Looking back through the archives, we actually haven’t seen much in the way of homebrew magnetohydrodynamic drives (MHDs) — which is somewhat surprising, as the core concept isn’t nearly as complicated as its syllable-laden name might indicate. You can see results with little more than a magnet, a couple of electrodes, and a bench power supply. The trick is turning these base components into something that might actually have practical value.
That’s where we find [Jay Bowles], who has gone down a bit of a MHD rabbit hole these last few months. His latest MHD unit is a considerable improvement over its predecessor by all practical metrics, and as an added bonus, really nails the look of a futuristic propulsion unit. Even though the all-electric thruster hasn’t gone on a mission to anywhere more exotic than a table-top aquarium, you could easily imagine a pair of them slung under some top secret stealth watercraft.
What’s the magic behind the MHD? When a charge is passed through a pair of submerged electrodes, it interacts with the magnetic field within the drive and causes the electrified water to be accelerated through it. It’s the Lorentz force in action, and if powerful enough, can propel a craft through the water with no noise or moving parts. Beyond the slight downside that it only works in salt water, you can see how there would be considerable interest in such technology. Unsurprisingly, the military has had their eye on large-scale MHD drives for decades.
For his latest MHD drive, [Jay] has arranged the magnets radially and given them a waterproof coating. The core of the thruster is 3D printed and designed so water can move through it smoothly, and the inner and outer electrodes are off-the-shelf stainless steel pipe sections. The idea is that, once the pipes start to break down from electrolysis, it will be easy to swap in a new set. The whole thing was designed with ease of assembly and maintenance in mind, and it shows.
In terms of performance, [Jay] says this new MHD drive reached an exhaust velocity of 50 cm/s and a flow rate of 3650 ml/s while consuming 30 amps at 25 VDC. He calculates that out to 0.2 watts per ml/s of flow, which is a big boost over the previous thruster, which needed 1.2 watts per ml/s. What does all this mean in a practical sense? Hopefully we won’t have to wait long for the answer, as it sounds like the next step is bolting a pair of these improved MHDs onto a boat and taking it out for a spin.
This all might seem like something of a change of pace for [Jay]. After all, the 25 volts getting pumped through these prototype MHDs is nothing compared to the high-voltage experiments we usually see from Plasma Channel. But his newfound interest in marine propulsion actually came about through an attempt to strap a pair of his multi-stage ion thrusters onto a RC catamaran over the summer, so it’s not quite as unrelated as you might think.
Interesting design. I like it. I will note that 361 stainless is 16-18% chromium, so as electrolysis occurs, chromium will end up in the water. The oxidation state and what other reactions that may occur will depend on a number of unknown factors.
Given his environmental pitch, I’m sure he’d appreciate that feedback.
Yeah he should check on that, there’s a chance he’s poisoning himself
Indeed, it will absolutely end up in the water! Having built a similar, smaller device a number of years ago and subsequently tested the water at a local laboratory, I can confirm that this design, and anything similar to it, produces copious quantities of heavy metal contamination in any water it operates in. That contamination also builds up on the device itself and will spread further through physical contact like any other loose heavy metal contamination.
It’s also not a “next generation” design. Stuff like this has been looked at for well over a half a century. It’s neat, but not practical. Some relatively simple thermodynamic calculations can show that this type of propulsion has a multitude of losses in spite of it’s apparent simplicity.
I’ve observed a distressingly high number of projects that use Chromium-bearing alloys in electrolysis projects with no regard for personal safety or environmental integrity. I can appreciate that the builders want something that doesn’t turn to rusty sludge within a few seconds of operation, but waste products containing chromium are persistent environmental contaminants do real, measurable damage to both wildlife and people.
Thank you for the write-up Mike.
For those that don’t know or didn’t, Mike’s comment is a good proof of reality.
Mike, I have a side project with a solar mirror. It’s a trough made amd held with a vacuum using dirt cheap mylar. The optics are basically perfect. At a size roughly 48″ long and 42″ inchs wide and about a foot deep or so, it’s probably the equivalent heat of a medium burner stove on a 1/2″ id black iron pipe, with decent sun light. I am trying to make something super cheap and very safe for people to make for themselves if need be to get usable energy. Do you have any suggestions as to how I might utilize that heat energy somewhat efficiently without it getting too complicated or expensive or beyond the means of typical person to create? I would appreciate any thoughts on the matter you may have. Thank you.
