Siemens CompetitionMHD Generator

[Shelnutt2]

http://www.collegeboard.com/student/pay/scholarships-and-aid/23619.html

I'm going to be entering into that (^^^) competition with one other person. We are going to be doing it on an MHD generator and its use for automotive transportation and as a home generator, amoung other uses.

We are just starting me research and design into this, but so far we are looking at building a disc stlye generator. We are looking to use Gallium as our liquid metal, as its relatively cheap comparted to other metals that are liquid at or around room temp, and its safer than Mercury. A PVC shell, and REM magnets. The thought has come up though that electromagnets might be a viable option, depending on the output of our generator. Heat will be used to circulate the gallium. This is a viable option as it allows for practically any heat source to power the generator. Also with Gallium's vary low melting point and very high boiling point, there is a wide range for the heat.

A MHD generator is based on the principles of Loretz force and the right hand rule.
F = Q( v x B )
* F is the force acting on the particle (vector),
* Q is charge of particle (scalar),
* v is velocity of particle (vector),
* x is the cross product,
* B is magnetic field (vector).

How can I do the cross product of v & B? A cross product requires 3 coordinates for v & B? Maybe I don't understand cross product.

Also what good is that equation to me? How does knowing the force acting on the particle help me? I would think my end result of this and maybe other equations would be to determine the power (wattage, voltage, amperage) of my designed generator?

I haven't been able to find much information on this subject, There are not that many websites. The public library doesn't have any books on MHD, and the University of Central only has 1 book on it in their library. My neighbour/friend is going to check out the book for me as he is a student there. Wikipedia has some information on MHD generators, but it lacks references (only 1 of the links work). MHD in general there is some good info but I think I understand the principles of MHD, I just lack example and in-depth information on them.

My purpose of this is project is to prove that a low heat MHD generator can be efficiently used for residential use. Basically, right now the only MHD generators that are out there are used in coal power plants where they use very hot liquid metal and on a grand scale of a power plant. Just think the benefits of a MHD generator for use in a car, in a household generator or any other application. The only information I could find on household generators was from Honda and it seems the average household generator, one that people run during the hurricanes, is only about 20% efficient in terms of energy production, which makes sense as the internal combustion engine on average for a gasoline engine is only about 20% efficient. Currently MHD generators are only about 20% efficient, or so wikipedia says, but they promise much greater efficiency if they are given the time of day. I don't know how efficient our generator will be but we hope to be around 20% or greater.

Any information anyone has on this subject, I'd love to know. I'm really hoping this project will land me at least in regionals. I still have one more year of high school to make it to nationals and I figure I need a better project that this to make it that far.

I'm posting this here knowing this is a great community.:)
edit:
Here is an autocad (trial edition) of what the generator *might* look like.
http://img211.imageshack.us/img211/6375/discgenmodelsmallne4.jpg

 

[xez]

Hmm where to begin?
I know a fair bit about the things you're asking about,
though I don't know if I can give a fully detailed response
tonight.

For starters:

a) Where are you getting gallium cheaply in any
useful quantity? I'm just curious since liquid metal
is handy for numerous purposes besides MHD type
pumps, and it would be nice to know where to get some.
I thought it was a bit expensive last time I looked at it
years ago.

b) You don't really need Gallium or MHD specific
references to do some simple calculations and experiments.
The lorentz force applies to any current flowing in a
magnetic field, even current in a wire, or current flowing
in liquid non-metals. For instance some common
conductive liquid non-metals would include electrolytes
like salt-water and suspensions of metal powder in
salt water, etc. etc. There's a classic motor that you can
make with a wire dipped in a beaker of salt water with
the current flowing through the wire down into the salt
water and back out through another electrode;
an external magnetic field causes a force on the current
and it spins the wire around the beaker. Anyway
the F = I cross B lorentz force is the same one that
makes any AC motor run, or makes the electron beam
curve to paint the picture in a cathode ray TV set, etc.
There's nothing really MHD specific about it. The only
thing that is MHD relevant is that the current flows in
a conductive liquid or plasma and therefore there's the
'hydro-dynamic' aspect to the motion of the charged
and current carrying liquid in the magnetic field.

