Need help/ suggestions regarding iron core AFPMG.

[Aamir]

We all know that when a load is connected to a PMG, an opposing magnetic field is created in coils which attracts ongoing and repels incoming magnet causing drag in generator. Has anyone tried using all N poles on one side and S poles on other side in an IRON CORE SINGLE PHASE AFPMG. I hope it will certainly reduce the drag because incoming magnet will be attracted instead of getting repelled but I think this configuration will also produce lower output than standard N-S-N-S configuration.

[joestue]

We all know that when a load is connected to a PMG, an opposing magnetic field is created in coils which attracts ongoing and repels incoming magnet causing drag in generator. Has anyone tried using all N poles on one side and S poles on other side in an IRON CORE SINGLE PHASE AFPMG. I hope it will certainly reduce the drag because incoming magnet will be attracted instead of getting repelled but I think this configuration will also produce lower output than standard N-S-N-S configuration.

no.

can't get something from nothing.
the highest efficiency very large transformers approach 99.7% efficient at 1 watt of iron loss per kilogram of iron, and 2 watts per kilogram of copper.

you can buy magnetic cores that have 1 watt per kilogram losses at 1.4T flux density, and those cores are useable up to 1.7 to 1.9T flux density. But the moment you break the magnetic path and make a motor, there are many other losses that show up, such that only very large motors can reach 95% efficiency again. large as in, 50hp or so.

[MattM]

Flux doesn't seem to travel out from iron all that well unless you energize it as an electromagnet.  When you energize it then it acts like a supermagnet.

[Aamir]

My magnets (25mm x 10mm N35) have arrived. Next step will be making some different size iron cores, stator and rotors but I am unsure about rotor thickness. I am thinking about using 6mm thick steel discs for rotors and make 25mm x 3mm round slots in them for magnets fitting. Is that ok for 8 magnets per rotor or should I use thicker discs for rotors?

[joestue]

basically the flux density through the magnet is going to max out at like 1T. so take the cross sectional area of the magnet and compare to the cross sectional area of the steel behind it where the magnetic flux flows, but and then allow for the flux density in the steel to be 1.5T

so if your 25mm diameter magnets are set inside of the outer perimeter of the disk, then the thickness of the steel can be as little as 3mm and you might detect a little big of magnetism on the backside of the disk.

6mm is plenty. don't need to make 3mm deep slots for the magnets to sit in, 1mm is fine.

[Aamir]

basically the flux density through the magnet is going to max out at like 1T. so take the cross sectional area of the magnet and compare to the cross sectional area of the steel behind it where the magnetic flux flows, but and then allow for the flux density in the steel to be 1.5T

so if your 25mm diameter magnets are set inside of the outer perimeter of the disk, then the thickness of the steel can be as little as 3mm and you might detect a little big of magnetism on the backside of the disk.

6mm is plenty. don't need to make 3mm deep slots for the magnets to sit in, 1mm is fine.

Thanks Joestue. I chose 3mm deep slots because magnets will be temporarily fitted in the disc and I don't want to use epoxy or super glue for them. If 1mm deep slot is enough to hold the magnets in place without any adhesive then I will use 5mm thick disc to make it lighter.

[Aamir]

I copied this image from another thread to ask a question. Which set of coils will have highest output, A or B? I thought B as magnets are exactly aligned with the center of coils but according to a member's statement in another thread, it should be A. In an iron core coil, output will be maximum when magnet is aligned with iron core but I have no experience about air core coils.

[Astro]

(Attachment Link)


I copied this image from another thread to ask a question. Which set of coils will have highest output, A or B? I thought B as magnets are exactly aligned with the center of coils but according to a member's statement in another thread, it should be A. In an iron core coil, output will be maximum when magnet is aligned with iron core but I have no experience about air core coils.

 3 phase Alternating current

[Astro]

As for plate thickness, you will achieve saturation at about a 3 mm thick plate with those magnets and them having some steel around them. Meaning not at the very edge of the plate, which is a little less of a concern with round magnets. If you are using square or rectangle magnets the closer you want to put them to the edge of the plate, the thicker you need the plate to be to reach saturation.

[SparWeb]

Hi Aamir,
The output is continuously alternating.  The rotor has to be turning for anything to happen.  Both A and B will have their moment to peak in the positive and then in the negative.  Your equipment will just measure AC current. 

The diagram does look OK for a 32-pole 3-phase alternator.  The 24 coils will be wired into phase groups, where every 3rd coil will be joined in series.  Each phase will have 8 coils together.  If the coil marked "B" is the start of the coils in one phase, then the other coils in the same group will be the ones with the magnet "red dots" centered on them.