Well the simplest option is a Thermoelectric/Peltier module assuming electricity is the goal – nothing complicated to understand or fabricate. If not just point it at a cast Iron pot and use that heat directly as heat (can even stand that pot on the Peltier for a small amount of effectively bonus electricity – the heat lost out the bottom of the pan might change a bit, but there was always some loss there.
The most efficient way for electric I can think of that won’t require that much effort is a Stirling engine or Thermoacoustic engine (much the same thing in principle though a Thermoacoustic should be easier to build its probably harder to tune in). You can get a kit to build those or make one yourself quite simply anyway – chasing peak efficiency is a whole other challenge though.
In theory a solar cell could be the target but I’d not suggest it seriously when you are by the sounds of it creating a very extreme focusing effect – the efficiency of solar PV drops off with heat and the panels construction can only take so much..
Thank you. Thermoacoustics sounds promising. I am looking into it now.
It’s has to be easy, simple and safe and at least 8% efficient if it can be from reasonably available cheap materials.
In the quick test configuration it boiled water inside the filled pipe almost instantly. I have concerns about safety using steam and I am also looking for something with less parts and maintenance. Thermoacoustics may be exactly what I need. :)
That’s an excellent project.
Back in the 1970s, my grandfather built an array of solar heaters for his house using parabolic reflectors that focused sunlight on black pipes willed with flowing water. Tracking was done with some simple servos, timers, and photocells. The energy from the hot water was pumped through heat exchangers to a non-eutectic phase-change salt solution in large barrels. The solar energy gathered during the day was used to heat the house at night. The house was quite a distance from the nearest town and large enough that heating with gas or electric would have been ruinously expensive. The solar heat is still humming away with minimal regular maintenance.
Wax would also work for heat storage, but it is more flammable. Water is usable, but it not as energy dense. Pumped air can also be used instead of water for the working fluid. It’s simpler, but takes a pretty hefty fan to move enough heat energy. I like your use of mylar for the reflector. My grandpa used aluminum flashing and that stuff got as hot as heck and had to be on ceramic insulators (re-purposed from power poles, I recall).
Gathering, transporting, storing, and using heat energy directly will always be the most efficient. If electricity is needed, then some sort of low temperature differential engine is best. It is important to match the size of the engine to the energy supply since the efficiency of the engine is proportional to the available temperature difference. Semiconductor energy conversion from heat is less efficient, more expensive, and limited in output, but is potentially easier to maintain. The same temperature difference applies. A geothermal heat sink for the “cool” end can help efficiency a lot.
A portable solar oven that can safely cook a potato and change a phone would be a dream come true!
Thank you Mike. I like your ideas. Whatever I end up doing I will try to share it with you later. Tracking will be timed with water flow. That’s the plan anyway. Wax is very interesting, but what your gramfather used seems well thought out. You have personal evidence of the final product.
Hexavalent chromium?
Multiple chromium and nickel species were identified in all samples we had tested. The ever-terrifying hexavalent chromium was measurably present in many samples and well past exposure limits in most where it was. Some electrode configurations were far worse than others. The alloys with high percentages of nickel and silicon produced wildly varying results. My suspicion was that crystalline structure (in the case of the silicon) and electroplating of the cathode (in the case of the nickel) changed the electrode mechanically during electrolysis and affected the interfaces with the solution enough that oxidation and reduction favored different paths. Using a frequency response analyzer and some low-noise amplifiers, we were able to measure the growth and breakdown of the plating and oxide layers. Electrolysis of salt water with stainless steel is wild stuff. Sadly, we never found a configuration that didn’t tear at least one electrode to shreds and scatter all manner of metal-bearing compounds into the solution.
Using chromium-free electrodes in areas expecting high oxidation limited some of the chromium issues, but others popped up. Traces of manganese, lead, tin, zinc, and copper in a veritable zoo of oxidation states all took their turns above the EPA limit. Pure iron electrodes were not acutely toxic, but they didn’t last long and made an awful, corrosive exhaust that ate right through any exposed copper. There were brief forays into conductive polymers and ostensibly nontoxic metalloids, but those all made an even bigger mess.
