# Axial flux generator

1. Aug 22, 2015

Hi, in an axial flux generator if the flux passing through any of the coils at any given time constant is in the same direction , can all the separate coils be connected in series as to create a single phase output?

in permanent magnet rotor this would mean that the rotor disc facing from one side would haev all magnets facin N to the coils and the other side disc would have all S facing the coils.

the output then should resemble that of a oridnary DC generator before rectifier.?

2. Aug 26, 2015

bump:)

3. Aug 26, 2015

### sophiecentaur

I don't see why not. There are many alternative designs for a single phase alternator (which is what you are suggesting). But the output waveform is nothing like as convenient for providing a good smooth DC output, which is what you can get with many poles - hence many phases of output AC - each one rectified to provide a closely spaced sequence of pulses.

4. Aug 27, 2015

in the electromagnet case if I feed the rotor flux with a field current which is with a high frequency , I suppose i would get the same frequency waveform out of the genrator?
because as the rotor passes each of the axial coils the flux has changed its strength due to the varying flux but since all the coils are in series the total induced current should also swing accordingly to how strong the flux has been on each passing coil?

5. Aug 27, 2015

### sophiecentaur

The output of such a generator would depend upon the frequency of the exciter field, the number of poles and the relative phases of each coil and the rotation rate.

But, if you are considering making the rotor with elecromagnets, are you not making things very difficult for yourself?. I thought the advantage of this sort of generator was its simplicity because of the permanent magnet rotor. (Or is this a thought ecercise?)

6. Aug 27, 2015

not exactly , the idea is more like that of a revolver.imagine five cylinder shaped stator laminations which are axial in their placement. each of the cylinders have a coil on it.
at one end all the five cylinders join together while at the other end they are each separated by some distcance. at the end were they are joined together there is another coil . the field coil.the rotor is made such that at this end were the cylinders join together the rotor has a close airgap between the stator cylinder ends.

at the other side the rotor has a sort of carving or extension should i say similar to those found in magnetic reluctance motors/generators.
as the field coil is energized flux is now running through the stator cylinders and through the rotor but the flux xan only loop back into those cylinders which are near the rotor extension at any given moment as the rotor extension passes by them.
so as the rotor moves the rotors extension is dragged past each cylinder , sort of like each cylinder is fired with a flux.as this happens fast enough and all the five output coils on each cylinder are connected in series the induced current frequency and waveform should resemble the one fed into the field coil.

what do you think sophie?

7. Aug 27, 2015

### sophiecentaur

I don't know what to think.
I thought I knew what the basic geometry of an axial flux generator and I can't reconcile what you suggest with that. Have you a diagram?

8. Aug 30, 2015

well I have attached an image , my drawing skills are bad thas for sure but in the image you can see a solenoid with a coil on it , imagine there are say five such solenoids placed around a rotor , each of these solenoid has two coils on it, one is the field coil the other one is the output coil.both all field coils and also all output coils are connected in series.
now take a rotor which has two different ends , at one end there is a disc so that all five exial solenoid ends would be magnetically covered at all times, the other end has something more like a blade or whatever you call it , so at this other end the flux that extends from the disc side of the rotor can enter the solenoid only where the rotor passes them by.
so you energize the field coils , and rotate the rotor and the flux should then go from the solenoid into the rotor and back into the solenoid at the other end, only since the rotor is rotating the flux can enter each of the solenoid as the rotor passes them by, creating alternating flux through each of the solenoids.
does this sound like its getting somewhere now?

#### Attached Files:

• ###### Solenoid-with-core.JPG
File size:
26.4 KB
Views:
188
9. Aug 30, 2015

### sophiecentaur

OK. I'm still not much wiser but I can say that, as a general principle, if the field linked to a coil keeps changing, then you will get an emf induced in the coil. The actual amount of emf and the AC frequency will depend on the actual design.
Your picture suggests to me that you propose to replace the permanent magnets of a conventional axial flux generator with a number of solenoids. Whilst this is quite possible, the idea negates the principle of avoiding taking power to the rotating pieces. If you wanted to do it your way, it seems you would need 2N slip rings and brushes (for N independent solenoids). This would spoil the 'pancake' design with a permanent magnet rotor and make it no more attractive than a conventional alternator (in which it's normal to take power via slip rings to the rotating exciter coils (but, even then, with only two rings, afaik).
In principle, though, you seem to be right about the basics (from my understanding of your vague description; a variable frequency exciting voltage and a rotating field will induce yet another frequency of emf in the output coil(s).
This puts me in mind of the old audio pitch changer which used a set of rotating play heads, which scanned a moving tape, on which the input signal was being laid down. Google Tape Pitch Changer is you are interested.

