Regarding Willib's comment, why not? Sure, it would be kind of a waste of that big diameter, but if you didn't have many magnets on hand or couldn't really afford that many, but had alot of wire for coils, you could try something like a dual rotor with 16 poles and 24 coils, (so basically a 32/24 set-up, but with every second magnet removed). It wouldn't give you true 3 phase, as only 2 out of 3 phases would be seeing some magnet action at a given time, but the advantage of having double the coils should make up for it. Plus, the linear speed of the magnets is «simulating» a higher RPM with a smaller diameter, so, more voltage should be induced in the coils. You can always rectify them separatly anyways, and I say you should have quite a nice output increase, because you have double the coils, with magnets passing faster over all of them. So, unless you absolutely must have a perfect 3 phase waveform...

Of course, if your going to mount that at the top of a tall tower, because of the bigger rotor you might have to make some parts stronger then you otherwise would have needed, but it shouldn't be a killer. Once you get the cash, you can always come back and add some more magnets if you want. It's unconventional, but I still say it could be a way to get some more power...

if yes, my reasoning parallels Electrondady's.  unless both legs of the coil see flux, you'll be dealing with more resistance per flux line.  even though velocity is up, average flux remains the same, and the coils spend time with zero flux.

with a sharper peak and only one coil leg under a magnet at a time,  the areas under the voltage-current curves are smaller.

As for the comments about needing 2 coils, one under a N pole while the other is under a S pole, I have to ask you how is it that it has been recommended many times on this board that people should test 1 coil to see what kind of output they get when trying to determiine number of turns?  

You don't need 2 coils operating at the same time - if I remember correctly the hampster generator on the otherpower web site uses multiple magnets and one coil - correct me if Im wrong.  Don't  some magnetos have  only one coil?  Is it possible to build a 4 magnet 3 coil 3 phase alternator in which only 1 coil is directly under a single magnet at a time?  In fact  I believe you can generate a pulse that could be rectified into usable DC power even if you onlu had 1 magnet (either N or S) passing past a coil.

I assure you one coil works ok and in fact a decent looking sine wave is produced if the magnets are not spaced too far apart.

Interesting comments.

post #11; johnlm said " In fact  I believe you can generate a pulse that could be rectified into usable DC power even if you onlu had 1 magnet (either N or S)passing past a coil."
correct... see;
1 coil, 1 magnet, 1 AC wave
and 1 magnet 1 coil

Well my geuss is everything went down hill. You may have 2X the speed at 2X the diameter, but since you did not use 2x the mags nor larger coils or larger magnets, you have 2X the dead space between magnets on the larger rotor. So did you really increase anything?

So 1 magnet may fly past the coil 2x as fast, then nothing (but should be a magnet) then another magnet then nothing. The poles are not alternating N S twice as fast on the coils because they are 2X further apart. Instead of N S N S,  D represents DEAD SPACE, do you now have N D S D N D S??? My geuss is yes. So if you used 12 magnets, you have 6 N poles, 6 S poles, and 12 Dead spaces. Where once at the smaller dia. you might have had a 1/4" between magnets, I bet you now have about 3/4" between them?

You do not have both a N and S pole over the same coil at the same time either, but you probably did when it was a smaller diameter rotor.

If the same size and same number of coils are used on the larger diameter rotor, again dead space in between them. Same as the magnets.

If using the same magnets but wanting to increase the speed of the poles passing the coils using a larger diameter rotor you would need to use more magnets at the same spacing between them as you would have had on the smaller rotor. In this case with the same number of same sized coils you would be alternating the poles 2X as fast at the larger dia. because you have 2X the poles. Perhaps 1/3 or 1/2 would be unused at any given time because of too few coils, but they would be the same spacing, N and S both over a coils legs same as on the smaller rotor. No dead space between magnets.

So my geuss is you got less usable power because of the large gap between magnets, poles not properly lined up on the coils, etc...

Perhaps the bolt in the center will make a difference.

How many Watts was the smaller rotor giving and how many watts from the larger is what I am wondering at the same RPM.
.
nothing to lose

Which is, quoting from my feeble memory:

Number of poles x revolutions per second x flux in tesla x area of 1 magnet in square meters x number of turns in 1 coil x number of coils

You will notice there is no mention of 'diameter' in the formula.

The diameter of the machine is not directly relevant in this equation.

I usually use the equation in this way:

If you know the diameter of the proposed machine, you can predict how fast it will turn, and the approximate power output.

Fiddle around with the formula, change the magnet area etc. until the output looks close.

Then have a look and see what magnets are available here in the 'real world'.

Then based on those magnets, figure out the smallest diameter that will fit them with proper spacing.

Flux, I agree, I have seen the waveform on a scope, with magnets quite far apart, and the same number of poles the sine wave looks 'peakier'.

All of this a long way of saying I would vote for the power going down.

Now if you used the larger diameter to put more magnets on the rotor, more power would result.

Which agrees with the formula (increasing the number of poles)

But use more magnets or copper in the increased space and you will get more power out.

No such thing as a free lunch!!!!

Just some of my thoughts :P

First of all I thought I had made everything pretty clear in the description of the experiment; however I did not directly state until several replys had been posted that in both cases the coil was exactly the same coil in both setups.  Some folks thought maybe their were different coils.  And in both cases the number of magnets and the spacing to the coil, and the RPM remained the same.  And I only asked what everyone thought the open circuit output voltage was.

Several folks got to the real heart of the matter, especially Monte when he cited (in words) the basic formula of V=NARPB/2 where
N= Number of turns
A= Area inclosed in the coils (in Meters^2)
R= Revolutions per second
P= number of poles
B= Field strength (in Teslas)

No where in this formula is there any factor for radius or diameter which would determine velocity of the magnets. And Revolutions per second (frequency) is only revolutions per second - a magnet mounted farther out on a plate  verses one mounted closer to the axis of rotation rotating at one revolution per second still makes only one revolution per second - there is no increase in frequency by mounting it farther out on the radius. The frequency is totally dictated by revolutions per second and P the number of poles.  There is however an increase in velocity of the magnet moving past the coil.  There would be an increase in frequency and output voltage however if the free space out there were used to add more magnets - whether you added more coils to go along with the added magnets or not. But I pointed out I did not add extra magnets in the spaces, and I only used one coil. If I had used two similar coils properly positioned with one under a S pole whebn the other was under a N pole, the only thing that would have been different in the above formula is N.  The number of turns would have been double and the output voltage would have been double what I measured in both setups.  A (single) magnet moving faster past a coil does not increase frequency, it probably does however increase the peak magnitude of the output voltage, but the width of the pulse is narrower due to less time the magnetic lines are cutting across the coil wires.  Same area under the curve as a corresponding lower peak magnitude but wider pulse.

I was careful to measure and state that the output voltage was measured in RMS. I didn't say this but it was measured on a Hewlett Packard RMS lab grade meter. RMS is the true EMF (elctromotive force - another term for voltage).  In electrical engineering one can prove using calculus that a narrow high peak pulse looking waveform with dead time between the positive and negative transitions (characteristic of what one would get with the same number of magnets spread apart more, but faster velocity of the magnet past the coil) can have the same RMS value as a true sine wave looking waveform with a lower peak measurement. The amount of power you get out of your alternator whether its measured as AC power or rectified DC power is contingent upon RMS voltage.  

Alot of good practical hands on experience and knowledge came out when folks started talking about getting the coils sized, configured and placed properly for various magnet sizes and shapes; but the purpose of this experiment was to see if there was any validity to statements such as "I have 24 magnets (12 for each magnet rotor plate), and 9 coils and Im going to use a larger (than needed to fit them in?) diameter plate to increase the velocity of the magnets to get more power".  A few folks, including Dan B mentioned (my interpretation here) that the most effecient design is one that uses a magnet plate sized to be able to get the magnets to properly line up over the coils with the coils usually sitting as close together as will comfortably fit given their size dictated by wire size, number of turns and shape.
My statement here: spreading things out by increasing the diameter of the whole system beyond that gains you nothing in output RMS voltage or power.

The answer is #3.  The RMS output voltages were (within measurement error) the same in both setups.

Please forgive any typos or spelling errors.

A very interesting subject.

I have read everything to this point including your original question and answers. But I got to admit that I'm still stumped. I'm probably over looking something really simple, but I'm still confused.

I don't know of a generator or alternator.... especially a pm generator that doesn't increase over all in power output the faster you turn it, so long as it doesn't fly apart. (In other words increase in output wattage)

In other words... it doesn't matter whether the pm alternator have 1 coil and 1 magnet, or 200 coils and 200 magnets... regardless of internal arrangements, Every generator or alternator I have ever owned will produce more actual output voltage and amperage the faster I turn it.

When you take any pm generator or alternator and spin it faster... Isn't the only thing your actually doing is increasing the velocity in speed that the pm flux passes over the field coils ?

Output power (volts*Amps into a load) increases with;
an increase in rpm,
an increase in the number of magnets,
an increase in the number of turns of wire in a coil.

Any one of which increase the AC frequency.

Output can also be increased with larger rotors together with larger magnets and increased wire size.

what is the price? maybe a new blades for us newbies?