Making a Simple Magnetohydrodynamic Generator, Need Help Optimising It

I don't understand 'steam engine' as MHD is an electrical generating concept, not a mechanical generating concept.

Its been many years since I looked at this technology. The first thing I remember is that the magnetic field is very intense, as it must essentially ionize the fluid to become a plasma. I'm not sure what the strength of your magnets is, but it doesn't sound very strong.

Secondly, if you are introducing conductive dust into the steam flow, perhaps you should orient the shown chamber vertically, in order to prevent the natural stratification that would occur. The fluid is considered mathematically as a 'perfect conductor', and then 'diffused' down from that, based on the conductivity. I think that by using steam with conductive particles, you are still having a high conductivity.

You might be better to use a steam turbine to pump brine (with particulate) through the magnetic chamber, and create your electrical plasma that way. I don't think MHD ship engines work effectively in fresh water, as the conductivity is lower. (I'm guessing)

but then, I don't really understand your setup. (steam engine... if you have the steam, you can run the engine directly... no mhd needed)

Chris

The steam engine generates mechanical motion, in form of an oscillating liquid piston passing through the MHD, which then extracts current.

Looks a bit like this:

But I'll get into all that in the next thread.

I'm wanting to work with salt water, not plasma, so the magnets only need to be strong enough to deflect the ions. I figure the HDD rare earths I have are probably 1-1.5 Tesla, and will be less than a centimetre apart. Good thing about a liquid piston; you can squish it down thin.

Has anyone found a good mathematical model for this effect? This might answer many questions and spare you many plumbing and boiler problems.

You were ahead of the curve with the solar option.

While room-temperature fusion is still a pipe dream <sorry for the pun, but not very>, there does appear to be some promise in LENR fusion in combination with MHD.

I have a pile of web pages you might find interesting, with a link to a fairly comprehensive MHD mathematical treatment at this web page.

My plan is to use supercritical water as the ion-transport media in both solar-powered and fusion-powered engines where the gaseous water is the only moving part.

Isn't supercritical water at about 36,000 psi? I'm wondering if there is a little difficulty in getting heat into water at that pressure, since that's at the edge of metallurgy as we know it....

Jon.

See here for a plot of temperature vs water vapor pressure. We're takling about temperatures in the 374 - 400°C range, and pressures in the neighborhood of 22,000kPa (~3200psi).

My major concern is the pressure - and I've been assured that there are off-the-shelf stainless steel alloys that can safely handle that combination of temperature and pressure.

I'm planning to use a µC to control an LENR reactor to produce and maintain an operating temperature in the "safe" region - if you're interested, you're welcome to take a look at the draft control code (updated periodically) at this web page.

Thanks for the excellent post.

> tapered nozzle with a 3mm square aperture

You'd recommend a downtapering nozzle rather than outward one I've pictured? Would that not reduce the flow rate?

Square for any particular reason?

> the voltage increased linearly, showing modest gains

What sort of figures were you getting for what kind of flows?

Did you calculate an overall efficiency?

> a pump to run salt water in a closed circuit past the MHD nozzle

Do you think this liquid piston would resolve those issues?

> this would increase resistance to the flow

Why so?

>You'd recommend a downtapering nozzle rather than outward one I've pictured? >Would that not reduce the flow rate?

Tapering increases the flow speed. It also increases the resistance thus reducing the pressure. Flow rate will remain approximately the same as we are talking about a (almost) incompressible liquid.

>Square for any particular reason?

I guess for ease of use. You can use flat electrodes and it's easier to get the flux though the water. I would use a square or rectangular tube too.

> this would increase resistance to the flow

I guess we are talking about fluidodynamic resistance.

I have another question though: When the fluid containing the ions flows past the magnets and thus generates an electric potential on the electrodes don't the ions get a magnetic pull slowing them down and thus creating a hydrodynamic resistance?

When the fluid containing the ions flows past the magnets and thus generates an electric potential on the electrodes don't the ions get a magnetic pull slowing them down and thus creating a hydrodynamic resistance?