10. Aug 30, 2015

, sadly I couldn't find much useful info about tape pitch changer, seems a rather old technology.
anyway , well no the rotor is without slip rings or any other rotation contact parts.the stator consists of these five solenoids (core and coil) and on each solenoid there is also a second coil which is the field coil.all of these field coils are connected in series since the flux through all of the sator solenoids is the same direction at any given time.
then the rotor has a shaft and two end plates, one is in the form of a disc which covers the ends of the solenoids at all times , the oher is in the form of blades which cover at maximum two solenoid ends at any given time.
the rotor is basically a flux path , just because of its shape the rotor connects the flux to the solenoids one by one as it passes by creating a changing flux (increasing decreasing) as it passes by the solenoids. a little similar to a magnetic reluctance generator, were the rotor only connects /disconnects magnetic flux paths.
why would i need to have the actual field windings on the rotor when i can just use the flux they create which to my mind has the same result.?

11. Aug 30, 2015

### sophiecentaur

Yes it was old technology and sounded very burbly!
Your description is getting us somewhere now. It still seems nothing like an axial flux generator, though. The flux due to the rotating solenoid seems to be largely radial and tangential- at the ends of the solenoid.
My previous comment about any varying field producing an emf still holds but it's the amount of change of flux that counts and your diagram doesn't have enough information to tell what's going on. If you are really serious about this then I suggest you learn to use a simple drawing package and produce a better diagram which could help your verbal description.
I will watch this space.

12. Sep 1, 2015

before i post a diagram , I just wanted to clear up that for a fixed rpm , a higher excitation frequency would increase the output of a generator like this right?
in typical ac generators with dc excitation the increase in frequency comes from the increase in rotational rpm and also results in higher output voltage and power.

13. Sep 1, 2015

### sophiecentaur

I can't really answer that without a diagram because it all makes very little sense to me yet.
I suggest you look at the diagram on that link about a pitch changer (rotating head and tape transport) and look for some analogy there. The age of the technology is not relevant.

14. Sep 2, 2015

ok sophie , I have attached two images , for now this is the best i can do , i think you would understand the workings from these.
so you have N number of solenoids placed in the stator , the higher the number the more precisely the waveform of the generator output is shaped.the solenoid are in the axial direction , now even though in the pictures each solenoid has one coil on it , imagine each has two , one bigger (the output coil) one smaller (the field coil) as said previously each field coil is connected in series and the same goes for output coils.
now the rotor is in the middle basically a simple form , shaft with a disc at one end and blades at the other. the end with the disc gest all the flux from all solenoids simultaneously, then the flux is linked through the rotor shaft (also a low reluctance material) and bends over to the other side were the blades are , now at this side the flux can only enter two solenoids at a time , and as the rotor rotates those two solenoids change all the time , if the rotor has sufficiently high rpm then the waveform fed into the field coils can be closely matched at the output + the rpm induced frequency , so basically we can get a desired output waveform and frequency by just controlling the field current waveform and frequency.

#### Attached Files:

File size:
106.4 KB
Views:
164
• ###### suuds 3.png
File size:
19.7 KB
Views:
130
15. Sep 2, 2015

### sophiecentaur

A conventional axial flux generator has two rotors with magnets arranged so that the field is axial- from a pole on one rotor to a pole on the other.(at least that's the kind I know. A pancake shaped stator in between the rotors has a number of coils which produce number of phases of AC. I do not see how your system is anything like that or how it can produce a suitable axial field. Your picture does not convince me, I'm afraid and your description confuses me. Sorry.

16. Sep 2, 2015

well ok , maybe my description is not the best and maybe some parts of this idea are not working , might as well just take the typical axial flux generator which you described and imagine that the N-S poles on the rotor whose flux cuts through the coils placed on the stator are not permanent magnets but electromagnets, the way you make the rotor in this case isnt that relevant only the flux path through the coils, core and the orientation of the whole generator geometry is.
take the rotor the way you know it think that it has electromagnets on it and the flux through the coils goes only in one direction at any given moment (AC excitation current)
since the stator has those many physically independant coils on it each coil sees a different level of induction since the current in the field windings goes up and down as in a sine wave.
now the result i think is still there , an output which follows input in terms of frequency and waveform only magnified in voltage and current as in a generator it should be.

17. Sep 2, 2015

### sophiecentaur

This starts to look to me more like a multiphase transformer with the primary rotating relative to the secondary.
I have no doubt that there will be some AC Volts across the output but where the energy has come from (excitation input or what's driving the rotor) is not clear. I think the answer to that could be to do with the relationship between the rate of rotation of the rotor and the frequency of the AC excitation.
Altogether, I have a problem seeing any application for this idea. What would you think it could be used for?

18. Sep 3, 2015

well it could be portrayed as a transformer to some degree , only the difference here is that the primary winding itself is not rotating as that would require sliding contacts but there is no need for that , instead take the flux which the primary winding creates and let that flux in the rotor which can then distribute the flux the way it needs to be done to get induction in a particular design.
well there is field current which is a tiny part of the whole energy input , which after the generator startup is then fed from the generator output as in ordinary generators, and then there is rotational torque supplied by some rotating power source , a turbine or whatever i just dont focus on that because for the generator inner workings its irrelevant.

as you said the rpm of the rotor always plays a part in terms of how much energy and how high of a frequency you get out from a generator , but the AC field current can then be used to control the output or set it to a particular desired frequency and voltage since on mechanical loads the rpm can vary so the ac field current could be used to compensate, also there are mechaical power sources with very low rpm , and normally that requires a lot of poles and a big generator , if this thing works might as well just have a much smaller generator and use a very high field current frequency , as in every transformer and motor if one increases the frequency the same voltage or output power can be kept but using less loops of wire etc.

well sort of like this the thinking goes , what do you think ?

19. Sep 3, 2015

### sophiecentaur

Surely it's totally relevant. It has to work or it's not worth discussing.
Either way, your generator needs sliding contacts - either for field or output power.
Changing the flux involves transferring energy (the field holds energy). Which way the Power flows will depend on the relative phases of the two supplies of power (the engine turning the rotor and your exciter device). I can't understand, from your description, why it should do what you claim.
Do you have any reference to anything like your idea or did you generate the idea on your own?
No one else seems to be responding to this so I would guess that no one else understands what you are describing either.
If you want to be sure that the idea has legs then you would need to understand how (in detail) a conventional alternator works (numerically as well as in principle) and how multiple alternators in parallel behave.

You could, perhaps, look at references to a Selsyn motor or a Magslip which has some similar ideas employed in the design.

20. Sep 3, 2015

why would i need sliding contacts or contacts at all ? arent there alternators on the industrial scale that use a field coil on the stator that induces current in a coilset placed on rotor then the current is ran through a rectifier on the very rotor and then the rotor has poles which are ctreated by the rectified AC that has been turned into DC.

instead of having sliding contacts and physical electromagnet on the rotor , why cant i have a stationary field coil whose flux is extended through an airgap into the rotor which then puts that flux through the output coils and the flux is lined back the other side into the field coil.
I guess this then begs the question , does it only work when the very source of the field rotates or can it also work if the source stays stationary but the flux created by that source is made to rotate ?

21. Sep 3, 2015

### sophiecentaur

Seems ok in principle. Can you supply a reference to a working system? I always assumed that slip rings would be needed. But then, what about your other coils?

But I still don't see where this is going. What is the final purpose?

22. Sep 4, 2015

About the reference to a working system , first of all here is a wiki article (yes I know about flaws in wiki but this doesnt seem to be one of them )
https://en.wikipedia.org/wiki/Alternator#Brushless_alternators

then ,
and here is a nice page for marine students that i just found http://www.marinesite.info/2014/03/how-does-brushless-alternator-works.html

so basically these types of alternators have two parts, the main parts works exactly the same as if the current for the rotor would be supplied directly via sliding contacts , but the first side works like a rotary transformer, you have a primary winding (stator exciter coil) then a core (rotor armature) and a secondary winding thats rotating.
If my thinking is correct then if you take a tarnsformer , an ordinary one and just rotate the core while its working no changes to the induced current happen , but if you physically move one of the two coils either the pirmary or secondary , you get additional induced current due to the work done and additional increase or decrease in frequency depending on the speed and direction you move the coil with respect to the half period of the sine wave at that moment ?
What do you think is this reasoning correct?

In my case , I dont want to have these two parts because here the goal is to have the same DC poles on the rotor but just have them there without any sliding contacts that shorten the life of the device.since the main output part still has the usual design of DC rotor poles and a stationary output coil on the stator this generator still produces an output frequency which is fixed and related to its rpm, the rpm of the rotor.

In my case what I do is take away the rectifier first so that the rotor has a time varying polarity changing field as the field coil has.normally if I applied such field to a typical alternator the output induced would be a " cripple" it would probably self cancel at times and so on. So the geometry of what I'm talking about is a little different as I described earlier with the example of cylinder type solenoids and a rotor.
In my case I only need to have that exciter coil (field coil) made flux enter the rotor so that from the rotor it could be distributed along the output coils as it rotates.
the question is can i just link the flux from a stationary field coil into a moving rotor and then the flux from the rotor would resemble the same flux wixh is normally on a rotor who has windings on it , or do i physically need another set of coils on the rotor to get current in them which then would set up a rotating field.

maybe another way to explain this would be , take a transformer for example, now imagine somehow you are able to rotate the core around while its working , in terms of current induced in the secondary nothing would change because the core is uniform all around , so the b field seen by the secondary coil would see no difference.
take that same transformer now but imagine the core has two symmetrical gaps of equal length but opposite sides in the core.the primary flux created by the primary coil is still linked to the core but now as you move the core it becomes something similar to a reluctance generator , and now the secondary coil does feel additional forces from the rotating core.

In other words , imagine my generator is a transformer with the only difference that instead of all the secondary power coming from the primary with no moving parts , in my case the primary is just a small coil for setting up the field and the core is the rotating part and the secondary winding is the output coil, so the primary winding sets the flux and frequency , the core is moved by a mechanical power source and the secondary's induced current is the result of the pirmary + the mechanical force from the rotor.
you would ask, " Why on earth one needs this" ? asnwer - I want a generator whose output frequency and even waveform to an extent can be fully controlled by the field current.

one of the positive sides of this could be a generator run on high frequency having the same power output with smaller dimensions than a generator with the same output power but an oridnary design.

something similar has already been thought out from one guy in the internet , only his version , which i talked much about here and there myself, has moving (sliding) contacts and in that particular design they cant be avoided due to the laws of physics sadly.

well I want ot hear what you think, i;ve tried to explain myself , although i believe I havent done the best job one could want , but I tried.

Last edited: Sep 4, 2015
23. Sep 9, 2015

well i think I found a generator that comes very close or actually is exactly what I'm talking about , it's called reluctance alternator.but even this generator produces a frequency which corresponds to its rotational rpm , mainly used in missiles.

24. Sep 10, 2015

### sophiecentaur

Don't be sad. Now you have introduced some hard facts (much better than the 'impressionist' sketches (lol) you may get more interest.
The brushless alternator is very interesting. Marine use would justify the extra expense and avoid the effects of salt air contamination on brushes. Marinising engines is a big deal (and makes them well overpriced, aamof). My ordinary 12V alternator was an ordinary one and needed attention after an incident with seawater in the engine compartment. This Delco Remy would have been impervious to that problem - just a bit more bulky and a lot more expensive probably.
So does a normal alternator - rotate a magnet in a coil and that happens. It could have an advantage if a particular impedance was required, perhaps, for instrumental reasons.
I have read and re-read your post and it seems to me that you are going to get a beat between the driving AC and the rotational frequency.
Is this a kind of magnetic amplifier? What is the application? What is the input, what is the output and what does it actually do?

25. Sep 11, 2015

in the case of delco remy video , from the drawings it's kinda hard to understand how the poles rotate with respect to the field coil which there is stationary ,I assume similarly to the way I proposed in my " impressionist" paintings :D:D field coil being in the middle and a low reluctance magnetic rotor just distributes that flux across the output coils as it passes them by.

as for your question , well it's not exactly a magnetic amplifier although some workings are similar , but in a mag amp the very power and sine frequency comes from an already existing sine wave AC source who is just controlled (transistor analog) but since this is a generator , the very power and frequency is created from a non electric source (mechanical) so in terms of this it's still an ordinary generator , a mechanical power source is transformed into electric power.the importance lies in to how it's done.
take for example the last video of the guy describing his high frequency alternator based on a faraday disc, as I've said earlier that idea comes as close as one possibly can to what I'm talking about here.
it's actually perfect except for the need for brushes and the very limited single loop output.
But if you spin it , you can excite it with any ac frequency (up to a certain level i suppose were loss would start to dominate) and it would output a current with that frequency only amplified so to speak of but in actuality not amplified simply generated since the power for the increase in current came from a rotational source.
I'm sure you understand of what I;m talking about.
I mean the brushes spoil the fun but in terms of a good simple robust generator i think this one is good it has very low internal resistance , and its output frequency and hence power can controlled by a simple resonant circuit driving the field coils.

When I did all my impressionist skecthes and all that talking I was just describing ideas of how to replicate this phenomenon found in the AC faraday disc generator only without brushes.since the faraday disc doesnt work without brushes as its only current generating part must rotate with respect to a b field.

What do you mean by this? like that the output frequency would always be that of field + rpm of rotor? , well if I could ge that without some sort of distortion in he waveform then fine , but i somehow doubt it , what do you think?