[Aamir]

Hi Aamir,
The output is continuously alternating.  The rotor has to be turning for anything to happen.  Both A and B will have their moment to peak in the positive and then in the negative.  Your equipment will just measure AC current. 

The diagram does look OK for a 32-pole 3-phase alternator.  The 24 coils will be wired into phase groups, where every 3rd coil will be joined in series.  Each phase will have 8 coils together.  If the coil marked "B" is the start of the coils in one phase, then the other coils in the same group will be the ones with the magnet "red dots" centered on them.

My main concern is that when an air core coil will produce peak voltage?
A) When the moving magnet reaches the center of coil so that magnetic flux flows through inside of coil.
B) Both sides of coil cutting magnetic flux simultaneously.

[MattM]

When a magnet is directly over one side of the coil there will be current.  When the magnet is in the center the forces mostly zero out as the forces balance.  But any point where the magnet cross over one side of the coil in any way there is an inequality, therefore current potential.  When magnets of opposite poles cross over opposite side - simultaneously - there is the greatest imbalance, and that would be your greatest current potential.  Any time you create current, however, there are magnet fields created simultaneously that oppose the movement of the magnets.  That resistance of your PMG to generate electrical power is referred to around this forum as cogging.  With that resistance comes heat, which means gauge of wire and heat tolerance of the magnets has to be taken into consideration.  Too much cogging can be a big problem to practical usefulness when it comes to a PMG, therefore the number of coils and magnets are carefully chosen to keep cogging at the lowest while keeping the phases in harmony.  Breaking your coils into separate phases also reduces cogging.  (The 3-phase design in this thread demonstrates harmony by keeping each leg ideally at 120 degrees apart.)  Adrian Kraatgen's reports well document multiple phase designs, and he usually gives you the breakdowns of the engineering with universal mathematics.  Once you see the patterns it will help you understand the choice of methods used in designs around here.  But you can also learn through trial and error.

[Astro]

When a magnet is directly over one side of the coil there will be current.  When the magnet is in the center the forces mostly zero out as the forces balance.  But any point where the magnet cross over one side of the coil in any way there is an inequality, therefore current potential.  When magnets of opposite poles cross over opposite side - simultaneously - there is the greatest imbalance, and that would be your greatest current potential.  Any time you create current, however, there are magnet fields created simultaneously that oppose the movement of the magnets.  That resistance of your PMG to generate electrical power is referred to around this forum as cogging.  With that resistance comes heat, which means gauge of wire and heat tolerance of the magnets has to be taken into consideration.  Too much cogging can be a big problem to practical usefulness when it comes to a PMG, therefore the number of coils and magnets are carefully chosen to keep cogging at the lowest while keeping the phases in harmony.  Breaking your coils into separate phases also reduces cogging.  (The 3-phase design in this thread demonstrates harmony by keeping each leg ideally at 120 degrees apart.)  Adrian Kraatgen's reports well document multiple phase designs, and he usually gives you the breakdowns of the engineering with universal mathematics.  Once you see the patterns it will help you understand the choice of methods used in designs around here.  But you can also learn through trial and error.

 Yup what he said. The short answer is, as he said, when the magnet passes over the edges of the coil (not the center) is when you get voltage and current. The magnets flux lines that produce voltage and current are not like a bubble that cover the whole coil. Fact that is why we put an opposite pole magnet on the other side of the coil (dual axial) is to make more flux lines pass through the wire. The edges of the coil when laying flat and looking from above are often called legs of the coil and that is important when trying to explain this. As I am explaining it the "sides" of the coil are if they were standing upright and we were looking down from above at them. If one magnet passes over the leg of a coil, you have voltage and current. If 1 magnet passes over one leg of the coil AND you have an opposite pole magnet on the other side (not leg), (dual axial) you will produce more voltage and current, due to more flux lines in between the 2 magnets. NOW if you have a magnet over each leg of the coil and they are of opposite poles, you are going to produce more voltage and current then if you just had one magnet over one leg of the coil. If you add opposite pole magnets on the other side of those magnets (dual axial) well you have taken the pma about as far as you can and are getting as much voltage and current as you can out of it, without making it different with different wire size, windings and mags.  The picture you are referring to is my project. I am shooting for a fairly low rpm machine and that is why it is designed like it is. I spent my entire life working with electricity and building things, so I had a pretty good head start when it came to how to design a pma for the application I wanted. That said I had never poured epoxy or anything. I will say this about the project, my plates will be done and ready for me to lay out the magnet arrangement on them today and so we will find out in the next month or so how this thing works, as I will be spinning it up in the garage. IF I HAD TO DO IT ALL OVER AGAIN, I would have built a much smaller machine first. My thought and focus was goals. I wanted x amount of power at x amount of rpms and that was that and I started designing and building. However I never thought of the experience part of it. Just because it looks good on paper does not mean it is going to be easy to build. I think building a smaller simpler one first would have been the way to go for a first build. Some people on here have built enough of them, that they would not have any problem building a pma like this, but they have experience. Plus they are pretty cool about sharing that invaluable experience.