Yuppers. TANSTAAFL (There aint no such thing as a free lunch)

Just to throw a bit of oil on the fire and an extra parameter in the equasion:

What if we would use two electrodes of different materials? This would add an electrical potential to the poles. I have a gut feeling that the flux created could add to the efficiency. I need to think about this a bit. :-/

Of course this means using a uni directional flow of the fluid.

Sorry for the misinterpretation...the magnetic field was applied across to 3mm square channel which tapered outward. The resistance to flow was getting the fluid from a 1" schedule 40 through the 3mm square.

Another post suggested different metals for the electrodes. I tried zinc and nickel. The effect was similar to using different electrolytes: different staring points at zero flow. When you check into the Physical Chemistry aspects, there is no great gain in performance in this temperature range.

By modest gains, I'm suggesting less than 5% increase in voltage over a range of 1 to 3 gallons per minute. Machining the valve was difficult. We wound up using two blocks of acrylic to mill out the channels. Running at higher rates began to separate the two pieces. Otherwise I expected to hit a saturation point as I decreased residence time in the nozzle. With additive manufacturing (3D printing) one might be able to draft and fabricate an MHD manifold with the electrodes already in place. You could include valves to open and close some of the lines to adjust resistance to flow. This won't get a big increase in voltage (per cell), but start hooking them up!

Just the same, it woun't be mich different than using a freshly charged NiCad (1.25V) vs a new alkaline (1.6V) battery. With solar as the source of energy, you might do better to build a power cell or use a Stirling engine to turn a dynamo.

I might still have my results from1979, but it would take a few weeks to try to dig them out of storage! I never thought MHD would come up, outside of Clancy's Hunt for Red October!

Welcome to CR4!

1. MHD generator tryout - YouTube

2. Mini MHD Generator - YouTube
   	More videos for mhd generators »

3. MHD Generator

So as I understand it, you are producing steam from a solar collector, injecting a conductive material, like a salt, into the steam, then passing the flow of this conductive fluid through a pair of magnets, and then extracting electrical energy from the fluid using a pair of conductors. Is that it?

Have you compared the efficiency of this system (in its totality) versus a regular piston steam engine? It seems to me that most of the energy of the steam you've produced is unproductively vented out the end of your nozzle. You would need the steam to do work on the ionic material, so that the steam's energy is efficiently extracted; i.e., super hot steam goes in and cooled steam/water comes out. The steam's energy would be used to break the ionic bonds, dissociating the ionic material into its positive and negative ions. The kinetic energy of the ions is then extracted by the conductors, producing an electrical current.

Or have I misinterpreted what you are hoping to do?

Ok, seems I should actually describe what I'm wanting to do in detail which is useful:

Solar concentrated on copper boiler tube, or flame.

Water boils, steam forces water out through MHD gennie.

Water enters insulated container, increasing air pressure.

Water also forced into small top reservoir, also increases pressure, but is heatsunk metal, so water cools.

When no more water in the boiler, steam stops being produced.

Pressure in boiler drops slightly below top chamber.

Spray of cold water forced back into boiler by air pressure, collapses steam.

Water sucked / pushed back through MHD and refills boiler.

Repeat as necessary.

=====

So, to be clear, it's the salt water which passes through the MHD and generates current, not the steam. The steam is just to move the water column.

Also to be clear I'm aware how much can not work with this idea, but if it does work, or can be made to, it'll be an ultra cheap and simple electrical generator home makeable by most people in the world largely from scrap. And having no moving parts, minus the liquid piston, low to no maintenance.

And looking at it, could actually be quite efficient. But if I can hit 10% solar to electric I'll be ecstatic.

I can see the heating of the water to steam.

I can see the steam powered expansion through the mag field into the chamber(s).

I can see cooling of the one chamber..

but then I see equilibrium being reached. I don't see a cycle.

I don't understand the purpose of the insulated chamber. to me, that only limits the flow of brine.

most thermal exchange systems (like a refrigerator) use a return path. I don't see the return path.

more explanation please.

Chris

> but then I see equilibrium being reached. I don't see a cycle.

