Design of a 1 kW axial flux alternator

[samacda1000]

I am looking for the design of a 1 kW axial flux alternator. I am familiar with Scoraig's design; however, was looking for something that had a higher capacity. Could someone help with this?

[fabricator]

In most areas with average wind speeds of 12mph, the otherpower 17' machine would be a pretty solid 1kw machine, you might want to do a search for Chris Olsons machine with a geared down alternator, that machine puts out a lot of power with less copper and cheaper magnets.

[windvision]

With a little research on this site, you will find many articles on machines that are 1 KW or larger. Enjoy the read.

[GoVertical]

Greetings, winter killed my testing of a new vertically stacked rotor and stator design. Basically it is three stators and four rotors stacked on the center axis. The outer rotors are metal backed and the two inner rotors the magnets are mounted on nonmagnetic material. It is away to increase coil count and use less magnets. Before the weather stopped testing I was able to verify that it does work. For a 8 inch diameter PMA it has 12 coils per phase using ¾ by 1 inch cylinder magnets. The increased coil count results in a lower RPM to achieve a charging voltage. I am looking forward to spring so I can continue testing. 

Photos of a 6 diameter 8 coils per phase

[GoVertical]

photo of 8 inch diameter PMA using PVC for stator spacers and case.

[gsw999]

Nice looking stators!! keep us posted, Thanks

[scoraigwind]

The power output depends on size of alternator and on rpm.  The larger ones in my recipe book can comfortably produce over 1 kW at about 200 rpm and could produce a lot more at higher speeds if you want to run your blades that fast.  You can also gear the rpm up like Chris does and get a lot more power with better efficiency if you want to get into that extra mechanical stuff.

Stacking alternators with multiple rotors and stators is less efficient use of materials than building one large diameter machine where all of the magnets interact with all of the coils.

It's pretty simple to build altenators that can kick out several kW by simply increasing the number of magnets and/or using larger magnets.  I have built some around 5 kW. 

 I often design larger alternators for people who send me emails.  I don't charge a fee for this.  I do it for fun and to help people.  I appreciate feedback but in many cases I don't get any because they put the alternators on VAWTS that presumably never work.  I can't understand why people do this when they can get good results with HAWTs.

[artv]

Hi GoVertical......I am thinking along the same lines ,only slightly different. Iwas thinking having a center disk of magnets ,with both pole faces exposed and coils on both sides of the magnet disc, laminated cores behind the coils to provide flux path, this way you could double the coil count for any amount of magnets.                                                       Hugh .....I'm sure you or many of the other members have seen this before is there any reason why it's not a current design??.........thanx artv

[Flux]

"I am thinking along the same lines ,only slightly different. I was thinking having a center disk of magnets ,with both pole faces exposed and coils on both sides of the magnet disc, laminated cores behind the coils to provide flux path, this way you could double the coil count for any amount of magnets.   " 

This is a common idea and normally it has no benefit. If you split the flux distribution with an air gap on either side you halve the flux for a given gap length. You need more wire, with higher resistance to do the same job. in addition you add two more mechanical gaps that constitute useless air gaps that the coils can't use.

Your proposal does have some merit for an axial machine as it to some extent balances the mechanical forces that can be a pain with a single rotor axial with a backing core. Even so you will still have large forces during assembly and it is a bit of a challenge holding magnets in some sort of non magnetic rotor.

I agree absolutely with Hugh, one large alternator uses materials far better than stacking multiple alternators on one shaft. I think this applies even more to VAWTs where low speed means large diameter alternators. At low speed you can go very large indeed in diameter without too many complications except that you may need to reinforce the discs to get stability if you go several feet in diameter.

I love to see people keep reinventing the wheel.

Flux

[GoVertical]

Greetings, I really like the ChrisOlson solution for more power at low RPM's.  Yes, a vertically stacked stator design is just a experiment with incomplete test results.  It does  use less magnets for a increased coil count and can be fabricated in a small footprint when compared with established fabrication techniques. I am working on a VAWT, but the proposed PMA design could easily be adapted for HAWT.  There are designs with 3 phase cog stator that laminate each phase in a single stator but they are said to be difficult to fabricate and increase the air gap.  By casting each cog phase coil in a vertically stacked stator would simplify fabrication and smaller magnets could be used. It is just a proposal.  The world needs renewable power why not investigate alternative designs. artv, can you provide a drawing of your design?     