[Aamir]

MattM and Astro, that's what I wanted to confirm. Thanks to the experienced members of this forum, I have learnt a lot about AFPMG. I haven't started working on my project yet because I am still learning .

[Astro]

MattM and Astro, that's what I wanted to confirm. Thanks to the experienced members of this forum, I have learnt a lot about AFPMG. I haven't started working on my project yet because I am still learning .

 That is what it is all about, learning.
I tried to figure out a better way of coil to magnet numbers and what these guys will tell you is a 4 mag  to 3 coil ratio is best. So an example is 3x4 for 12 coils (with 4 coils per phase) and since we took the coils times 4, we take the mags times 4 and get 4x4, for 16 mags.
 As I said I spent a little time, proving that theory and from what I came up with, yeah it is correct. The thing I found interesting is when you start getting up there in coil and mag numbers, the difference becomes larger. For example, I thought about doing a 18 coil. 3 x 6 is 18 so 4 x 6 would be 24 mags. So 6 more mags then coils. But I went with a 24 coil stator. 3 x 8 is 24 and so 4 x 8 is 32 or 8 more mags then coils. On a 16 mag 12 coil, the difference is 4 more mags then coils obviously. It is simple math, but it plays a large part in what determines the needed rpm to get the desired voltage and current. It is my hope and by running many numbers several times over, that my 24 coil, 32 mag pma should perform really well for the turbine I am trying to build.. To be honest, it has not been cheap to build and I am glad from a complexity (building it) stand point, from a cost standpoint and from a shear size weight stand point, I am glad I did not try for anything larger.  My plates weigh about 20 pounds each. My mags weigh almost 5 pounds. My stator which is a non spinning mass, but it weighs 13 pounds. So it starts adding up fast.
So....... that said I should have 50-60 pounds or so of spinning mass in the center, and that is going to play into how much I am allowed in weight for the wings or rotating mass further out from center. I think that is going to play a big part in obtaining and sustaining the desired rpm of this vawt build. Since a vawt is much like a old hit and miss engine, when you think about it, because not all wings have positive air pressure on them at the same time. Also because they are not like a hawt that has tapered blades and thus keeps the bulk of it's rotating mass towards the center. But I like you found myself reading a lot and thinking a lot about everything before I picked up a tool or bought one thing for the project.
 It is by far the most complex project I have undertaken by myself. In a build like this I would normally just build the stator, figure out wire size from it to the building and play with the controls of it all to make it work properly (so the electrical part of it). Not have to worry about air foils, the physics of a flywheel, what bearings to use, or even make the mold to cast the stator in. All that was always left up to other people who specialize in those areas.

[Aamir]

Has anyone tried coils like this in AFPMG? If I make coils like this having same diameter and thickness of magnets (23mm x 10mm or thicker) and use 8 coils between 16 magnets (8 magnets on each side) for single phase output, will it work?

[Astro]

(Attachment Link)

Has anyone tried coils like this in AFPMG? If I make coils like this having same diameter and thickness of magnets (23mm x 10mm or thicker) and use 8 coils between 16 magnets (8 magnets on each side) for single phase output, will it work?

 The question is not will it work, because as we have established, a magnet passing over a copper coil produces an electrical current. The question is how well does it work. That is the question and really the only question in building these things. Even the ones from China that get poor reviews on youtube work, they just do not work very well. Intended use plays a part in how well something works as well. Look a butter knife will work as a screwdriver in a pinch, but it will not work as well as a screw driver and will be pretty useless in building a house. Point is the right tool for the job.
 That said, why are you going for single phase and not 3 phase? Second thing is, if you recall MattM explained that how a magnet produces electrical current as it passes over a coil is...... it creates an imbalance of forces between the two legs of the coil. So if the magnet is passing over both legs at the same time, do you think that is an imbalance or is it an equal force on both sides at the same time? Or in other words think of yourself standing on an inner tube out in the middle of a lake. If you have equal pressure or force on both of your legs, you are balanced. If you have more weight or force on one leg, you created an imbalance and are going to tip over. So again, the greater the imbalance between the two legs of the same coil, the more the induced electrical current will be. In the inner tube example, we want to tip over as fast as possible, because that means we have a large imbalance.

[Aamir]

A few months ago, I found a website about Transverse Flux Generators ( I am not allowed to post the link ). As my goal is single-phase iron core PMG, I think this design is the best as it has the shortest path for magnetic flux in C-shaped laminated iron cores like an AFPMG, but it doesn't need steel discs for magnets to complete the magnetic flux through coils which means lighter rotor and simple design as compared to dual rotor AFPMG. Not enough data is available about this design, so I am relying on experts' opinions/ suggestions. Thanks.

[Aamir]

Published 1 January 2019
Physics, Engineering
IEEE Transactions on Industrial Electronics

"The new transverse flux generator enjoys a coreless rotor leading to a reduction in manufacturing complexity, as well as a decrease in the mass of the rotor, which in turn results in a better active power density. On the other hand, the number of permanent magnets used in the proposed machine is half of that utilized in a conventional transverse flux generator, which reduces the construction cost, especially in large-scale transverse flux generators. Moreover, in the proposed machine, the leakage flux originating from the inactive magnets would considerably be less than that in conventional machines."

[MattM]

Seems silly to go to such a high expense without figuring how to add a secondary coil with its own c-shaped stator coils to pick up the opposing fields in its own phase.  They would just have to run the C-shapes stretched a little so that the two coils would only run inside their own respective C's.

[joestue]

The comments above about the number of coils to magnets getting out of hand when the pole count rises isn't exactly a problem.. Its just that the 3/4 coil magnet ratio maxes out at .86 winding efficiency compared to a theoretical .96+  2 pole 36+ slot generator. it doesn't matter if its an air core or a steel core.

The higher pole count 1:1.1 ratio magnet to coil generators an get as high as .96% winding efficiency, but they often have high frequency (well by definition) and thus high eddy currents. For example, you can fit 30, .25" wide magnets inside many small 36 slot 56 C frame motors.. and easily get 1 hp at 1200 rpm.. when the original 6 pole 60hz induction motor from the same frame got 1/3rd hp max at 1200 rpm at less efficiency.

But in a slow speed direct drive wind turbine eddy and hysteresis and high frequency issues.. none of those are present.

Now as the magnet count increases and the frequency rises you end up with more leakage inductance which means that at some point you short the windings and the rotor don't stop. ... but if you add a mechanical brake those problems are solved. (this isn't an issue in an air core motor because the inductance is so low.. (But as a result the resistance is so damn high)

From what i can see there is a happy medium where for a given volume of magnet there is an optimum rotor diameter. (number of poles be damned) and i think it increases quick. larger seems always better until you factor in the rest of the volume and cost of the generator and how much air it blocks.

basically it doesn't really matter how much you plaster the rotor with coils and magnet cores, the maximum benefit you get is about 10% from going from the 3/4 coil magnet ratio to a much better ratio, but you increase the rotor diameter and you get a whole lot more torque for free.


please note that i am not talking about jumping from 1 layer winding to 2.
you can fit two, 3/4 coil magnet ratio coils on top of each other.. increase the "end turns" and then squish them into each other and you can for all intensive purposes double the power output for a given set of magnets.

the reasoning is that by increasing the end turns to allow for the coils to overlap.. you also have much increased surface area while practically doubling the amount of copper flowing radially directly through the magnetic field, out of it, hopping over the other set of coils, and flowing straight out. so you can keep the current the same while doubling the available area to cool the coils.

[Aamir]

Seems silly to go to such a high expense without figuring how to add a secondary coil with its own c-shaped stator coils to pick up the opposing fields in its own phase.  They would just have to run the C-shapes stretched a little so that the two coils would only run inside their own respective C's.
(Attachment Link)

MattM, you are right, but I have something else in mind. How about using layered windings on the center of C-shaped laminated iron cores as in a bicycle dynamo and connecting them in series to get low current high voltage output? I think that will do the job.

[Aamir]

The comments above about the number of coils to magnets getting out of hand when the pole count rises isn't exactly a problem.. Its just that the 3/4 coil magnet ratio maxes out at .86 winding efficiency compared to a theoretical .96+  2 pole 36+ slot generator. it doesn't matter if its an air core or a steel core.

