how to detect eddy currents and measure their negative impact

[brandnewb]

Please excuse me if I am totally blundering. But I would really like to see if my premise has any merit or not. Also I am going out on a limb here so I might be mistaken on the physics behind it all. That subject never was my strong point ;(

My premise is that traditional laminated sheets for coils are sub optimal. The sheets still conduct electricity. Because there is a tiny gap between the sheets the effect of eddy currents induced by magnets rotating over them is reduced but not eliminated.

I would like to introduce a totally new way of going about things in this day and age but before I do so I will need to be able to demonstrate that the concept is indeed an improvement over traditional.
Something I am certainly not ready yet to without a doubt through out into the world at this stage. {2} But I want to believe!. Famous quotes The X files{2}

Please see the only stator of a 3 phase single stator alternator that I like to end up with. The rotors (magnets are on both sides but just not displayed here)

The idea is to have the coil disk engulfed with aluminum powder filled resin on the outskirts and the coils them self with iron powder filled resin.

I am imagining the incredible heat dissipation ability increase over just plain resin should I have understood things correctly and there are no unforeseen road blocks ahead.

Also for each magnet group (1 pair in both sides making 4 magnets in total) there are 3 coils. 1 for each phase. So basically there is a continues loop with minimal interruptions.

I have been trying to demonstrate in a convincing manner that the iron powder I am using does not (or at the very least much less than traditional laminated sheets) produce eddy currents because the powder is not electro conductive.

Early test results suggest I am on the right track but I am also known for making huge blunders so I need to take this a bit more scientific before I start making bold claims.

Please see the latest test setup I used. Earlier setups that made me think this is indeed worth exploring are horrendous and unwieldy.

The distance between the magnets is roughly 28mm and the field strength is around 120 mili tesla where I can get my probe at.
The central disk is non compacted (just gravity filled) iron powder. To more closely resemble the state it would be in when cast in resin.

Something like this.


In this current configuration there central bearing will contribute to the central disk rotating when I use a hand held drill to spin the magnets over the disk.
But it could also be that there is also the eddy current induced drag we all fear.

I did try rotating the magnets rather fast back and forth back and forth for like 1.5 minutes at full rotation drill speed and there is simply no heat build up that I can detect.

How do I make sure I am on the right track here?

{1] @SparWeb, please allow me to quote a line from your signature.
"No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it."
If coping quotes is a no no then please delete this addit{1}

[mab]

I kind of feel you've answered your own question: if you want to measure eddy current losses then you measure the drag induced in the rotating rotors or the heat generated in the medium under test.

Not sure what you mean by laminated sheets - unless you mean laminated iron cores which are not usually used in AF Generators. But if you are referring to laminated iron cores then to compare with your powder cores, you would need to do the same eddy current tests on a laminated iron core too.

[brandnewb]

please find here an example of what I meant with laminated sheets in coils

https://www.fieldlines.com/index.php/topic,149398.msg1062917.html#msg1062917

I just no longer believe this is still a good way of going about things.
{1}the sentence above is far to harsh. I meant to say I no longer believe this is still the most ideal way of making alternators.{1}

I guess I meant to ask how can I convince the rest of us?

{2} I'll revisit this when I have the iron powder disk completely isolated so only eddy currents can make it rotate. No longer also bearing friction{2}

[brandnewb]

Does anyone know where I can buy premade laminated sheets in Europe that are used in coils? I can't seem to find any.

I will need this to compare results.

[JW]

Hey nice CAD drawings...   I run Solidworks  2019 and its awesome. As far go,s getting material you have to imposes/compromise   

JW

[joestue]

Does anyone know where I can buy premade laminated sheets in Europe that are used in coils? I can't seem to find any.

I will need this to compare results.

The problem is any attempt to cut slots in the core will short the edges together and this affects the losses far more than anything else does.
saturation of the tooth edges also affects losses far more than the bulk steel does.

figure 5 to 10 watts per kilogram for the core losses of a well designed motor, at 60hz.

[brandnewb]

Thank you JW for your compliment.

I will admit though that I have been trying to migrate from this old limping software I use now to blender. I hear the open source blender has gotten add onns that will blow anything out of the water known to man.

But the migration trajectory is too steep for me at the moment so I'll just limp on for the time being using my trusty old setup.

Want to hear something unsettling? No matter I will tell anyway. My kid recently showed interest in what I was doing and I felt motivated to teach. But then I remembered I still use this old software that really should not be tought about anymore ;(

It got me motivated again to move to blender and all its power. I know of nothing better yet. But somehow I am soo busy I seem to not be willing to make the time yet.

But anyway JW. Would you please elaborate on "As far go,s getting material you have to imposes/compromise"? That one has gotten me stuck at the moment ;(

Kind regards

Hey nice CAD drawings...   I run Solidworks  2019 and its awesome. As far go,s getting material you have to imposes/compromise   

JW

[brandnewb]

Thank you Josh for pitching in.

Can you please try and curve your contribution in such a way that I can also make sure I can incorporate it in this scientific trail?

Here are the question marks
* Where do I buy?
* What do I buy?
* How do I prevent all the things I have no idea yet you were talking about.
* How do I setup an experiment that will convince on a scientific nature that I am on the right track or to the contrary?

Does anyone know where I can buy premade laminated sheets in Europe that are used in coils? I can't seem to find any.

I will need this to compare results.

The problem is any attempt to cut slots in the core will short the edges together and this affects the losses far more than anything else does.
saturation of the tooth edges also affects losses far more than the bulk steel does.

figure 5 to 10 watts per kilogram for the core losses of a well designed motor, at 60hz.

[brandnewb]

I could really use some advice regarding if I am using the multimeter correctly to demonstrate that the laminated plating in a transformer still conducts electricity.
And not even a little if I interpret the readings correctly.



Now no matter how close one puts the probes together in this particular iron powder, except for the probes touching of course, the multi meter always display .OL, does not beep and the red light does not light up.

I have great difficulties finding laminated sheets that I can use in my testing to be able to compare results. I might even hit a road block here.

Are there any other tests I can perform to make it possible for me to just trust I am not making a huge blunder going with this iron powder?

{1}Also please try and envision how much more close contact the powder can become with the wires as oppsed to laminated sheets. In fact with powder one can really focus the field enhancing properties exactly there we one needs it. The wires. And not inadvertently create paths for the field to slip around the wires rather than through them when using laminated plating.

How can this not be a huge improvement over traditional? But then again I do make blunders and physics is not my strong suit.

I really hope one is willing to help identify where I am missing the mark here.

{1}

[brandnewb]

to make sure I get my point across here is a super fast and loose model of the concept. fast though it actually was not that is why it is loose

the lighter colors are resin filled with a material that can conduct heat very well but does not conduct electricity and also does not conduct magnetic flux.
The darker colors are iron powder filled resin.



And of course a continues loop of 3 phase coils.

Also the rest of the disk made of this lighter part is not shown of course.

[mab]

Ok, so you are using iron sheets to make a laminated core.

Your tests on the transformer core are probably reading correctly: when they assemble the 'E' and 'I' laminations, they weld the E's and I's in two lines as a quick and cheap way of holding the core together - so that electrically connects the laminations to each other at two points.

The purpose of using laminated iron is to provide a good magnetic path without providing an electrical path for the eddy currents ; the eddy currents always run in a plane perpendicular to the lines of flux, so as long as the laminations run parallel to the lines of flux they block the eddy currents, even if the laminations are electrically connected to each other at a single point.

The transformer manufacturers get away with connecting at two points as the points are magnetically symmetrical.

If you really want to do the work of experimenting, then the easiest way would be to find an old bolted together EI transformer and pull the laminations apart, cut them to shape, but electrically seal the cut pieces before assembly into the shapes you want- to avoid the multiple shorts Joestue advised you about.

[brandnewb]

yes thank you mab for your support.

However I am not using any sheets to make cores at this moment as I cant find where to buy yet. On display was a transformer that I ripped out of an old microwave.

Regarding alignment of sheets. You made me wonder if I should let the powder cure in a magnetic field perpendicular to the one it will be rotated by over. Just for giggles if nothing else.

[brandnewb]

I will let this matter rest now.

I have made every effort to convince my self. And although I am not fully confident yet I will also just accept that proving it beyond a shadow of a doubt is just not in my interest at the moment as it is rather difficult to do so.

My bold claim is that I have figured out a way on how to improve on ancient notions.

If one wants to challenge that then please do so as I am always open to be stood corrected. In fact it is important that this gets peer reviewed else risk never being adopted while it might need to be.

[Mary B]

I will let this matter rest now.

I have made every effort to convince my self. And although I am not fully confident yet I will also just accept that proving it beyond a shadow of a doubt is just not in my interest at the moment as it is rather difficult to do so.

My bold claim is that I have figured out a way on how to improve on ancient notions.

If one wants to challenge that then please do so as I am always open to be stood corrected. In fact it is important that this gets peer reviewed else risk never being adopted while it might need to be.

If you plan on playing with that microwave transformer as is be VERY careful, it develops lethal voltages!

[brandnewb]

understood mam! I will put it back in a dusty corner where everyone knows there be dragons.

[JW]

 i stumbled over this-

"By Lenz's law, an eddy current creates a magnetic field that opposes the change in the magnetic field that created it, and thus eddy currents react back on the source of the magnetic field. For example, a nearby conductive surface will exert a drag force on a moving magnet that opposes its motion, due to eddy currents induced in the surface by the moving magnetic field. This effect is employed in eddy current brakes which are used to stop rotating power tools quickly when they are turned off. The current flowing through the resistance of the conductor also dissipates energy as heat in the material."

https://en.wikipedia.org/wiki/Eddy_current

[Mary B]

understood mam! I will put it back in a dusty corner where everyone knows there be dragons.

People have lost hands to the high voltage DC those can generate when rectified... some lost their hearts if it grounded across their body... EXPLODED...

[brandnewb]

yes Mary! I am really taking your warning seriously. When building a diy transformer I got dangerously close to the edge and vowed to not do that anymore.

Anyway I found a manufacturer in the Netherlands that produces the sheets. But they do not stack them and since the order will be tiny the costs will be relatively high.

Now I am willing to do it because I am curious. But they do not punch circles. So I can get a square of 15cm. Or a rectangle in basically any length but maximum 15cm tall. Orienting the direction of the sheets will be my responsibility as I need to do the stacking my self and that is not easy the guy on the phone told me. Easy to loose track and start messing up the orientation.

several thickness of sheets to choose from as well as well as composition.

I guess the first thing to determine is what are typical sheets used in axial pm alternators. Does anyone have a suggestion?

One thing the guy did tell me to make me less hopeful is that even though losses are reduced never completely eliminated. The amount by how much depends on the composition of the sheets and quality of stacking.

If left to me then that means there will always still be that shadow of doubt so is it even worth going down this path?

[MattM]

You need a guy with a CNC table with a plasma cutter.  Any shape is then possible.

[brandnewb]

Ahh finally I understand it. You have a keen eye Mary. The full bridge rectifier you saw in the background was there all along and is a part of the alternator test bench. The transformer was only a short term invader not invited to the party.

I plan on using the full bridge rectifier to also test the difference between star and delta after rectification.

understood mam! I will put it back in a dusty corner where everyone knows there be dragons.

People have lost hands to the high voltage DC those can generate when rectified... some lost their hearts if it grounded across their body... EXPLODED...

[brandnewb]

Ok Josh,

I think I owe you an apology as at first I thought you were speaking in riddles. But the more I get to understand the explanation of what you meant by mab I think this can actually work. WHen having enough material to cut into pieces.

So the idea is to rip the microwave transformer apart. cut it to bits and try and shape a similarly shaped disk with it as long as the parts are not connected to each other?

The only problem I can think of is that I will never have enough pieces to come even close to the powder disk. And even if I could get close then it would only be for several mm rather than the 15 that was on display.

And the sheer amount of work that would take is also offputting to me if I understand the concept correctly.

Would you please elaborate a bit?

Does anyone know where I can buy premade laminated sheets in Europe that are used in coils? I can't seem to find any.

I will need this to compare results.

The problem is any attempt to cut slots in the core will short the edges together and this affects the losses far more than anything else does.
saturation of the tooth edges also affects losses far more than the bulk steel does.

figure 5 to 10 watts per kilogram for the core losses of a well designed motor, at 60hz.

[JW]

So the idea is to rip the microwave transformer apart. cut it to bits and try and shape a similarly shaped disk with it as long as the parts are not connected to each other?

   No

[brandnewb]

wow, I just realised I am still on the right track.

As I used AI to help me find some more background:

Look what happened:

----------------
me:
does iron piowder conduct elecgtricity

AI response
Yes, iron powder can conduct electricity. However, its conductivity is not as high as that of solid iron due to the air gaps between the particles, which can impede the flow of electrons. The conductivity of iron powder can also be affected by factors such as the size and shape of the particles, the degree of compaction, and the presence of any coatings or contaminants on the particles.

me
if the iron powder does not conduct electricity. will it produce eddy currents when ran over by magnets?

AI response
Eddy currents are created when a conductor is exposed to a changing magnetic field. Since iron powder is a conductor, it can indeed produce eddy currents when a magnet is moved over it. However, the eddy currents in iron powder would be significantly less than in a solid piece of iron due to the air gaps between the particles, which impede the flow of electrons.

If the iron powder is so finely divided that it behaves as an insulator rather than a conductor, then it would not produce eddy currents. Instead, each particle of iron would respond individually to the magnetic field, resulting in magnetization of the particles rather than the generation of eddy currents.
---------


Well I can definitely see that the powder starts behaving like a magnet after running experiments.

I am not letting this go just yet

{1}I will also start asking other questions But just like me AI is known to blunder so do not sell your house yet so you can bet big on its advice!!{1}

[brandnewb]

you know what. I will let this go for now.

I have ran early tests that even though not explicitly as I set out to do it did show me that moving magnets that are not attached to the same bearing as the powder disk  (but they came from the side so not officially part of the experiments I was set out to run) produced no drag as the disk stayed still

Also when spinning magnets at drill speed for a few seconds until the powder disk started rotating at the same speed due to bearing friction. Then reversing the drill direction and doing the same.
And then doing the above again and again for like 1.5 minutes did not produce any heat that I could detect.

I have now officially convinced my self that using this particular iron powder will not have issues with eddy currents.

Also when taken into account how much better we can channel the magnetic field, when using a powder like substance rather than a solid, to run through the copper wires rather than around them as opposed to conventional methods I would really like ones suggestion how this can not be an improvement over traditional? And let us not forget the additional heat dissaption it can bring as oppsed to standard resin.

I mean I am sure I must be overlooking something if no one here seems the least bit intrigued.

[joestue]

You can lookup iron powder core datasheets to get the losses, then just assume yours will be worse given that your iron powder is likely not in anyway optimized with regard to hysteresis losses.

https://www.micrometals.com/products/materials/-26/

So for example at 400hz and an ac flux density of 1T, the losses are around 0.8 watts per cubic centimeter. At 60hz its just over .1 watts, which would be 13 watts per kilogram losses at a flux density of 1T at 60hz. Good quality non grain oriented silicon steel is probably lower losses.. which is why no one makes motors from mix 26 iron powder cores, not to mention the physical differences.

Anyhow. At 1 T mix 26 appears to be around a permeability of 20, which is plenty good.

Lets assume you can make an iron powder core with lower permeability, because its a lower density. So the eddy current losses are lower, but the hysteresis losses are higher. So lets presume its .1 watt per cubic cc at 60hz at 1T.