c) Cross product? It's just basically the right hand rule
for the directions of fields and forces, etc. Check these
links and you should be able to understand it:
http://mathworld.wolfram.com/CrossProduct.html
http://planetmath.org/encyclopedia/CrossProduct.html
http://en.wikipedia.org/wiki/Cross_product
http://en.wikipedia.org/wiki/Right_hand_rule

Basically the short of it is that the force on the current
is at right angles to BOTH the current direction AND
the magnetic field direction, so, for instance, a current
flowing along in the X-Y plane with a magnetic field
directed in the Z plane which is uniform all over the
X-Y plane can be bent into a circular shape since it'll
be forced always to move at right angles to its current
path in XY and it won't have any Z directed force since
it'll stay in XY plane since the magnetic field is in the Z
plane and the XY plane is at right angles to Z...

Anyway since inverting the direction of current flow
OR inverting the direction of the magnetic field polarity
will reverse the direction of the force on the
moving charge you need the cross product to determine
whether the charge will move left or right assuming
the magnetic field is up and down oriented. Of course
if the magnetic field isn't at right angles to the current
the cross product will give you the right magnitude
and direction of the force too. And in the extreme case
where the current flows PARALLEL to the magnetic
field, there will be no force at all on the current.

d) And just to be clear, I = current = Q * v where Q is
charge and v is the velocity vector, of the charge, so
ordinary electrical current is the (Q*v) you're talking
about.


e) Hmm about understanding what use the lorentz
equation is in understanding various kinds of MHD
generator... well I was about to draw an analogy to
an MHD pump, but that'd probably just be confusing.

Basic reletavistic electrodynamics tells us that:

* a charged particle moving in an magnetic
field experiences the lorentz force.

* a strip conductor carrying an axial current normal
to a magnetic field will generate a transverse
electric field across the width of the strip due to the
Hall effect which is just the lorentz force on the charges
in the conductor pushing the electrons to the top/bottom
of the strip rather than having them spread uniformly
across it.

* Since a metal is like a gas of free electrons, the metal
moving physically across a magnetic field will cause
a hall effect voltage at right angles to the
metal's motion and at right angles to the magnetic
field due to the lorentz force acting on the 'free' electrons
in the metal.

Basically you just have to keep track of which
lorentz or other electromagnetic / relativistic effects
cause forces on the charge carriers in the fluid, and
also understand what currents and voltages you'll end
up with as a result of your design's geometry and
flow and circuitry. It's not too complicated if you
understand a few simple rules of electromagnetics
like the lorentz force, hall effect, and induced currents/
voltages etc.

f) As for low heat and efficiency? Well study up a bit
on the thermodynamic principles relating to heat
engines and temperature differentials of the
working fluid versus energy and efficiency.. There's
only so much heat energy available, and some of that
heat energy can be converted into kinetic flow energy,
and some of that kinetic flow energy can be converted
to electrical energy, but overall you can't get more
energy out than there is thermal free energy
to start with based on your initial and final temperatures
and so on... Be clear on what mechanisms you're
using to get the gallium flowing in the first place
since that's an efficiency factor just as the losses in the
MHD kinetic-to-electrical energy converter will also be.

http://www.collegeboard.com/student/pay/scholarships-and-aid/23619.html

I'm going to be entering into that (^^^) competition with one other person. We are going to be doing it on an MHD generator and its use for automotive transportation and as a home generator, amoung other uses.

We are just starting me research and design into this, but so far we are looking at building a disc stlye generator. We are looking to use Gallium as our liquid metal, as its relatively cheap comparted to other metals that are liquid at or around room temp, and its safer than Mercury. A PVC shell, and REM magnets. The thought has come up though that electromagnets might be a viable option, depending on the output of our generator. Heat will be used to circulate the gallium. This is a viable option as it allows for practically any heat source to power the generator. Also with Gallium's vary low melting point and very high boiling point, there is a wide range for the heat.