The higher pole count 1:1.1 ratio magnet to coil generators an get as high as .96% winding efficiency, but they often have high frequency (well by definition) and thus high eddy currents. For example, you can fit 30, .25" wide magnets inside many small 36 slot 56 C frame motors.. and easily get 1 hp at 1200 rpm.. when the original 6 pole 60hz induction motor from the same frame got 1/3rd hp max at 1200 rpm at less efficiency.

But in a slow speed direct drive wind turbine eddy and hysteresis and high frequency issues.. none of those are present.

Now as the magnet count increases and the frequency rises you end up with more leakage inductance which means that at some point you short the windings and the rotor don't stop. ... but if you add a mechanical brake those problems are solved. (this isn't an issue in an air core motor because the inductance is so low.. (But as a result the resistance is so damn high)

From what i can see there is a happy medium where for a given volume of magnet there is an optimum rotor diameter. (number of poles be damned) and i think it increases quick. larger seems always better until you factor in the rest of the volume and cost of the generator and how much air it blocks.

basically it doesn't really matter how much you plaster the rotor with coils and magnet cores, the maximum benefit you get is about 10% from going from the 3/4 coil magnet ratio to a much better ratio, but you increase the rotor diameter and you get a whole lot more torque for free.


please note that i am not talking about jumping from 1 layer winding to 2.
you can fit two, 3/4 coil magnet ratio coils on top of each other.. increase the "end turns" and then squish them into each other and you can for all intensive purposes double the power output for a given set of magnets.

the reasoning is that by increasing the end turns to allow for the coils to overlap.. you also have much increased surface area while practically doubling the amount of copper flowing radially directly through the magnetic field, out of it, hopping over the other set of coils, and flowing straight out. so you can keep the current the same while doubling the available area to cool the coils.

Joestue, thanks for such a nice and detailed explanation. What's your opinion about my idea of using layered windings in C-shaped iron cores?

[Aamir]

I will be more than happy to hear comments/ suggestions from other members too.

[MattM]

MattM, you are right, but I have something else in mind. How about using layered windings on the center of C-shaped laminated iron cores as in a bicycle dynamo and connecting them in series to get low current high voltage output? I think that will do the job.

You need a whole rotation to complete one complete polarity change, so not anywhere near as well.  You may want a higher frequency.

Honestly, not a fan of the whole concept because of the distance from your magnet to the changing flux.  The closer you get your wires of the coil to the flux, the more power you will realize from the magnets.

[Aamir]

MattM, you are right, but I have something else in mind. How about using layered windings on the center of C-shaped laminated iron cores as in a bicycle dynamo and connecting them in series to get low current high voltage output? I think that will do the job.

You need a whole rotation to complete one complete polarity change, so not anywhere near as well.  You may want a higher frequency.

My requirement is high voltage dc, so frequency is not an issue. We can use low forward voltage drop Schottky diodes and MOSFETs for this purpose, but we will see about this later.

"Honestly, not a fan of the whole concept because of the distance from your magnet to the changing flux. The closer you get your wires of the coil to the flux, the more power you will realize from the magnets."

With a minimum air gap, aren't coils with iron cores supposed to be producing more power than air core coils?

[JW]

Your missing a logic theory component. I have lots of experience with 300ampdc switching.

For DC the current is switching is instant so lightspeed in air.

You need too understand PWM.     

[Aamir]

Your missing a logic theory component. I have lots of experience with 300ampdc switching.

For DC the current is switching is instant so lightspeed in air.

You need too understand PWM.     

I know and understand PWM ( pulse width modulation ). I have lots of experience in SMPS with active and passive PFC, low-frequency and high-frequency inverters, and online UPS. PWM is not a concern here because the dc output of this generator will be unregulated.

[JW]

I was just pointing out that a dc field can collapse at the speed light etc.

Look at the aspect/ I have smoked over 3000usd on mosfets in the my experience.


 

[Aamir]

JW, you are right. A slight mistake and MOSFETs/ IGBTs won't get another chance. I mentioned MOSFETs as a reference. I have no intention to use them in my project.

[Aamir]

Hi to all members.
I have a question i.e., I have this C-shaped powdered iron core. What is the best way to complete the magnetic path through this core?
A: Two separate magnets on both sides with an iron bar as shown in pics.
B: A rectangular magnet with opposite poles on the same side.

[MattM]

A straight iron bar from the N to the S.  That creates a complete circuit.

[Aamir]

A straight iron bar from the N to the S.  That creates a complete circuit.

Thanks.