All that being said, there are likely some avenues left to explore. I recommend a thorough review of the literature and an abundance of caution before exploring them.
Maybe graphite foil as the electrodes? Or would that just break down too?
@Beowulf Shaeffer
Graphite was tried. The graphite/clay material that is used in pencils rapidly decomposed into beautiful slivers and sheets when even small currents were used. Some more pure graphite blocks used for casting and motor brushes worked slightly better. Pitting was a huge problem. I suspect that the non-crystaline structure of the compressed graphite was just too porous and brittle. I didn’t have a large diamond or any nanotubes at hand to try.
@Mike
Did you try anything similar to carbon felt or woven fiber? Not that it would necessarily help, though it’s used in other situations for surface area, but I’m curious what it’d do. Oh, or maybe the kind of carbon electrode made with coke and tar for aluminum production?
Otherwise, i dunno, maybe there’s some metal oxides that could work? Ceramics? Something using ion conductors? Not that I think it’s going to be great, but you’d think there would be something. Maybe capacitive coupling and swapping the arrow in the appropriate spots.
I assume Graphite or Platinum electrodes would be much better. The latter being a whole lot more expensive of course…
That was my second thought. My first thought is all the chlorine this will release into the water. That can’t be good for marine life. Perhaps it will be diluted enough that it won’t matter even if MHD drives become popular, but it’s still something to think about…
And sodium hydroxide
Hi Cliff. Thank you for the chemistry analysis. It seems more yt video rewrite.
I believe it will indeed produce the nasty hexavalent chromium ie Erin Brokovich. I think you’ll find plain steel will last just as long on those conditions.
The chromium species will certainly include hexavalent, but the mix will depend, as I said briefly and elaborated on but Mike and others, on the exact conditions. Also be aware that chromium oxidation state can change due to conditions (Pretty much any reducing agent for the system pushes the equilibrium to trivalent, which is less toxic, but still not good to have dissolved in your seawater, and is more more stable than hexavalent. Moderately strong oxidizers push it to hexavalent)
I don’t see how positive something that breaks down water can be.
He forgot something basic: The terminals. He cut the cylinders without leaving anything so he could connect them.
He used too much filament printing everything in one piece.
He should have broken the design into several parts, so as not to use supports.
I see a pretty big lack of 3D printing knowledge here.
Or at least tree supports.
A few sloping faces in the right place and he could still have done single piece printing, but with no support needed. Model the “upper” ends of the tall gaps and spaces like a gothic arch, if it absolutely needs to be flat for a close fit to the magnets then print small sections to glue in after printing so as to block out the bottom of the arched regions.
Beyond the “slight downside that it only works in salt water,” there’s the fact that it’s comically inefficient without incredibly strong magnetic fields. This one gets less than half a watt of mechanical output from 750 watts of electrical input.
Just borrow the magnets off this bad boy and one will really be going.
https://youtu.be/gD3dMzv1vIQ
Adjusting for the power supply, it would probably be closer to 1000 Watts out of the wall.
Any way to get a magnetic field intensity between to magnets?
3.65 kg/s x 0.5 m/s = 1.825 kg-m/s^2 which translates directly to 1.825 Newtons of thrust.
If you can maintain a steady forward speed of 0.5 m/s with that amount of thrust, then your output power is equal to 0.91 Watts, and the efficiency of the MHD thruster is about 0.12%.
Sounds like whale flatulence, an earthquake…. anything but a submarine.
“One ping only.”
Now if he could just modulate the power supply to make it play Paganini, then he’d really have something.
“Beyond the slight downside that it only works in salt water”
well, that and the ‘slight’ downside of running a powerful electric current through saltwater breaks the water down into some pretty corrosive stuff that pretty quickly destroys your electrodes.
Great story,,, I love when someone takes a BOOMER invention and updates it ( popular science,, popular mech.) ,60s–70s,,, they just created a water raceway between terminals,, worked good. Purpose????
Anyway nice start