A MHD generator is based on the principles of Loretz force and the right hand rule.
F = Q( v x B )
* F is the force acting on the particle (vector),
* Q is charge of particle (scalar),
* v is velocity of particle (vector),
* x is the cross product,
* B is magnetic field (vector).

How can I do the cross product of v & B? A cross product requires 3 coordinates for v & B? Maybe I don't understand cross product.

Also what good is that equation to me? How does knowing the force acting on the particle help me? I would think my end result of this and maybe other equations would be to determine the power (wattage, voltage, amperage) of my designed generator?

I haven't been able to find much information on this subject, There are not that many websites. The public library doesn't have any books on MHD, and the University of Central only has 1 book on it in their library. My neighbour/friend is going to check out the book for me as he is a student there. Wikipedia has some information on MHD generators, but it lacks references (only 1 of the links work). MHD in general there is some good info but I think I understand the principles of MHD, I just lack example and in-depth information on them.

My purpose of this is project is to prove that a low heat MHD generator can be efficiently used for residential use. Basically, right now the only MHD generators that are out there are used in coal power plants where they use very hot liquid metal and on a grand scale of a power plant. Just think the benefits of a MHD generator for use in a car, in a household generator or any other application. The only information I could find on household generators was from Honda and it seems the average household generator, one that people run during the hurricanes, is only about 20% efficient in terms of energy production, which makes sense as the internal combustion engine on average for a gasoline engine is only about 20% efficient. Currently MHD generators are only about 20% efficient, or so wikipedia says, but they promise much greater efficiency if they are given the time of day. I don't know how efficient our generator will be but we hope to be around 20% or greater.

Any information anyone has on this subject, I'd love to know. I'm really hoping this project will land me at least in regionals. I still have one more year of high school to make it to nationals and I figure I need a better project that this to make it that far.

I'm posting this here knowing this is a great community.:)



edit:
Here is an autocad (trial edition) of what the generator *might* look like.
http://img211.imageshack.us/img211/6375/discgenmodelsmallne4.jpg

 

[Shelnutt2]

A) Well I was wrong about that. The cheapest I could find gallium was $1k for 500mg. I had read a few things, including http://theodoregray.com/PeriodicTable/Elements/031/index.html#sample2".

B) Wow, now that you've said that, this generator is a whole lot simpler than I first though. I mean its just basic principles.

C)Doh! So basically, its just for velocity, I have X,Y,0, 0 being Z axis? If that true, then I feel dumb. I mean its easy to take the cross product then. I was thinking that the velocity only had an X and Y value, and was thinking, that then there is no Z so how can I do it. The invention of the 0, yes its a new concept;).

D) I was wondering how to calculate current, besides using I = W/V. So the electrical charge of the compound I use * the velocity it is at = amps? Did I understand you correctly?

E) So basically its just the hall effect? That is all a MHD generator plays off of? The voltage is produced by the rating of the magnetic field and how strong the hall force is? The amps come from how fast the liquid is flowing?

F) Well from my understanding of the Internal Combustion Engine, the great loss of energy is heat. More energy is via heat than via work on the piston. So why not capitalize on this. The heat differential in a fluid will make it flow. The hotter fluid natural rises as the hotter it is the less dense it is. Flow is created. I understand that I can't get more energy out of the system than I put all I am trying to do is create a more efficient generator, based on the downfalls of current generators (which are powered by internal combustion engines).Oh and thank you for your response. It actually was very helpful. Your actually the first person I've talk with that knows or cares about this subject matter. It seems most people never looked into or know the principles behind MHD.

 

[Shelnutt2]

Ok I've gone through two designs and I'm on my second test model.

Here is my current issue.
Salt water doesn't work. at least I don't think. I got a reading of 1.5volts off of it, but I had bigger issues.

I have wires comeing out of my generator to measure the voltage on. Well on one set of wires, I was measuring the voltage and the other set I wasn't doing anythign with at the moment. My friend turns on the propane to heat up the center pipe, in order to start circulating the salt water. Then when I touched one set of wires, the other set devolps a leak. When I take the probes of the non leaking wires, the leak stops. I repeat 2 more times, to see what happening and check voltage. On the 3rd time, something different happened. When I touched it a small bit of smoke/steam/something white came from where I was measuring the voltage. One the other set of wires, a jet stream of water came flying out of the "small leak", enough that it hit the propane tourch and knocked it out.

We think electrolysis was taking place. What do you think? Seems logical to me, as when voltage and current is put to water it separates the atoms. I was measuring the voltage right after the magnets.

So I need a new solution. I tried a water/ammonia solution, it didn't work, I got 4 milvolts. I tried a salt + ammonium solution and got .5 milivolts..

So either something is wrong with my generator design, or I need a higher conductivity in my fluid..

 

[xez]

Hmm sorry for the delay in replying. I think you may be
right that you're seeing the saltwater decomposing under
electrolysis.

I'm not sure at the moment what other conventional
electrolytes to test; I'm sure there probably are
some others that would give you less troubles with
electrolytic dissociation and yet which have high
conductivity, ionic mobility, and low viscosity.

Besides essentially aqueous / alcohol electrolytes you
might consider other conductive fluids.

Clearly you're applying substantial heat, though I don't
know what temperature ranges are your limits.
Tin/Indium/Bismuth/etc. alloys melt at pretty low
temperatures, and you can get them either as
electrical lead-free solders, or casting metals for
arts/crafts, etc. I believe there are certain additives
and alloys that substantially lower the melting point,
and I probably have a chart of these somewhere that I
could look up for you if you'd like. I'm pretty sure they'd
often be in the 100C to 250C range.
I'm sure not all of the options would be suitable because
of viscosity, crusting, or other issues.

Looks like (not that I suggest lead bearing metals, but
I'll throw in miscellaneous ones for reference):
280F = 138°C = 42Sn-58Bi
356F = 66% Tin & 34% lead alloy.
360F = 65% Tin, 35% Lead & Antimony
361F = 183C = 63Sn-37Pb
450F = 232C = 95Sn-5Sb


Another option could be molten salt eutectic mixtures;
some of these have surprisingly low melting points,
though you'd have to find a combination that was
reasonable in corrosiveness, viscosity, mixture stability,
etc.
http://ras.material.tohoku.ac.jp/~molten/molten_eut_query1.php


Of course you could use a pure plasma, but then you'd
need at least somewhat of a moderate vacuum in the
apparatus if it's going to be bigger than tiny, and then
electromagnetic RF coils or low frequency AC/DC electrodes
to start the plasma, then appropriate electrode assemblies
to extract energy from the plasma, etc. It doesn't seem
like a good prospect for something that's supposed to
be efficient on a small/crude scale, though it's certainly
the kind of thing they believe they can make efficient
e.g. in tokamak fusion generators.

Conductive slurry mixtures of things like DENSE
metal powder or carbon powder mixes in a fluid that's
just present enough to cause the whole thing to exhibit
liquefaction behaviors but not sufficiently fluid such that
the fluid dramatically increases the resistivity of
conductor-to-conductor current flow could be interesting.
Especially if you had very small conductor particles, and
a fluid that changed density a lot with temperature over
the working range of temperatures in your device such
that it'd be a good pump fluid. Micron-size-realm carbon
black in something like ammonia or alcohol or so, perhaps.

I'll think more about electrolytes and such and let you
know if I think of anything good.

Part of the reason the industrial generators use high
temperatures is probably that the maximum efficiency
of any heat driven engine (as far as I know) is the
Carnot Efficiency from thermodynamics theory, and that
is:
maximum efficiency = (T_High_K - T_Low_K)/(T_Low_K)
where
T_High_K = Kelvin (absolute) temperature of the high
temperature input of energy to the engine.

