transistor.

    I have ordered some car ionizors from China for the experiment. Hopefully they would be here in about 2 months which sort of lines up with the contest time frame.

    There are other source of high voltage module such as igniter, flyback and Tesla coil kits. I feel that this might be much lower power, very light weight and designed to be to operated hours at a time. Of course I could be totally wrong when mail ordering random junk from China especially when they have to label their products as real.

    While I can wind my transformer and build my own high voltage generator, this would make it a lot easier to duplicate the design.

    I don't have the full spec, but I would expect that they are designed to operate more than a few minutes at a time. This is what it is supposed to look like inside. I might tweak the design a bit.

    Here is a quick update: The ionizer finally showed up yesterday and took a bit less than 2 months.

    I have sketched the schematic. Seems to be a push/pull driver.

    The primary and secondary windings (3 in series) are on separate bobbins for high voltage insulation. It also means that the primary can be modified too.

    The ozone generator looks like a neon tube but with a pale blue glow and a wire mesh on the outside.

    There was a blown transistor. Could have been me reversing the polarity or that way from the factory. The transistors are 2SC945 - generic NPN with a ECB pin out. I have replace them with 2N2222 and that seems to work.

    The 0.1uF was supposed to be 630/400V according to the silkscreen on the PCB. It is supposed to be a film cap and there would be some circulating current in this topology. Haven't check if it runs hot - probably not given the low power.

    The ionizer seems to start around 2.5V and uses 0.09A at 12V.

    It doesn't have enough power to drive the voltage multiplier. I guess back to the camera flash circuit.

  • Failure Log #1 - Magnetohydrodynamic drive

    K.C. Lee04/30/2017 at 21:38 0 comments

    Fleming's left-hand rule for motors is a way of visualizing the direction of motion of the force produced by the cross product of the magnetic field and current. The current in Magnetohydrodynamic drive just happens to be conducted through water.


    In layman's term, apply voltage across 2 electrodes and a magnetic filed and you have a repulsive force. The amount of current can be increased by either increasing surface area of the electrode, decreasing the separation or increasing the voltage applied or (the conductivity which I am not going to do here).

    Build a few of these and control them with the necessary high voltage driver circuits.


    I recycled plastic tube from dollar store glue stick. These glue sticks are low quality and dried out.

    I cut 2 pieces of tin foils and lied the inside the one tube. I use a rare earth magnet from old HDD.

    Two bench supplies in series: That's 65.2V + 63V = 127V DC at 0.23A

    This is the other end of the set up:

    Zoom in

    The tiny bubbles are due to electrolysis. Very disappointing result. For 127V x 0.23A = 29W, I would expect the bubbles get push a bit farther back due to the water current.

    The voltages are getting to be a bit dangerous and it looks like this needs to be scaled up by a lot to be of practical use.

View all 10 project logs

Enjoy this project?



ActualDragon wrote 05/18/2017 at 17:33 point

hey, have you given any thought to trying out the magnus effect?

  Are you sure? yes | no

Morning.Star wrote 05/02/2017 at 12:39 point

*To separate strong magnets like HDDs, work a couple of bits of cardboard between them, fold them to enclose each side and then slide them apart laterally. Nice and safe with flat profiles that break easily. And to remove them from the bearers never lever them from underneath, they are secured with a dab of lacquer and will break. Use a flat-bladed screwdriver and try and slide the magnet sideways by levering against one of the stand-offs.

I've recovered loads of those things over the years, Alnico, then Cobalt Samarium and finally Neodymiums. These are particularly bad for breaking as they are sintered powder with a plated surface and are very brittle.

I cant wait to see what else you have up your sleeve then. :-)

  Are you sure? yes | no

K.C. Lee wrote 04/30/2017 at 21:14 point

Looks like this is a failure.  I was playing with 127V DC supply and at 0.23A (~29W), I am getting more bubbles from electrolysis than force.  :P  Even then does not seem to provide enough thrust to move those bubbles.

  Are you sure? yes | no

Yann Guidon / YGDES wrote 04/30/2017 at 23:01 point

"Bubbles" looks nice for a boat name.

  Are you sure? yes | no

Morning.Star wrote 05/01/2017 at 07:09 point


  Are you sure? yes | no

Morning.Star wrote 05/01/2017 at 07:07 point

MHD, nice idea :-) Dont give up!

The bubbles are caused by current, you need to use a higher voltage and much lower amperage to avoid them. Water is diamagnetic by the way, not magnetic or non-magnetic as most people assume. It's very faint, but it's there.

Float a test-tube of water on a cork disc in a basin of water and bring a strong magnet near it to see the effect - it will move away from either end of the magnet.

This could be what is interfering with the expected result. ;-)

  Are you sure? yes | no

K.C. Lee wrote 05/01/2017 at 16:40 point

The problem is that the amount of thrust is proportional to current.  I think adding salt like other attempts is kind of cheating a bit, so I cranked up on the voltage to compensate.  This means the efficiency is very low due to most of it ended up turning into heat.

Don't worry, I know some other way to produce thrust.

  Are you sure? yes | no

Yann Guidon / YGDES wrote 05/01/2017 at 21:03 point

Is this solution called water ?

and does it involve igniting the bubbles ?...

  Are you sure? yes | no

K.C. Lee wrote 05/01/2017 at 22:32 point

It will also have no movable parts and high voltages. I watched "The World is not Enough" last night, so it has to be a boat.  There won't be any explosions.

  Are you sure? yes | no

Morning.Star wrote 05/01/2017 at 22:42 point

@Yann Guidon / YGDES I'd be interested how that would work out. Being as the electrical energy goes into expanding the water into a gas (about 1:1000 in volume) and then goes into heat as it condenses into water again when burned, I'm not sure how much net thrust would be obtained.

This I believe is the mechanism behind the odd 'pop' caused by igniting hydrogen in a test tube compared with other flammable gasses.

@K.C. Lee Adding salt will increase electrolysis and your foils will also dissolve really quickly. I'd recommend stainless steel if you want it to last very long in any case although distilled water is nearly non-conductive and non-reactive.

  Are you sure? yes | no

Morning.Star wrote 05/02/2017 at 06:55 point

There is also the magnet you are using has a N-S-N-S configuration. They are brilliant for small AC generator rotors, here it would cause an eddy? in the drive chamber rather than thrust.

  Are you sure? yes | no

K.C. Lee wrote 05/02/2017 at 09:30 point

You might be onto something about the magnet configuration.  It was from the voice coil of a HDD.  The NS poles are not on the opposite side of the flat surfaces as I have assumed.

NNNNNNNNNNNNNN   but rather   N---------------S

The flux was indeed perpendicular, but the orientation means that the thrust is cancelled out when I run the tube along the N--------S direction.  I'll need to find a different type of magnet before I can retry.  Really don't want to stack these magnets as they are fragile. Once stacked, they would be very hard to separate.

My conclusion about needing lots of power still true.  Since F = MA  the whole thing has to be light weight and that means that this is not what I am looking for.  I already have a different idea that would work a lot better.  I did that years ago so I know it will produce a few grams of thrust.  I just have to do a bit more construction.

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates