windgen.org vawt discussion

Bernd

"BTW. i think my english must be terrible for you and the other users.  :("

It's a better than my german and better than a lot of english speaking people I have seen on the net

"You think so because the flux then went out of the magnets in only one direction ?
But using halbach arrays i'm not sure how height the "one direction" flux really is and they are mostly very expensive."

 Yes that was my thinking, maybe this wouldn't require the coils to be on 3 sides of the magnets 

Steve
"I have been thinking about how to make this generator easier to assemble and maybe more water proof.  I've come up with these ideas.

1) Use tall rectangular magnets, like 4cm tall x 2cm wide x 0.5 cm thick.

2) Stand the magnets on their ends, flat sides facing each other attach them to the rotating plate and support the top with some temporary structure. (tape?)

3) Tape around the edges of the magnets and fill the space between them with epoxy or fiberglass resin.  There is no need to have space between the magnets and this will add much strength.

4) Remove the temporary structure at the top

5) The shape of the tall magnets will let the coils go along the two long sides and the end, giving 83% coverage of the sides.  The coils can mate with the magnets from directly below and can be inserted and removed without changing the coils shapes.  The coils could even be potted (put in epoxy or resin) to keep the water off of them."

I think I understand what you mean but making this may be a lot easier said than done

Here is a rough image of what I think you mean, side section with magnet as blue
Cheers

Bernd wrote:

---

Hello Steve and thanks for this simulation.

"... The blade sizes were different, the radius was different and the wind speeds were probably different ..."

The radius was the same. I used 25cm too at torque test.
i guess the windspeed was around 3,5 m/s.
There is a high peak at 340 degree. This high torque peak is in reality definitely not present.
But it is nice to see how well the curves cover mostly.

Bernd

BTW. i think my english must be terrible for you and the other users.  :(

---

I went back and looked at the pressures and air speeds at that torque peak and I did not  see anything unusual.

The difference might be due to surface properties that I don't simulate.  My simulations are done with a laminar flow simulation so the surface is assumed to be very smooth.  Turbulence can still occur but the size is on the order of my mesh size, which is 2mm at the blade surface and it quickly expands.   Other simulations try to model the small turbulence at the surface.  I don't use this type of simulation because I have not found a good turbulence model that is as accurate as the laminar flow simulation.

It could also be that my front shape is different than yours.  I used NACA0018 up to the point (0.1284,0.0762) and scaled it to be 6.5 cm wide.

My wind speed was almost twice yours at 6.7 m/s.

I'm constructing the power curve now.  I have a few more points to get and then I'll post that to see if that agrees with your tests.

Your English is fine.  I can easily tell what you mean.  When I go to your board I have to use the translator for everything.

 

-- Edited by sjh7132 on Monday 11th of January 2010 11:47:33 PM

Hello Steve and thanks for this simulation.

"... The blade sizes were different, the radius was different and the wind speeds were probably different ..."

The radius was the same. I used 25cm too at torque test.
i guess the windspeed was around 3,5 m/s.
There is a high peak at 340 degree. This high torque peak is in reality definitely not present.
But it is nice to see how well the curves cover mostly.

Bernd

BTW. i think my english must be terrible for you and the other users.  :(

Caleb wrote:

---

I'd say those simulation results were remarkable!  Great job!

---

Thanks!  I should probably admit the fact that they values are just about the same is pure coincidence.  I was only interested in the shape of the graph.  The blade sizes were different, the radius was different and the wind speeds were probably different.  It would be interesting to see if they were supposed to be the same value or not.

 

I'd say those simulation results were remarkable!  Great job!

Here are the results of my 6 RPM (almost static) torque test.   My results have some noise in them because I had a lot of turbulence come  and go and I didn't average long enough, but I don't think the data is that bad.  I put Bernd's results below so you can compare.

There are some some common features in the graph, but they are also quite different.  Part of the reason is I simulated the blade at a 25cm radius, and Bernd's torque test was done at a smaller radius, with a larger blade.

When there are forces that aren't in the forward direction at the blade ends (as shown in the 3rd diagram), these forces do not affect the torque in proportion to the radius.  See the 4th diagram for an example.

The units on my graph are n*m per meter of blade length.

Bernd wrote:

---

just now i see that in your simulation picture the "Leitflaeche" (the bended blade) still goes a little bit deeper into the
Vorfluegel as expected. The radius into the inner circle should be on all sides the same.
I mean the distance between bend blade and inner circle. You can mount the bend blade so that it is on the line
with the end of the vorfluegel and the bend blade then doesnt dive into vorfluegel .
But i dont know whether this is really important.
I hope you understand what i mean.

Bernd

-- Edited by Bernd on Sunday 10th of January 2010 12:31:52 PM

---

Yes, I noticed the bad Leitflaeche position after I was 9 hours into the simulation.  I didn't think it was bad enough to restart.  I did correct the Leitflaeche position before I started my other tests.

 

just now i see that in your simulation picture the "Leitflaeche" (the bended blade) still goes a little bit deeper into the
Vorfluegel as expected. The radius into the inner circle should be on all sides the same.
I mean the distance between bend blade and inner circle. You can mount the bend blade so that it is on the line
with the end of the vorfluegel and the bend blade then doesnt dive into vorfluegel .
But i dont know whether this is really important.
I hope you understand what i mean.

Bernd

-- Edited by Bernd on Sunday 10th of January 2010 12:31:52 PM

hello Steve, i'm happy to see that you are simulating the c-rotor.

Yes the blade into my torque test is much bigger than into prototype.
The reason is that i used a blade of a previous test, where i tryed to optimize the blade dimensions for a single blade rotor.
The blade bend still match the radius of the rotor, at least somewhat. But this big blade was only a special test.
Better you use the normaly dimensions for simulation.

"... Your torque test blade seems to have the Leitflaeche extending into the Vorfluegel. 
It was an important feature of your prototype blade to have a space there.... "

You are right, but there is always a gap into the Vorfluegel, at all blades i ever made. Also into the blade at the torque test.
Maybe its hard to see and a little bit smaller than usefull, but it still has a gap too.

btw. i measured at teststand the power using the gap into the nose and a second test where is closed the gap.
I closed the gap with some thick paper so the air couldnt get on the other side through the gap.
The improvement using gap is in average about 22 %, at higher rpm more than at lower, that we can call really significant.
The gap is one of the important things of this rotortype.
Without gap the blade still works as a simple drag type blade.
Here the testing results with open (durchstroemt) and closed (geschlossen) gap using a single blade rotor.
"Drehmoment" = torque, "Leistung" = Power





".... The simi-circluar shape of the inside of the nose seems perfect for this.... "

I agree with you. I dont measured it until now, but it seems that the better air flow
through the half round inner circle increases the power.

P.S. I'm happy to see that you have register at my own board.

Bernd


-- Edited by Bernd on Sunday 10th of January 2010 11:45:35 AM

Bernd,

My C rotor torque test is still running, but I cheated and looked at the data I have so far.  We agree somewhat, by I seem to have an angular offset from your data. (even after I correct for our differences in coordinates.)

I noticed from your photo, that your torque measuring setup seems to have a much smaller radius, or the test blade is larger.  Is that true?   Does the radius of the Leitflaeche still match the radius of the turbine?

I think my numbers might be different because I ran one blade of your prototype, not of your torque measuring jig.  If you give me the measurements of that setup, I'll correct my model for the torque test.

The torques in my test are quite a bit better than in yours.  I think there is a real reasons for this:

Your torque test blade seems to have the Leitflaeche extending into the Vorfluegel.  It was an important feature of your prototype blade to have a space there.   When air hits the Leitflaeche, much of it is funneled forward towards the nose.  (because of the curve of the Leitflaeche, this imparts no torque on the system.)  When the air hits the nose it is curved through the semi-circle and out the other side, changing the air's direction 180 degrees.  That change of momentum imparts a forward force on the blade.  The simi-circluar shape of the inside of the nose seems perfect for this.







-- Edited by sjh7132 on Sunday 10th of January 2010 08:30:43 AM

stonebrain wrote:

---

However I think realworld testing is faster and more reliable(but not so easy neither).

---

I'll agree that the jury is still out on if real world testing is more reliable or not.  However Bernd has given me his drawings and he has a nice static torque graph.  I have made a computer model of his blade and am virtually repeating his static torque test.  (mine is actually spinning at 6 RPM, but I think that's close enough.)  We will see how close real life and simulation are.

I have the static test running and also 3 other tests of this turbine running at 100, 110 and 120 RPM.  The wind is 15 MPH, or 6.7 m/s.  It seems the turbine wants to turn much faster because this is only a TSR of 0.45.

I'm not sure if simulation is slower than real life.   Every 24 hours I can complete a test of the turbine at 4 different TSRs.  I can also look the airflow and pressure during those tests if I get unexpected results.

If it turns out that data is accurate this will be a nice tool, and probably get much more data in less time than testing in real life.

The question is, can we trust it?

Steve



-- Edited by sjh7132 on Sunday 10th of January 2010 05:37:47 AM

-- Edited by sjh7132 on Sunday 10th of January 2010 06:30:47 AM

Bernd wrote:

---

I also thought about the option to share the coilring into two parts for better dissasambling.

---

I have been thinking about how to make this generator easier to assemble and maybe more water proof.  I've come up with these ideas.

1) Use tall rectangular magnets, like 4cm tall x 2cm wide x 0.5 cm thick.

2) Stand the magnets on their ends, flat sides facing each other attach them to the rotating plate and support the top with some temporary structure. (tape?)

3) Tape around the edges of the magnets and fill the space between them with epoxy or fiberglass resin.  There is no need to have space between the magnets and this will add much strength.

4) Remove the temporary structure at the top

5) The shape of the tall magnets will let the coils go along the two long sides and the end, giving 83% coverage of the sides.  The coils can mate with the magnets from directly below and can be inserted and removed without changing the coils shapes.  The coils could even be potted (put in epoxy or resin) to keep the water off of them.

Just some ideas.

If my description isn't clear, I can draw some pictures.

 

Hello Stonebrain, i'm happy too,  to be here in this nice board :)

Fasten the blades stable is really important. But what you can see is only the prototype.
I mounted the blades intentionally with only one arm. So i can see where there are weak points.
I'm not a fan of "rotating tanks", so i first try to build the blades lightly.
To my surprise this prototype survived a big short storm last summer, a little "Tornado".
The rotor turned around with 800 - 1000 RPM and doesn't crash, unbelievable.
Normaly it turns between 30 to 150 RPM.

I'm also not a great fan of simulation, i think turning in reality gives better results and experiences.

"... About the rotor,I say it's a selfstarting darieus ..."

For me a darrieus is a lift type Rotor. The C-rotor is more a drag type with small tsr.
You can compare the c-rotor most to the savonius. 
The c-rotor has a great torque at stand and starts easy with only one blade. It seems it is really similar to the lenz turbine but.....
i hope a little bit better * :) *  Test will follow

Greetings

Bernd


-- Edited by Bernd on Saturday 9th of January 2010 09:17:24 PM

Bernd,
Happy to have you with us on the board.
That's the kind of stuff I like to see.Great pics too.Thanks!

I've bee staring at you C-coil generator for a while,trying to make a mental image of what's happening.
It's weird,that's why I like it.
Well,in fact the magnet is going through a 3/4 coil and is changing direction everytime the next magnet is passing,so there is no reason it doesn't work,and maybe not bad at all.
On the other hand the field of the neighbourmagnet will push the fieldlines in a short loop from N to S,on the same mag,and you may miss quite a bit of the fieldlines.

About the rotor,I say it's a selfstarting darieus.Like most vawtrotor designs,the only way to know how they work is to test them,except when you are a computer genius and find it out by simulation.However I think realworld testing is faster and more reliable(but not so easy neither).
The good thing about real,fullwind testing,is that you can see how it flies to pieces.So you can improve the construction.That's as important as things like efficiency.

I'm afraid your airfoil mounting is susceptible to vibrations/wobbling/oscillating.
But testing will show.
I think the mounting of the blade to the arm must be much more rigid.But allready you improved it compared to the first version.
If you like weird ideas,my idea is to make the blades from the mounting to the wingtop tapered(very narrow to the top).Thisway you will have more mass of the blade closer to the mounting,what make it more stable.
For weird things look at my 'sailwing darieus'(in the thread with the same name)

cheers,
stonebrain

-- Edited by stonebrain on Friday 8th of January 2010 11:57:18 PM

-- Edited by stonebrain on Friday 8th of January 2010 11:59:16 PM

turnymf wrote:

---

To me the magnets look like discs, maybe they are aligned  ns sn ns sn ns  to each other and thus pushing the flux out to the coils

---

Yes, that's exactly what's happening.  If you go back a few posts, Bernd posts a field plot.

 

Bernd wrote:

---

hello to all

i think most of you are searching for the disadvantages of the ringgenerator. :)

---

No we really are not trying to find things wrong with the ringgenerator.   It's just new to us so we have lots of questions and are trying to figure out how it works and how it compares to our plate type alternators.

Some advantages I see are it might give a higher frequency output for a slower turning shaft because you get more poles in the same space.

Also the physical construction of two spinning plates with a stationary stator in between on our alternators is always a challenge.  The ringgenerator does not have that problem.

During the construction of the ring generator, you have to deal with the repulsion of 2 or 3 magnets at a time.  We have to deal with the attraction of 16 magnets at once which is much more dangerous.

I think both designs have their uses.

 

I also thought about the option to share the coilring into two parts for better dissasambling.

" ... If a outer ring was iron would that cause cogging? ..."

If you want to put iron behind the coils, to increase the flux, then the iron must be rotate too.
If the iron would be fixed, you get eddy currents (i used the translator for "eddy" :) )

"... Maybe a Halbach array as a mag ring may make making the stator easier... "

You think so because the flux then went out of the magnets in only one direction ?
But using halbach arrays i'm not sure how height the "one direction" flux really is and they are mostly very expensive.

Bernd


-- Edited by Bernd on Friday 8th of January 2010 12:09:00 PM

-- Edited by Bernd on Friday 8th of January 2010 12:20:31 PM

I mentioned about the magnt life because I remember reading about storing magnets on steel plate

I think it may be possible to pot the coils if the stator was made of two or three pieces ie:cut on the radius
Then enclose the rotor in the stator

If a outer ring was iron would that cause cogging?
Maybe a Halbach array as a mag ring may make making the stator easier 

hello to all

i think most of you are searching for the disadvantages of the ringgenerator. :)

Of course there are those, but you search at the wrong end. The disadvantages are not beeing in bad power generating or less efficient.
The disadvantages are mechanically, for example if building the C-Coils, because they are much more difficult to build than
the easy ones find into dual rotor generators.
Also to find a way to mounten the magnets isnt easy, because you couldnt use metal for mounting and each magnet repulse
to the other with around 20 kg x 48 pieces = ~ 1 tonn (my big generator)
The same also applies for mounting the coils. You also couldnt use metall mounting, all things have to be antimagnetic.
Also the gap between magnet and c-coil, this generator has a gap at three sides of the magnet and not only one like other constructions.
So you must therefore ensure that all 3 gaps allways remain the same. Mechanicaly its more sophisticated, but all solvable.

I'm excited that it is possible, using this construction principle, to build a very powerfull generator only made from wood (or something comparable),
copperwire and magnets without necessity to use any piece of iron or other metal.

You mustnt use expensive equipment to build this generator, in simplest case just a saw and some bolts. :)

Bernd


-- Edited by Bernd on Friday 8th of January 2010 01:58:53 PM

hello turnymf

"... Would using magnets in this configuration reduce the life of them? ... "

No i dont think so. Using the magnets power doesnt reduce there lifetime, also the same for mounting like NS - SN - NS- SN - NS....


Bernd

Hello Ed, thanks for your comment

"... The original tests I did with it was with a single blade to see how it reacted throughout the full 360 degrees of rotation.  
Also to measure the stalled torque at any given point in rotation.   
It will actually start from the rearward position ( from a dead stop ) and fly directly into the wind and accellerate.  ... "

It sounds like a test of the c-rotor. :) I also measured the stalled torque at every 10 degree at stopped Blade.
Here is the outcome. "Drehmoment" = torque





"... Basically a linear alternator set in a rotary configuration.  ... "

I never seen it like that, but your are right. :)

"...  I found the closer the magets were to each other ( NN or SS ) the flux intensity was pushed out considerably, increasing the output. ..."

Yes thats right too. But for example half the space between the magnets didnt give you the double flux.
Also the induce of voltage has it highest point exactly between the magnets. If you reduce this space to much,
you reduce the space where voltage could be induced.
So you have to find the right point between number of magnets (magnet mass), flux enhancement and enough space to induces voltage.
currently i find it at the point where the space is double the magnet thickness.

" ... Unfortunately, it wasn't as potent as running the magents face to face as in the dual rotor arrangement.  ..."

Yes the flux height is less than a plate generator, but using this generator you have automatically much longer effective coil
legs (3 times more) and more space to mount copper.
Another point is the that half of the flux from the magnets in plate generator goes through the iron plate an is not used to iduces voltage to coils.
Just only one pole per magnet is used, the flux from the other one is going useless through the iron.
Also the ratio between effective copper which induces voltage, and copper which doesnt induces voltage is much better at the ringgenrator.

At the ringgenerator all flux (from 3 of 4 sides of the magnet) goes through the coil legs.

"... The ones I built worked quite well but they were rather expensive because of the magnet count needed to produce the same output as the dual rotor. ..."

My experiences are that you have to use nearlly the same qantity of magnet mass, maybe a little bit more at the ringgenerator.
A friend of mine plans and build iron plate dualrotors.
Acctually he build a 6 Kw modell. Often we calculate the same model as ring- oder dualrotor. He calculates the dual and i'm the ringgenerator.
Mostly we get similar results for price, size and diameter.

Some words to the "efficient".
There are two kinds of efficient. The first is how big a generator must be to put out a certain amount of power.
Here is a car alternator one of the best, he puts out much power and is very small,
but he also has a bad power coefficient of around 50%.

The second kind of efficient is how good and effectiv a alternator converts mechanically energie into electrical.
This"efficient" is the important one. Here both, dual and ringgenrator, can be go up to over 95% at part load.
I will design the coils of my big one to be at 90% at full load and clearly above 90% at part load,
exactly as much as a good constructed dualrotor can deliver.


Bernd


-- Edited by Bernd on Friday 8th of January 2010 09:58:34 AM

-- Edited by Bernd on Friday 8th of January 2010 12:22:16 PM

I love your ringgenerator Bernd
Turbine's nice too
Looking forward to seeing the new stator on the new unit

Would using magnets in this configuration reduce the life of them?
Cheers

Ed, I moved this conversation to 'Ed's Mystery Circuits'.  If you are willing to keep giving clues, we can continue there.  Otherwise I'll just wait.

sjh7132 wrote:

---

Nice you hear from you Ed.

Your 500 amps from 12 watts
 sounds like it would require superconductors.

The rest of what you wrote sounds like the 'free energy' guys on youtube got you you. :-)  It is true that the current every where in a series circuit is the same, but you use voltage, and power is current * voltage so each device uses power.

I look forward to seeing what you are up to.

Steve

---

Set up a light bulb on a battery and measure it on both input and output...  I'll overlook the free energy comment, ... for now ;O).   Super conductor?  Maybe...  Look into the series LC circuits for a better clue, when brought to resonance the impedance drops to zero which forms a psudo super conductor ( short circuit ).   Also, a parallel LC circuit when brought to resonance becomes total impedance ( open circuit ).   A parallel LC only needs the initial input to continue, the energy in the system is simply recycled over and over again.   Still needs input to compensate for losses but far less than you might expect. 

 

Nice you hear from you Ed.

Your 500 amps from 12 watts sounds like it would require superconductors.

The rest of what you wrote sounds like the 'free energy' guys on youtube got you you. :-)  It is true that the current every where in a series circuit is the same, but you use voltage, and power is current * voltage so each device uses power.

I look forward to seeing what you are up to.

Steve

Nice work Bernd .    The Lenz2 runs very nicely with one blade.   The original tests I did with it was with a single blade to see how it reacted throughout the full 360 degrees of rotation.   Also to measure the stalled torque at any given point in rotation.    It will actually start from the rearward position ( from a dead stop ) and fly directly into the wind and accellerate.  

Your alternator is an interesting design as well.   Basically a linear alternator set in a rotary configuration.   I found the closer the magets were to each other ( NN or SS ) the flux intensity was pushed out considerably, increasing the output.    Unfortunately, it wasn't as potent as running the magents face to face as in the dual rotor arrangement.   The ones I built worked quite well but they were rather expensive because of the magnet count needed to produce the same output as the dual rotor.   Very interesting project though.  



-- Edited by windstuffnow on Friday 8th of January 2010 07:08:40 PM

"my" nose is similar to NACA 0018 i think. :)
about the first 20% of the NACA 0018 are looking like "my nose". :)
But if its to difficult, simulate with the half tube.

Bernd

-- Edited by Bernd on Thursday 7th of January 2010 10:27:49 PM

Bernd wrote:

---

Steve have you ever simulate the c-rotor ?

Bernd

---

No, I'm working on it, but your nose shape is giving me problems.  Can you suggest a NACA number and scale to get me something close?  A few coordinates would also work.

 

-- Edited by sjh7132 on Thursday 7th of January 2010 10:05:40 PM

Steve have you ever simulate the c-rotor ?

Bernd

Bernd wrote:

---

i have two ideas

first is, that a "sharp nose" also could be increase the power of a lenz turbine ?

---

I have tried a sharper nose (elliptical) on the Lenz and it doesn't work as well.   The cylindrical front end seems to be very important for generating 'lift' that pulls the blade forward.

Going up wind the blunt end does build up quite a bit of pressure, but the lift at other places in the rotation more than compensate for it.

I have been thinking about having an object ahead of the Lenz blade, but not connected to split the air stream.  I've tried one configuration so far and it didn't help the power.  I might have used the wrong shape or not had it in the correct position.

[See attached images]

i have two ideas

first is, that a "sharp nose" maybe also could increase the power of a lenz turbine ?

second is, that i should measure a lenz Tubine vs. the C-rotor on my testbench, especially because they are build similar.
I can use the same blade (straight or bend) an the same "front" and just have to mount it different and give the lenz the specified angle.

also i can test the "sharp nose" vs. the "normal nose" of the lenz. i think that would be a nice and interesting test

I wonder if the lenz would be also runing with only one blade like the c-rotor does.
Thats one of the properties of the c-rotor which had impressed me.
It starts by itself and runs around with only one blade.
(So i have to correct what i have written before, its not only a drag profile. :) )

Bernd


-- Edited by Bernd on Thursday 7th of January 2010 09:48:23 PM

"... The absolute numbers are not  correct but you can see if something is better or worse than something else "

Exactly ! 
To get results "in real life" here in Germany these days its much too cold.
Today we have -17 Degree Celsius. :)

btw. where is this discussion board located ?

Bernd

Bernd wrote:

---

Because of my "simple wind", made by two 40cm table fans, i'm only
able to compare two results, like before and after changing something.

---

You testing is like my simulations.  The absolute numbers are not  correct but you can see if something is better or worse than something else.

Because of my "simple wind", made by two 40cm table fans, i'm only
able to compare two results, like before and after changing something.
The wind from the fans doesnt hit the blades completly, but to compare soemthing it is pretty good.
To get exactly results of power coefficient and peak of power you have to use "better wind".
The mechanical power during measurement was from 0,5 to 1,2 Watts, depending on the tested blades.

Bernd

Bernd wrote:

---

With the simple Break (it is intentionally a simple rope brake) i am able to measure the torque an together with the
RPM i can calculate the mechanical Power. So i can see immediately if a improvement works or not. :)

---

What have you seen for power so far?

 

Not really a NACA profile, we just use a "sharp nose" in order to reduce the air resistance. It still remains a drag profile.

".... Danka  (soon I'll be ready for your board.  ha ha).... "
Well, thats the beginning. :)

Bernd


-- Edited by Bernd on Thursday 7th of January 2010 06:13:33 PM

Thank you for the data and drawings.  Which NACA airfoil nose did you use?   Or is it just similar to that shape?

This sounds like this might be a problem for my genetic algorithms to solve.

Variables I see are:
1) Vorfleugel size and shape
2) Vorfleugel angle
3) Leitflaeche length
4) Number of blades (solidity)

Vorfluegel (half tube C) translates to 'slat'   (That translation doesn't seem to fit well)

Leitflaeche (the bent blade) translates to 'baffle'   (That makes more sense)


Danka  (soon I'll be ready for your board.  ha ha)

hello electrondady_1

".... it would appear  that on the inside dia.of your magnet ring ,
you have used a spaceing that is approximately twice the magnet thickness,
is that relationship/ratio something you have experimented with?..."

Yes, i tryed some different spaces between the magnets and i found that at a space at the outer diameter, double the thickness of
the magnet, would be a good compromise between power and number of needed magnets and the price.
But it must not be the absolutly best ratio.

Bernd

hello Steve,
my prototype C-Rotor has diameter 50cm, so radius 25cm is correct.
The length of the bended Blade L (we named it "Leitflaeche") is 13cm.
The diameter of the half tube C (we named it "Vorfluegel") is 6,5cm.
All the dimensions i got during empirical trying on my testbench.



All dimensions give back the current state of knowledge.
Each of the 3 Blades (on the upper picture only two) have a total cord length of ca. 18cm.
So the solidity (i prefer the percentage value) is around 35 %.
On the testbench i could determine that a half round "Vorfluegel" has not been optimally.
It tryed some other shapes an i found that a aerodynamic shape of the "Vorfluegel" is much better for a good power coefficient than the half round tube.
So today we recommend a shape like the "nose" of a NACA profile.
Also the bended shape of the "Leitflaeche" is a result of trying at the testbench.
We found that a bended shape, with aerodynamic "nose" increases the power coefficient significant. Now some pictures from my prototype:







All dimensions have not yet been determined exactly.
You are invited to participate in further development. :)

Bernd


-- Edited by Bernd on Thursday 7th of January 2010 06:14:56 PM

electrondady1 wrote:

---

just as in a conventional air core coil,  each leg is intersecting flux of opposite direction.


your c shaped coils still intersects the flux lines at 90 degrees .
so a current is induced.

not all the flux lines can be utilized
there is a limit as to how much of the magnet the coil can encompass.

i can not guess as to the overall electrical efficiency but aesthetically i think the design has real possibilities

in regards to magnet spaceing ,
it would appear  that on the inside dia.of your magnet ring ,
you have used a spaceing that is approximately twice the magnet thickness,
is that relationship/ratio something you have experimented with?

where's my pencil , where's my calculator!

---

I was wondering some of the same things edaddy.  This makes much more sense to me now, and even thinking of it the classical way of flux through the loop works.  The most flux comes when the coil is between magnets, the most voltage as it transitions from magnet to space between.

It seems that putting the magnets closer would squash the sphere more and push the flux out more.  Also that slight tilt from being in a circle helps force more to the outside where the coil is.    I wonder what would happen if you used those two effects a little more.  Maybe shrink the diameter of the alt so the magnets are closer and the ratio of inner distance to outer distance is more.

Really it's pretty efficient.  The only flux lines that can't be used are those that skirt from N to S between the coil and the magnet.  Kind of opposite of our alts where we miss the lines that take the widest path.

All kinds of fun things to think about.

there is a convention that proposes that flux leaves the magnet from the north pole and travels to the south pole.

and left to themselves, the flux lines of a magnet will form a sphere

so in placeing magnets with opposing pole as you have done,  the sphere is deformed into a flattened sphere or dounut shape

each magnets flux lines are traveling in oposite directions

just as in a conventional air core coil,  each leg is intersecting flux of opposite direction.


your c shaped coils still intersects the flux lines at 90 degrees .
so a current is induced.

not all the flux lines can be utilized
there is a limit as to how much of the magnet the coil can encompass.

i can not guess as to the overall electrical efficiency but aesthetically i think the design has real possibilities

in regards to magnet spaceing ,
it would appear  that on the inside dia.of your magnet ring ,
you have used a spaceing that is approximately twice the magnet thickness,
is that relationship/ratio something you have experimented with?

where's my pencil , where's my calculator!
















-- Edited by electrondady1 on Thursday 7th of January 2010 02:36:52 PM

-- Edited by electrondady1 on Thursday 7th of January 2010 02:40:41 PM

-- Edited by electrondady1 on Thursday 7th of January 2010 02:48:35 PM

-- Edited by electrondady1 on Thursday 7th of January 2010 03:00:26 PM

Bernd, I know we talked about the dimensions of your c-rotor through you-tube, but I don't think I understand them correctly.

I believe R is 25 cm.
What are D and L?

Is the shape correct?

Thanks

Ah, you did say the magnets were NS SN NS SN but I was thinking NS NS NS NS.
Your field plot makes that clear.  This pushes the field lines way out so your coils can intersect them.

This makes sense now.

Thanks for taking the time to explain.

" ...but that's because Voltage  depends on the CHANGE in flux going through the coil... "

As i told before, thats the way with iron coils. There the flux goes through the iron core and not through the copper wire.

The "changing in flux" is unimportant for the induction directly into copper wire.
Only the height of flux is important and how fast it move across the wire.

You can generate DC Voltage when you take a wire through a constant magnetic field as long as you take it in constant direction.
The DC polarity only changes when the direction of move is changing or when the direction of the magnetic field is changing.

Its completly different from the well known induction into iron coils !!

" ... Those are some nice magnets in your photo.  What is their size?  Do you know how strong the surface field is?  N40, N50? ... "

the dimension of the magnets are : 40 x 40 x 15mm
They are made from N52 Material.
The expected benchmarks of this generator:
Output Power from 20 RPM (hope 15 RPM) up to 500 RPM
at 500 RPM it should be around 5 Kw


Bernd


-- Edited by Bernd on Thursday 7th of January 2010 12:44:05 AM

Those are some nice magnets in your photo.  What is their size?  Do you know how strong the surface field is?  N40, N50?

" .... Until the field lines are crossing a wire, the flux through the loop is not changing .... "

Yes thats right, but it must not be changing !! Only the crossing (as soon as possible) through the coils leg, directly through
the copper wire, is inducting the voltage. It must not be a loop ( first faraday induction phenomen)
That is very different to the induction into iron coils !!

Bernd



-- Edited by Bernd on Wednesday 6th of January 2010 11:54:48 PM

-- Edited by Bernd on Wednesday 6th of January 2010 11:57:47 PM

Bernd wrote:

---

The peak of voltage induction occurs when the magnetic flux accts directly on the coil leg,  not on the coil hole !!

---

That is true, but that's because Voltage  depends on the CHANGE in flux going through the coil.  Until the field lines are crossing a wire, the flux through the loop is not changing.   I think we probably agree but are thinking about it in two different ways.

I still think that most of the flux between your magnets is not being used, and I wonder if tall thin magnet might be better since it would cause the field lines to bend out more and intersect your wires.  (Or maybe a ring magnet)

When I get some time, I will do some experiments and maybe put them on YouTube, or at least post photos here.

I believe the equation V = -N d_flux/dt still applies to your alternator, where d_flux is the change in total flux going through the loop and N is the number of turns.

(Sorry I don't know how to put Greek letters on this page.)

thank you for your nice comments

i really understand that most of you must be wondering how this type of curios generator should be working.
As i heared first time of this construction, i thought the patent holder would be crazy. :)

To understand how it is working we must know that there are two types of magnetic induction.
The second type of induction is the type everyone knows. Its named " "first Faraday induction phenomenon "
This principle is used in every iron electric motor oder generator. The height of the induced voltage depends
on the height of the flux in the iron coil core and how fast the flux changes and also how much wiring is used
and how big the coil is.

" The first "faraday induction phenomenon " is the one wich is rather unknown. This type is important for ironless coils like the one in
my generator or the plate/disk generators.
The height of the induted voltage depends on the height of the flux which acts on the coil legs and how fast the flux pathes over the coil legs.
Also its important how much copper wire length is present.
The flux directly affects to the coil legs. You can read it if you search for the "Lorentz effect".
This type of induction is the one "we use" into ironless generators.
It is not very important how much flux you can drift through the hole of the aircoil, although many people think so.
The peak of voltage induction occurs when the magnetic flux accts directly on the coil leg,  not on the coil hole !!

The ringgenerator has the same direction of flux through the coils like the plate generator.
I will show you some pictures wich can explain it.

"...........It looks like there is room between each leg to fit. It would add more complexity to the assembly but it is doable...."

Yes thats right. My prototype has much unused room between the coil legs.
But he was only build to take expirences and to get measurements.
If i would fill the unused space with copper too, the results would become much better.

" ...... The voltage produced is highly dependent on how close the wires get to the magnet.  In our plate alts it only depends on the turns of the loop and flux...."

Yes thats right. There ore some different things you must note.

" .....The actual area of the C has nothing to do with the voltage (or power) created.  It probably has more to do with the length of the C.... "

Completly correct (if i understand you well ) How much voltage inducts depents on the length of copperwire into the coil legs
which is taken by the flux of the magnet.

Fotos:

The first foto shows "the next Generation" of my ringgenerator.
Its not me, its a young friend of mine. Bernd himself is 46 :)




The following picture shows how the flux is floating.
There are much more really detailed information on my discussion board.
Also very accurate calculations. You only have to learn german. :)
Or ask me. :)



Bernd


-- Edited by Bernd on Wednesday 6th of January 2010 11:37:29 PM

-- Edited by Bernd on Wednesday 6th of January 2010 11:48:08 PM

Bernd
It's a fascinating design. I was wondering if you have tried adding a second set of coils to make it a two phase alternator. It looks like there is room between each leg to fit. It would add more complexity to the assembly but it is doable.

sjh7132 wrote:

---

---

I've seen claims that the C-rotor beats the lenz2.

-- Edited by sjh7132 on Wednesday 6th of January 2010 02:06:37 AM

 

---

I for one would be curious to see what the C-rotor looks like on your simulation software.