T_Low_K = Kelvin (absolute) temperature of the low
temperature output of energy (e.g. exhaust or point
where your temperature differential is highest between
the hot input and cold side of the pump).

http://www.ac.wwu.edu/~vawter/PhysicsNet/Topics/ThermLaw2/CarnotMaxEff.html

 

[xez]

Oops that above contained a typo; this is the correct
Carnot heat engine efficiency formula. That's the
maximum possible, and real engines are usually quite
a bit less efficient.

maximum efficiency = (T_High_K - T_Low_K)/(T_High_K)
where
T_High_K = Kelvin (absolute) temperature of the high
temperature input of energy to the engine.

T_Low_K = Kelvin (absolute) temperature of the low
temperature output of energy (e.g. exhaust or point
where your temperature differential is highest between
the hot input and cold side of the pump).

So if T_Hot=700K, and T_Cold=300K,
Max Efficiency = (700K-300K)/700K = 57%, and of
course it gets lower as the temperature of T_Hot gets
lower and lower relative to T_Cold.

 

[etherpulse]

This topic is exactly something I am dealing with in an industrial process I am working on. In my case, we are trying to creating an MHD accelerator and decelerator (essentially a motor and a generator) and are having a problem deciding on what liquid metal or electrolyte to use. I’m an electrical engineer, but the chemistry involved here is out of my area of knowledge. We have tried several electrolytes, both aqueous and non-aqueous, and have real problems with degrading the electrode (as in it disappears in a minute or two) due to electrolysis. We are just starting to explore liquid metals and haven’t tried anything yet. My online research shows that there are many gallium compounds available, but data on their electrical characteristics is limited. The are used mostly in the semi-conductor industry it seems, and some are used are a replacement to mercury in thermometers and in liquid metal switches in thermostats, and similar applications. We have also discovered that mercury is reasonably cheap, and may be OK for a prototype, but we are going to need something less toxic for a practical system.

If you will post any progress you make in this area, I will do the same. Maybe we can come up with a good solution.

 

[Shelnutt2]

etherpulse,

I gladly will post up my information as I get it.

xez thanks for information again.

So here is where I am at. I need a solution still. I've tried IPA solution (91% IPA, 9% water) + NaCl solution. It was no good. I could not get a reading above 50 milivolts with it. I also tried an ammonium solution (what you buy at the store), its concentration is unknown, with NaCL and it turned out worse at 30 milivolts.

Two more things for my system I've realized. I need a solution that has a very low heat capacity. The lower the heat capacity the more current I will get.

AS for the hat I am putting into the system, http://www.bernzomatic.com/bernzomatic/consumer/jhtml/gasTypeComparisons.jhtml. I'm using a propane torch right now. If I'm doing the math right,
2,498 BTU/cubic foot. I'm only heating a .5 inch (1/24 foot) part of a 9/16 inch (0.046875 feet) OD pipe.
So then Volume= (Pi^2*r*h = 3.14^2 * 1/24th * .046875 = 0.01924 cubic feet of what I'm heating. 2498/.01924 = 129833 BTU's I'm using. So basically, my generator is completely inefficient.

I need to find a way to get the electrons moving, using less heat. Assuming I did the math right.

I think my first step is finding a solution. Now adding metal powder to IPA seems and interesting idea. I know I can get access to aluminum pellets. Could those be used to make aluminum powder and then added to my mix?

Gallium is the perfect solution but it way to expensive. etherpulse, you said you found that its easy to get a hold of cheap mercury. May I ask where? It seems any place I've seen it for order was either by the 100's of kg's or I had to have a teaching license and go through a bunch of security to order some, and it wasn't all the cheap either.

One chemist I talk to suggested the use of a salt bridge in order to avoid the problem of salt water, but the only problem with a salt bridge is that the I couldn't see a way that it would work, as water by itself is highly resistive, so I failed to see how I could get a current going, and volts, and I just couldn't see how it would work..Although the chemist couldn't grasp the concept of this MHD generator either..so..