[GoVertical]

Stacking alternators with multiple rotors and stators is less efficient use of materials than building one large diameter machine where all of the magnets interact with all of the coils.

Greetings,  magnetic lines of flux extend out of each end of the magnet. The vertical stacked approach is a way of adding more coils while using less magnets. Can you please provide more evidence so I can better understand what you are saying.

[Flux]

I am not Scoraigwind but I will give you my take on this.

A magnet can only give a certain flux density into a magnetic circuit. That circuit is basically an iron circuit ( or the magnet itself) and an air gap. For an example if we took one of the common magnet shapes and built a single rotor machine with a laminated core to close the flux path, your coils will be in the air gap formed on one side of your magnets.

If you modify the arrangement to use both sides of the magnet then you will have 2 air gaps in series. If the gaps are the same as for the single rotor case the flux density will be lower. To get the same you need to halve each gap.

You can work with one big gap or two smaller ones but the total; flux available from the magnet is the same ( you can't have your cake and eat it).

I fail to see what the number of coils has to do with the argument. What matters is the total flux and the number of turns linked. if the flux is the same you get the same emf with the same number of turns. It makes absolutely no difference if the total turns in one phase is in N coils or 2N coils.

In either case you can adjust the total turns in the coils to get exactly the same emf.

To go even further you can always get the emf you want with any magnet set up at any speed, what you need for power is to be able to generate that emf in a coil of low enough resistance to produce the current you want.

By splitting the flux between both ends of the magnet you will need more coils to get the same emf but you will add more copper length and increase the resistance.

In the ideal case there is no net gain or loss but in real life you need to consider leakage flux that doesn't link your windings and you also have to remember that you need a mechanical gap for it to rotate in. This for safety needs to be about the same with one stator or two and it is wasted in that you have no coils linking this. By creating two extra useless gaps you will inevitably be able to use less of the flux with your two ended design. This and the added flux leakage will work against you and it will be more difficult to get the result that way than the conventional way with an air gap on one side of the magnet only.

Unless you can grasp the concept of magnetic circuits and working the magnet at its BH max point you will struggle to follow this but I hope it gives you something to think about.

Don't get me wrong, your idea will work fine and you can build a good alternator that way but you will get lower efficiency from your magnets and you complicate the mechanical design.

Flux

[artv]

Thanx Flux ,I didn't realize the flux could be divided like that.              GoVertical I don't know how to draw pictures sorry. I was going to use stacked magnets to increase the flux stenght, 5 per stack ,then take plywood cut holes in it to hold the stacks ,but leave both poles of the stack exposed. I have lots of ferrite mags but no neo's, thought I could use them but now I'm not so sure.............let us know how your build turns out........artv

[GoVertical]

  Greetings artv, here is possible method for rotor fabrication without using glue. It may help. I hope to see some photos as your project progresses. Best regards

[scoraigwind]

I agree with Flux.  You need to think of the magnets as rather similar to batteries, and the air gaps as rather like light bulbs.  There's no free lunch whereby you can get the same current in two light bulbs in series - they will both be dim.  You also have to put more batteries into the circuit.  You need to have just as many batteries to make those bulbs glow brightly. So you can't sneak extra coils into the magnetic circuit without losing out on flux density in the larger (total) air gaps.

It really is better use of those extra magnets to spread them out on a larger diameter disk.  Some people find this easier to understand when they consider that the magnets are moving faster that way.  However you look at it, the multiple smaller disks create less voltage in each of the coils than a bigger disk with double the number of magnets.

Using larger disks, the power of the alternator increases in leaps and bounds (what is often wrongly called 'exponentially' but is actually the square).  Double the magnets and you get twice the voltage per coil.  Double the number of coils and you double the voltage again.  So you can get four times the power output with the same current in the coils. By stacking the alternator into 2 stators you only get the doubling of coils but not the doubling due to each coil seeing all of the magnets going past.  All this is with the same coil losses (heating per coil) so you actually get much better efficiency too.  Twice as many coils means losses will double whereas output is fourfold.