This is the point where it might not work. The heatsunk top chamber will be at whatever maximum pressure is reached by the boiler. If the pressure in the boiler drops even slightly an amount of colder water will spray back in, collapsing steam, dropping the pressure, spraying more water etc. As this happens the just below boiling water in the insulated chamber will also be sucked back into the boiler through the MHD, again generating current.

So it all comes down to the pressure in the boiler dropping slightly when all the water has been forced out and there's no more steam being produced. Maybe it'll tip on its own, maybe I'll have to force it somehow.

BUT

this is next week's mission. Right now I just want to get a basic proof of concept on the MHD alone, using a bucket and gravity.

> I don't understand the purpose of the insulated chamber.

The water has to end up somewhere. Being insulated, energy won't be lost, and being sealed it'll build up a bit of pressure to help forcing the water back into the boiler. Not that it'll really need it with the vacuum pressure.

> I don't see the return path.

The water oscillates, like a piston, rather than moving in a circle.

Chris...

SUGARANDFAT is trying to sneak up on a fluidyne-like engine. I'm not sure that his current design is what he wants, but he already has resources available to deal with that issue.

What he's looking for (as am I) is help with the MHD aspect of using a fluidyne to produce electrical power.

If you're interested in solar thermal engines, I can provide some food for thought - but let's limit this thread to the MHD query.

What about just doing some tests?

I have several hard drive magnets. Getting copper should not be too hard. Salt is a non issue of course.

I have a hunch that a high salinity solution - a brine - should work best due to the highest possible content of ions.

Using a couple of DMMs for measuring volts and amps will give me power output. No load voltage and voltage with different loads can give me an idea of efficiency and scalability.

What I do worry about though is the corrosion of the copper in the very salty water. Coating the copper is not an option I guess.

What parameters would you vary to test efficiency?

> What about just doing some tests?

Starting in on that tomorrow. Got 6 magnets out of some hdds yesterday, will get the few other things I need for a basic test.

> I have a hunch that a high salinity solution - a brine - should work best

Is there anything better for this than salt? Salt is fine, but anything that can be tweaked..

> corrosion of the copper in the very salty water.

Possibly, time will tell. They'd be super easy to replace.

> What parameters would you vary to test efficiency?

Dunno, test or optimise? For the latter I guess magnet configuration and aperture, nozzle, rate of flow and salinity, etc.

To measure the efficiency I'd be looking to folks such as yourself with a better understanding of these things. But I've lined up a hackerspace here in Leipzig, so will have good access to meters and diagnostic equipment.

NaCl salt is easy to come by and should be able to at least give you a POC.

Other salts like CuSO4 might be a better solution although you might see a much higher migration from one pole to the other. Switching the coper plates regularly should be an easy fix. :-)

I need to look into this. Also thinking of FeSO4 and AlSO4 but I'm not sure these will dissolve well in water.

Let's do some basic testing first.

Ammonia and Sodium Hydroxide might also work well - and both have the advantage of being universally available (ammonia can be distilled from urine and NaOH can be leached from wood ash in third world contexts).

I'm guessing that volume, flow, and field strength are also significant determinants of output.

Haven't had much coffee yet, but I think the output would be long cycle AC (the solution will be flowing both ways), so the migration should even out.

Welcome to CR4!

Welcome to CR4!

Thanks! This appears as if it might be a good place to find help and share solutions to interesting problems.

For initial testing, you might not need to worry about corrosion. For long-term use with NaCl, cupro-nickel will be more corrosion resistant and will have good thermal conductivity. For NH₃, steel tubing would be better. I forget what works with NaOH.

Carbon steel will work fine with ammonia, and also with the caustic soda up to about pH 12, higher pH will result in the formation of ferrite ion in solution, along with corrosion of the steel. If you insist on using sodium chloride brine (not the highest conductor, but a good one), then you should simply include in solution a copper inhibitor if copper is to be the electrode material. Tolyltriazole solutions are effective on copper, but make sure solution pH is in the correct range.