But, you aren't near 1T in the entire core, and less than 60hz.

I think it will be difficult for you to measure the losses without an oscilloscope, or some careful thermal measurements.

If you have a drill press and a disk of magnets you can spin above a sample disk of iron powder, you can bolt the iron powder down to the table and spin the disk at 2000 rpm or so, then measure how much it warms up.

[brandnewb]

Our domain is 2.5hz maybe 5. Does that change anything?

[joestue]

Our domain is 2.5hz maybe 5. Does that change anything?

Should be a whole lot more than 5hz. That's only 5 magnets per second passing over your coils.

Anyhow just assume the losses are linear to the frequency.

[JW]

https://www.google.com/search?sca_esv=593477997&q=hysteresis&spell=1&sa=X&ved=2ahUKEwiYrZe8qamDAxX6nokEHbSEBGcQBSgAegQIBRAC&biw=1280&bih=559&dpr=1.5

By Lenz's law, an eddy current creates a magnetic field that opposes the change in the magnetic field that created it, and thus eddy currents react back on the source of the magnetic field. For example, a nearby conductive surface will exert a drag force on a moving magnet that opposes its motion, due to eddy currents induced in the surface by the moving magnetic field. This effect is employed in eddy current brakes which are used to stop rotating power tools quickly when they are turned off. The current flowing through the resistance of the conductor also dissipates energy as heat in the material.

sorry for posting this quote twice

Ok here's another

 

In thermodynamics, a change in the thermodynamic state of a system and all of its surroundings cannot be precisely restored to its initial state by infinitesimal changes in some property of the system without expenditure of energy. A system that undergoes an irreversible process may still be capable of returning to its initial state. Because entropy is a state function, the change in entropy of the system is the same whether the process is reversible or irreversible. However, the impossibility occurs in restoring the environment to its own initial conditions. An irreversible process increases the total entropy of the system and its surroundings. The second law of thermodynamics can be used to determine whether a hypothetical process is reversible or not.

[brandnewb]

Thank you joestue for this clarification,

What it did is for me to be able to trust that at the very least I am not planning on something that will degrade the performance more than help it.

On the other hand if there is no need for field enhancing additives I think I had better not use them to begin with. As it will make it harder for me to keep the disks where they need to be.

@everyone, Please please do not get what I am about to say in the wrong way. I do not wish to be misunderstood for someone that keeps nagging not willing to admit defeat.
But I still think there is no evidence either way. Either coils in a disk like I suggest are a step forward or traditional still reigns supreme. I guess we will never know ;(

Our domain is 2.5hz maybe 5. Does that change anything?

Should be a whole lot more than 5hz. That's only 5 magnets per second passing over your coils.

Anyhow just assume the losses are linear to the frequency.

[Mary B]

Iron powder potted coils WILL cog! Ferrites I use to eliminate RFI are magnetic and they are compressed powdered iron... one of the low frequency ones since I am using them on 60hz issues....

[brandnewb]

do you mean cog as in a magnet wants to stay attracted to the coil and it is difficult for it to move to the next one?

Because if that is what you meant then that is also my definition of cogging.

But can you please show me examples where each phase was so close together that it would be surprising that cogging still occurred?

{1}i.e if stacking coils in an axial arrangement has been around for ages then I will admit that I have not done my homework right. I will blame youtube!! {1}

{2}cogging always occurs to a certain degree. I mean coil wires under load will also be a part of it. Especially when one uses field enhancing additives. I mean to explore where it becomes an issue. I am almost certain that when spacing coils apart like traditional then yes cogging will become an issue.

But when we go for a continues loop then those matters are dealt with aren't they?

Please just someone help me out of my misery.

{2}

{3}And no I will not start introducing Ai again for now as it is rather expense and the answers are usually still garbage{3}

[brandnewb]

One thing I do wonder about, what is this poll functionally meant to do

[MattM]

How often do you want readings?