I am explaining this at length because I get asked about this roughly once a week, and so I feel it really needs a thorough explanation.  I hope it is clear but by all means ask questions and I am sure we can get into more detail.  Please don't think this is just a case of insisting on doing it the same old way due to resistance to change.  Both the HAWT and the large diameter alternator are actually better than the VAWT and the multi stator alternator.  That's why we talk about them so much and hope to help people avoid wasting time on failed ideas.  VAWTs became obsolete hundreds of years ago, but the idea seems to catch people's imagination anew every week.  (Sorry I seem to have turned this into a rant :-)

Hugh

[GoVertical]

“flux”
I am not Scoraigwind but I will give you my take on this.

A magnet can only give a certain flux density into a magnetic circuit. That circuit is basically an iron circuit ( or the magnet itself) and an air gap. For an example if we took one of the common magnet shapes and built a single rotor machine with a laminated core to close the flux path, your coils will be in the air gap formed on one side of your magnets.

If you modify the arrangement to use both sides of the magnet then you will have 2 air gaps in series. If the gaps are the same as for the single rotor case the flux density will be lower. To get the same you need to halve each gap.

Ans: I do not see the association you are trying make between a single and multi rotor configuration.


You can work with one big gap or two smaller ones but the total; flux available from the magnet is the same ( you can't have your cake and eat it).

Ans: The center rotors are part of the magnetic circuit that rotor configuration create. The coils of the added stators are just placed in the lines of flux path that has been already been created. 

“flux” I fail to see what the number of coils has to do with the argument. What matters is the total flux and the number of turns linked. if the flux is the same you get the same emf with the same number of turns. It makes absolutely no difference if the total turns in one phase is in N coils or 2N coils.

Ans: I would rather have a discussion. The vertical configuration is acting the same as the traditional fabrication technique. I mentioned coil and magnet count because in a earlier post it was said that it does seem to be a efficient use of material. To fabricate a 3 phase with 8 coil per phase using the traditional technique would require 64 magnets and a large diameter stator. The vertical fabrication method would only require 32 magnets and can be fabricated in a much smaller foot print.

“flux” In either case you can adjust the total turns in the coils to get exactly the same emf.
To go even further you can always get the emf you want with any magnet set up at any speed, what you need for power is to be able to generate that emf in a coil of low enough resistance to produce the current you want.

Ans:  Yes, increasing coil and magnet count results in a higher output at a lower RPM.
Stator resistance can be address by using a cog stator configuration. Lower stator resistance, better efficient at higher RPM's.

“flux” By splitting the flux between both ends of the magnet you will need more coils to get the same emf but you will add more copper length and increase the resistance.

Ans: The lines of flux are not being divided and remain the same as with the traditional method.  The coil placement is just taking full advantage of the lines of flux path created by the vertical rotor configuration. As I understand it the Femm simulation software does not allow for nonmagnetic martial backed rotors.

In the ideal case there is no net gain or loss but in real life you need to consider leakage flux that doesn't link your windings and you also have to remember that you need a mechanical gap for it to rotate in. This for safety needs to be about the same with one stator or two and it is wasted in that you have no coils linking this. By creating two extra useless gaps you will inevitably be able to use less of the flux with your two ended design. This and the added flux leakage will work against you and it will be more difficult to get the result that way than the conventional way with an air gap on one side of the magnet only.

Ans: I am not sure what you trying to communicate, can you provide a visual aid to help me understand?

Unless you can grasp the concept of magnetic circuits and working the magnet at its BH max point you will struggle to follow this but I hope it gives you something to think about.

Ans: I am researching BH max point and I am trying to understand why you would even mention it.

“flux” Don't get me wrong, your idea will work fine and you can build a good alternator that way but you will get lower efficiency from your magnets and you complicate the mechanical design.

Flux

Ans: Different points of view are always a good thing. As I said earlier the vertically stacked stator and rotor configuration is just an experiment and not fully tested. I am looking forward to spring so I can complete the project. Best Regards

[GoVertical]

I agree with Flux.  You need to think of the magnets as rather similar to batteries, and the air gaps as rather like light bulbs.  There's no free lunch whereby you can get the same current in two light bulbs in series - they will both be dim.  You also have to put more batteries into the circuit.  You need to have just as many batteries to make those bulbs glow brightly. So you can't sneak extra coils into the magnetic circuit without losing out on flux density in the larger (total) air gaps.

It really is better use of those extra magnets to spread them out on a larger diameter disk.  Some people find this easier to understand when they consider that the magnets are moving faster that way.  However you look at it, the multiple smaller disks create less voltage in each of the coils than a bigger disk with double the number of magnets.

Using larger disks, the power of the alternator increases in leaps and bounds (what is often wrongly called 'exponentially' but is actually the square).  Double the magnets and you get twice the voltage per coil.  Double the number of coils and you double the voltage again.  So you can get four times the power output with the same current in the coils. By stacking the alternator into 2 stators you only get the doubling of coils but not the doubling due to each coil seeing all of the magnets going past.  All this is with the same coil losses (heating per coil) so you actually get much better efficiency too.  Twice as many coils means losses will double whereas output is fourfold.

I am explaining this at length because I get asked about this roughly once a week, and so I feel it really needs a thorough explanation.  I hope it is clear but by all means ask questions and I am sure we can get into more detail.  Please don't think this is just a case of insisting on doing it the same old way due to resistance to change.  Both the HAWT and the large diameter alternator are actually better than the VAWT and the multi stator alternator.  That's why we talk about them so much and hope to help people avoid wasting time on failed ideas.  VAWTs became obsolete hundreds of years ago, but the idea seems to catch people's imagination anew every week.  (Sorry I seem to have turned this into a rant :-)

Hugh

Greetings, thanks taking the time to respond. I do not view your response as a rant. For me I am merely working on VAWT design. The biggest problem has been low RPM. I was unable create a faster blade design, so I am proceeding with a PMA configuration that require a lower RPM to achieve higher output. The limited testing I was able complete before the cold weather stop my progress has suggested that this approach has merit. The lathe I use for testing was unable to drive the PMA higher than 300 RPM when trying to charge a 12 volt battery. So I trying to match the PMA with low RPM, high torque blade configuration. I am searching for a low cost, easy to fabricate design. There are commercial VAWT designs, namely PACWIND and MAGLIFT and HAWT have along proven track record. I do understand what you are saying regarding the center rotors. For my project the cost fabricating of the vertical configuration off sets any efficiency loss that may happen from lower magnet count. Any one reading this should know that the vertical configuration I am work on is just a experiment that has not been fully tested. When I have the opportunity to complete the testing I will post the results. Thanks again and best regards.           

[fabricator]

“flux”

Ans: Different points of view are always a good thing. As I said earlier the vertically stacked stator and rotor configuration is just an experiment and not fully tested. I am looking forward to spring so I can complete the project. Best Regards

This is  not a question of one point of view or another it is a question physical reality.

[JW]

GoVertical;

 Those are some nice pictures, thanks for sharing 


JW

[ghurd]

I am searching for a low cost, easy to fabricate design.

A pair of steel disks would cost a LOT less in neo and copper.

[Antero]

There has been several attemps to create a real VAWT.

I have been looking and reading everything about them since 2000.
This was perhaps the best attempt;   http://www.turby.nl/
They had first a very big big site, to introduce the product and the test results.
Now they have only this site.

I have once tried to order Pacwind Delta VAWT to my test. I phoned and send several... E mails etc.
They never answered to any of my calls or E mails. That was really fortunate, because now I have read
several cases about it..

Etc.

THIS is a law of physics;
To create for example the same true wind power of 2kw 3,8m HAWT, VAWT needs to be 2m dia and 6m height= 12m2.

How to manage the storm sitution, which HAWT survives with tail etc

My adwice; forget here and now the VAWT, its a waste of time and ..

VAWT is for special circumtancies, for example we have here in Finland;
 http://www.cypresswind.com/
They develope this for emergency power in very high telephone masts to telephone companies, they do not sell this to any other customers.

Other one here in Finland, which I have been in contact several times. For very special.. circumstancies too;  
http://www.windside.com/index.html

Antero

[joestue]

Greetings, thanks taking the time to respond. I do not view your response as a rant. For me I am merely working on VAWT design. The biggest problem has been low RPM. I was unable create a faster blade design, so I am proceeding with a PMA configuration that require a lower RPM to achieve higher output.

At some point you have to be able to understand why the 120 rpm turbines in hydro electric dams are 40 foot in diameter.

anyway, this thread's been hijacked enough..