As to the "cycling" of your steam engine/lquid piston, why not simply use a shutter to turn on/off the light (energy) input, thereby assisting the "collapse" of the steam pressure? Also, you are making a "boiler", and certain safety devices come into play. Water purity is also an issue and salt salt content in the presence of elevated temperature and pressure with mild steel may result in Stress Crack Corrosion of mild steel when (1) high heat stress is present, (2) chloride over 15 ppb (microgram/L), and the dissolved oxygen in the water is above 50-100 ppb. Good luck. Don't hurt yourself. Make sure the device has a safety relief valve somewhere.

The electrodes can be anything conductive, but I'd want to keep the main tube and in fact pretty much all the metal involved as copper due to thermal conductivity.

There will be some steel parts in form of a few little spigots and fittings etc, but even if these do corrode they could be easily and cheaply replaced. I'll probably be using bike parts.

Any conductive agent for the water is definitely going to have to be something readily available the world over, I always neglect to mention in my threads that all this is not so much for my own use as everyone else's, as in appropriate technology.

> why not simply use a shutter to turn on/off the light (energy) input

Considered it but: would waste input energy, it takes a surprisingly long time for steam in a shaded container to collapse from air temp, and it would be probably not the easiest mechanism to build.

> certain safety devices come into play

The pressure in this thing should stay pretty low, not high enough to blow a plastic bottle let alone anything metal. The temp shouldn't really come much above 100 C either, less the thing gets jammed somehow, but as the pressure increases it by definition removes the water away from the heat.

The pressure in this thing should stay pretty low, not high enough to blow a plastic bottle let alone anything metal. The temp shouldn't really come much above 100 C either, less the thing gets jammed somehow, but as the pressure increases it by definition removes the water away from the heat.

Keep in mind that most of the plastics that you might recycle begin to turn to mush in the neighborhood of 100°C (DAMHIKT).

Yeah, I've created a few RC helicopter canopies by putting PVC sheets in boiling water and then pulling them over a mold. The plastisizing temp of many plastics is around 120 degrees C. I needed to pull quiet hard to get them over the mold.

I think SugarAndFat was just providing an analogy. I guess he meant that the pressure would not be high enough to blow a plastic bottle at room temperature. PET bottles can resist about 1 MPa (10 bar) though.

Understood - but recall that what he's described is a closed system in which he's only moving the liquid water away from the heat; and that the pressure increase he's after raises the boiling point of the water and the temperature of the steam above what it would be at 1 Atm.

Plastic will never conduct heat the way metal does. You throw away efficiency where others try to find it. Obviously, by the rules of thermodynamics that govern all heat engines, the wider the separation of temperaures between the extreme hot and the extreme cold will provide for higher cycle efficiency in moving the liquid piston.

You're right. I was assuming/hoping that S&F was thinking of plastic only for his MHD section (between the hot and cold "heads"). The hot and cold heads definitely need all the thermal conductivity that can be managed.

The concept of using cold water to condense steam in a cylinder/piston to allow atmospheric pressure to return the piston back into the steam end of the cylinder (which was now a vacuum) was one of the first ways steam engines were built, before they figured out to make the piston double acting an to put steam alternately at each end.

I remember seeing an even earlier steam water pump where a chamber, with appropriate valving, was alternately filled with steam to push water out, and then allowed to condense, drawing water into the chamber.

This information came from some old turn of the century encyclopedias my parents had--I was very interested in steam engines around age 10.

Another similiar form of moving water is with the little toy boats that were operated with a candle and propelled by the steam from water in a "putt-putt" manner. Lee Valley tools in Ottawa used to sell them a couple of years ago--they were a reproduction from pre WWII toys. The principle seemed to quite similiar to the steam pump and probably not unlike a coffee percolator.

Your idea of moving water around in this manner could probably be made to work, although the efficiencies would be low.

Jon

Yeah I saw some designs for the earlier steam water pumps, guess this is kind of based on them.

The toy boats work by flash boiling water a small amount of water, I don't think this setup in a solar collector would likely reach high enough temperatures quick enough for that.

This may well prove inefficient, but I suspect it might resolve at least a couple of the issues of the earlier engines. But I'll give it all more thought when I get to it.

> I'm suggesting less than 5% increase in voltage over a range of 1 to 3 gallons per minute.

So, completely useless then? I'm wanting this thing to produce useful amounts of power, as in at least tens to hundreds of Watts...

> You could include valves to open and close some of the lines to adjust resistance to flow.

I'm looking to keep this all super low tech and easy for people to construct themselves. Developing world and all that.

> you might do better to build a power cell or use a Stirling engine to turn a dynamo.

I thought mhd had this super high potential efficiency? Is that just for plasmas?

> the pressure increase he's after raises the boiling point of the water and the temperature of the steam

The pressure should be well less than 2 bar. The water will start leaving the boiler as soon as steam is being produced, and you're only looking at 300 ml extra water in up to a 2 litre bottle, so not much.

Plastic bottles actually get harder in the presence of boiling water, and shrink a bit. So as long as the temperature or pressure don't get much higher than I'm planning it should be sound.

> plastic only for his MHD section (between the hot and cold "heads"

Is there a cold head? The chamber where the water ends up is insulated to prevent energy loss, so that when the water re enters the boiler it's still just below boiling.

This isn't anything like a Stirling cycle, it's more like two opposing pistons, with the somewhat pressure buildup in the bottle acting as the return, like charging a spring.

Regarding the Newcomen steam pump which this is kind of like, the main reasons they were such low efficiency, apparently, is that the cold water spray cooled the chamber walls and collapsed a fair amount of the next batch of steam, and that the piston housing was poorly made at the time and didn't seal well.

The first issue I think is resolved here partly by the wicking inside the boiler which would largely prevent the cold spray coming in much contact with the copper, but even if it does you're just sucking heat out of the metal into water which will then be boiled, so you're not actually losing anything...

And being a liquid piston there aren't really any issues with sealing or friction. Maybe a bit of turbulance, but not much even of that.

Is there a cold head? The chamber where the water ends up is insulated to prevent energy loss, so that when the water re enters the boiler it's still just below boiling.

That's what the (uninsulated) upper cyclinder looks like to me.

This isn't anything like a Stirling cycle, it's more like two opposing pistons, with the somewhat pressure buildup in the bottle acting as the return, like charging a spring.

Hmm. I think it might be. :-)

>>This isn't anything like a Stirling cycle, it's more like two opposing pistons, with the somewhat pressure buildup in the bottle acting as the return, like charging a spring.

>Hmm. I think it might be. :-)

Might be the opposing-pistons-with-spring idea, or might be a Stirling cycle? I agree it sounds interesting. The output would be low frequency AC. At high frequencies (a really hot pipe), with the brine changing direction in the MHD narrow conduit frequently, I would worry about cavitation chewing up the surfaces.

Remember my first post pointed out this was worked on by the Soviets and Israelis in the mid-1970s using 4000-5000 degC plasma. Sorry, no developing world solution here!

What has worked well in the developing world are solar box cookers (scouting and church groups have lead the charge in pre-making and distributing these inexpensive appliances). Meat is cooked without combustion, only collected solar energy. The ovens are made from cardboard boxes lined with aluminum or tin foil. Since the meat cooks but the box doesn't burn, my guess is the internal temperature is between 130 and 230 degC per box. If the ground is covered with a blanket after a night of radiant heat loss, you might have a temperature gradient that can really get a Stirling Engine moving!

I've seen small versions sitting on top of tea cups turning a propeller. There are also plans online to build a Stirling from food or soda cans. Can you scale up the design using material that might be at hand: paint cans, trash containers or 55 gallon drums? You might drive a fan placed under the rafters of a school building or clinic or pump water from a deep well into a distribution tower and still have enough torque to run a salvaged automobile generator to produce 10A, 12VDC to charge a battery, power LED lighting or a radio!

The MHD system will be great under other circumstances. Keep the thought on the back burner for now.

Sorry, no developing world solution here!

I disagree - and think you're too quick to write off the abilities of people in developing areas. There are those who only need to be shown how. Consider, if you would: this effort in Pakistan and another in the Sudan...

...and, of course, there's always William Kamkwamba's windmill to prove the point in the clearest possible way.

I think you helped make my point: the cited efforts in Pakistan and Sudan are both Stirling Engines. Kamkwamba's windmill is based on a simple set of ideas and built with on-hand materials.

The windmill story is a great example of walking before running because the man who built it got attention that allowed him to learn more and hopefully will do more. Kamkwamba was interested in helping his people and opted for the power of the wind as described in books he found in a library.

I'm also not suggesting the people in developing areas are unworthy of technical advances or unable to produce electromechanical devices. Consider a heavier version of technology transfer: the Buran Orbiter. Under the Freedom of Information Act, the Soviet Union gained plans for for the NASA Space Shuttle. In trying to build in accordance with the plans, they found many parts and materials not available in the Soviet Union. So this became a very expensive project. The Soviet/Russian government had no use for the this thing It was the wrong technology at the time .

Do you say mhd is inappropriate for the developing world due to it not being viable anywhere on a home makeable scale, or that there they'd lack the resources to build anything even close?

The first could well be true, but I'm only gong to give this project any mind if it can be made from an old bike and some plumbing fittings. If I do end up getting a solution out of all this, it will definitely be in the form of a very simple and easy to build device makeable from scrap and things you'd find in a corner store, with a comprehensive animated tutorial on its construction and use.

For the sole reason that, as you imply, anything less would be completely useless.

I would like to add that certainly in developing wolds efficiency is less important than having a working device.

I have seen many people decide not to do or help out on a project because the efficiency is too low in their opinion. Not doing a project has an efficiency of 0. Doing a project with an efficiency of 5% is is an endless amount higher. ;-)

I'd also like to add that a project with a very low cost can certify a low efficiency. If a higher efficiency means a much higher cost, it's not worth it.

In other words: if I can build something from scrap that has an output of only a few hundred watt using a free energy input, I will build it.

What I'm trying to get at here is the home built MHD adds a lot of work and effort with little improvement in return. You can supply more power for your trouble by digging holes in the ground put a plastic trash can liner in, fill with water, throw in some salt and electrodes and harness the energy of the wet cell farm for low power applications that need electricity (LED lighting, radio). Use the mechanical systems (windmill, Stirling) where you can get the best energy transfer.

A fuel cell makes sense because it's 40-60% efficient in ambient temperatures.

Use the right tool for the job. Generate electricity when and where you need it and do it safely.

> the home built MHD adds a lot of work and effort with little improvement in return.

If that's the case, and frankly it probably is, then yeah, never mind it.

However if it can be made to work, in a form that is easily replicable, then it'd be an amazing improvement on other heat to electrical generators.

> Generate electricity when and where you need it and do it safely.

That's the intention. Some people need electricity and have heaps of sunlight and not much else, especially at certain times of year. Also, anything that can be made to work with solar would likely also work with gas or other flame.

> Use the mechanical systems (windmill, Stirling) where you can get the best energy transfer.

Wind power is great and I'll be working on that too. Making any kind of Stirling can be a bit of a bugger, and a highish power one really not easy at all. I've seen high end ones that run to 40%, but an off the shelf kitset, still more expensive than most people in the world could afford, usually hits 5-7% overall. If something can be lashed up out of some magnets and copper wire which achieves the same numbers it would make electrical production much more viable where most people in the world live.

a few 100 watts would be spectacularly successful

the underlying premise is that the biggies would leave money on the table, by ignoring low cost generation techniques

the reason people generally don't pursue low efficiency projects, is the juice isn't worth the squeeze

too much money for too little return

Agreed Bert.

(I'd be pretty ecstatic with 5% and a few hundred Watts.)

Ok (I think we're both on the same page). I'm happy to help make your point.

I'm particularly pleased with the Pakistani and Sudanese successes because both efforts grew out of my web pages. Kamkwamba built his windmill at age fourteen, working from drawings in a (mobile) library book written in a language he didn't understand. I enjoyed Stewart's humor, but I'm still awed by the ability of a 14 year old kid to grasp the principles he needed from nothing more than line drawings.

Our challenge is to push useful technologies as far forward as we can manage (using whatever development tools are available to us) - and then reduce instances of those technologies to sets of line drawings that people limited to a bicycle repair shop implementation capability can put to use.

Most engineers seem fascinated by the coffee cup Stirling mentioned earlier - and as a toy it is fascinating - but it's worth noticing that it simply isn't worth scaling up. Nearly all Stirlings' operating basis is the expansion and contraction of a fixed volume of air, and that imposes serious limitations on their usefulness for doing real-world work. Sadly, if we scale up a toy, then we end up with nothing more than a bigger toy.

For those familiar with Stirlings who don't like hearing that, let me point out that their ability to do work is closely related to the internal pressure applied to a piston - and that for air dP/dT (amount of change in pressure resulting from a change in temperature) ≈ 0.34564 kPa/°C which is disgustingly small.

By way of contrast, for water gas (supercritical water) dP/dT ≈ 235.88 kPa/°C, way more than 600 times the pressure change for the same difference in temperature.

Note that if air gas is replaced with water gas it's still a Stirling engine, but now it's some 600+ times more powerful.

In terms of pushing ions into/through an MHD generating head (or even just pumping water), this represents a huge increase in output.

Of course what you'd have then wouldn't make a very good toy.

Okay, let's say we can safely harness a supercritical steam system (with bicycle shop methods and materials) to pump something past the MHD head. It sounds like you can sacrifice some efficiency by having two circulations: the supercritical steam and the ionized fluid. This lets you deal with the problems of high pressure and corrosive media separately.

As far as the what to pump question goes, you aren't going to use plasma gas. An electrolytic solution with electrodes produces electricity (there are tables for combinations of electrode metals and salts and what voltage is produced under zero flow conditions) and any other fluid can only present a hazard for personal or environmental damage. It's pretty obvious there is no other research out there, so start with salt water for version 1.0. Later, you might find something else that works even better: version 2.0.

You need to test if the voltage generated continues to increase with increased speed through the MHD head or is there a saturation point due limitations on ion mobility. If a saturation is reached before you get to your maximum flow, you may want a longer path through the magnetic field and possibly with more electrodes rather than one long electrode. This will increase residence time in the nozzle and allow for more "collection" on the electrodes. Because your resistance will increase, have the bicycle shops make a manifold with multiple MHD valves in parallel to balance the resistance. Then add more devices as needed.

Under such conditions, you'll get 100W at 5%. You might need a lot of MHD valves, but you should be able to get there. Good luck!

<very big grin>

Excellent. Now it's time to shake out some more expertise to address sugarandfat's original request for help optimizing an MHD generator head; and for me to go back into learning mode.

I wonder if we might benefit from having some EE involvement in this discussion...

Any chance of a rough sketch of the configuration you figure'd work best?

I'm thinking gravity feeding the water's not likely to get enough flow to get any kind of telling result, so might rig up some kind of hoover based vacuum pump.

Also, I have a pretty much unlimited supply of strong magnets (no shortage of munted hdds in the Leipzig hackspace), so can be stacking a bunch of them in series and parallel, get something a bit more than 1 T.

Sorry, don't have a sketch for what you want to accomplish. My MHD valve was proof-of-concept. I wasn't trying for the output levels you're after or in optimizing performance.

My valve had 3 regions: Inlet (threaded for 1 inch Schedule 40), Channel (1/8 inch square, 3 inch long to get laminar flow), Outlet (tapered from 1/8 inch square to 1.5inch square). Electrodes were placed in the outlet on opposite walls near the end of the channel. The 1T magnetic field was placed across the other two walls with. The inspiration to do this was a paragraph in an engineering handbook. I didn't have a line drawing at the time.

Hopefully your plans will be more inspirational and informative.

>Our challenge is to push useful technologies as far forward as we can manage (using whatever development tools are available to us) - and then reduce instances of those technologies to sets of line drawings that people limited to a bicycle repair shop implementation capability can put to use.

So true. Welcome to CR4.

Quick clarification: is it really going to produce AC? I would've thought that since the ion deflection is a result of the magnetic field, which remains static, that the direction of the fluid flow is not important, that the ions will always flow in the same direction?

This is being based, as is most of my thinking, on ignorance.

It's only AC in as much as it will change polarity depending on the direction of the fluid flow.

"... the direction of the fluid flow is not important ..." - but it is!

Imagine a charged particle moving, say, east-to-west in a 'downward' magnetic field - and say its path is diverted to its left (i.e. northwards).

Now (after some time) another charged particle (of the same polarity) comes west-to-east. The magnetic field is still 'downwards', so it'll also veer to the left - and thus be moving southwards, towards the other electrode.

Ok cool, good to know.

What are the issues going to be working with this kind of current?

Just run it through a full-wave rectifier (bridge rectifier) and treat it as intermittent DC.

Use the lowest voltage drop diodes you can get (which will handle the current - which probably won't be much so it should be easy). Unfortunately, you'll lose the voltage dropped across the diodes - power lost (heating the diodes) will be = (voltage drop x current).

What is a rough estimate of the solar to work efficiency?

From that you can calculate the required solar collector area where the insolation rate is known for any given location.

I believe that Lenz's law will apply to the induction fluid. That is - as you put a load on your circuit there will be a magnetic force generated in the induction fluid that will appose the fluid flow. The energy will translate from the kinetic energy of the induction fluid to circuit power through the electro-magnetic induction process.

This will probably cause a "pile up" of the ions in the induction fluid reducing the homogeneous distribution of the ions in the induction fluid.

This may not be the most appropriate technology for producing power; but it is an interesting study.

You are a lucky man to be able to have the resource to apply your imagination to applied science.

I look forward to seeing the results.

Gavilan

> What is a rough estimate of the solar to work efficiency?

Absolutely no idea whatsoever. You could tell me anything and I'd believe it.

At this stage I'm just looking to get some preliminary numbers, it might have to come down to someones else to really get the thing cranking.

> You are a lucky man to be able to have the resource to apply your imagination to applied science.

My only real resources are mobility and limited demands on my time. But guess that's most of what one needs..

Was in the hackerspace today measuring plastic bottle volume and siliconing things. Should be able to put a multimeter across the thing tomorrow and see what happens.

A few days in bed with a mild cold. I rest because I can. Back into it soon, maybe tomorrow seeing how I'm feeling.

Well, first results in, and they're not really positive.Salt water (pretty much saturated) between six hard-drive rare earth magnets (field strength dunno, but maybe around 2 Tesla?), flux aperture 4mm, water aperture 2mm, rate of flow about 50 ml per second. Put a volt meter across it and it went from -0.115 V to +0.145 V (it's not a very good volt meter).So, basically, practically nothing.

Clearly I need to make the thing bigger and a lot more grunty, but the problem is the magnets being quite small.

Had a thought tho; as the water in the liquid piston steam engine I was wanting to plug this into would be at almost 100 C, this will largely increase it's electrical conductivity, as well as the saturation point of the salt.

This is about the only data I can find on the subject:
http://www.analyticexpert.com/wp-content/uploads/2011/03/Graph1.jpg

But looks like about 3 times increase for the NaCl at 100 C.

I'm in the process of rigging up a much gruntier version which will hopefully give me 5-10 times the rate of flow, this with a couple more magnets and the temperature will either give me something interesting or show there's no real future in this.

Thoughts?

Can't you run your first rig with hot water (not necessarily boiling - just, say, 70 - 80°C) to see if there's any appreciable difference?

I kind of suspect I'm going to have to do this best case scenario even to get a voltmeter to register the thing. But hopefully I'm wrong about that.

The water will probably end up at about 90 C by the time it gets to the nozzle, I'll do a room temp run as point of reference and for the sake of science etc.

This paper is worth reading. Seems to suggest that the conductivity increases by 1 - 2% per °C.

Should be pretty easy to set up an experiment - just a beaker, couple of electrodes, ohmmeter, thermometer, heater & salty water.