magnets: to stack or not to stack

[brandnewb]

So in the context of a dual magnet disk single coil: when using a non steel backed / non powdered array of single magnets of type N45 60x10x5 at an 3.75 degree interval and an inner radius of 600mm.

then any spacing between the upper and lower part of the magnet disks larger than 20mm seems to no longer make sense.

You see the field is so drastically better then. Almost no more side ways lines. For how I interpret the data I have the reduced coils space does less to reduce the voltage than the increase in voltage does by the field.
And as a bonus now we have less resistance in the coils because we can now reduce the coil length as now every turn counts more than it did before. This saves on the length thus resistance. Which will reduce the chance of overheating and in general will allow for higher amperage .

Reading on single magnet array with counter part 40mm distance (I ran out of spacers so could not simulate current coil width)                        
id   field polarity   surface   measurement point   distance (mm)   milli tesla   sensor angle   increase over   By %
V   N   top   a   0   321   0   singe/array   1.26182965299685
V   N   tp   b   0   338   0   singe/array   3.36391437308868
V   N   top   a   10   72   0   singe/array   22.0338983050847
V   N   top   b   10   70   0   singe/array   22.8070175438596
V   N   top   a   20   39   0   singe/array   95
V   N   top   b   20   38   0   singe/array   111.111111111111
V   N      g   0   14   0      
V   N      g   17   3   0      
V   N      h   0   18   0      
V   N      h   17   8   0      
V   N      I   0   17   0      
V   N      I   17   4   0      
                        
                        
Reading on single magnet array with counter part 30mm distance (with 300 winds / 600 turns this is as narrow  as I can make the coil I think)                        
id   field polarity   surface   measurement point   distance (mm)   milli tesla   sensor angle      
V   N   top   a   0   324   0   singe/array   2.20820189274448
V   N   tp   b   0   332   0   singe/array   1.52905198776758
V   N   top   a   10   80   0   singe/array   35.5932203389831
V   N   top   b   10   78   0   singe/array   36.8421052631579
V   N   top   a   15   71   0   singe/array   108.823529411765
V   N   top   b   15   66   0   singe/array   112.903225806452
V   N      I   0   17   0      
V   N      I   17   2   0      
                        
Reading on single magnet array with counter part 20mm distance (I think the reduced space for coils is well made up and more by this increase in flux)                        
id   field polarity   surface   measurement point   distance (mm)   milli tesla   sensor angle      
V   N   top   a   0   337   0   singe/array   6.30914826498423
V   N   top   b   0   346   0   singe/array   5.81039755351682
V   N   top   a   10   132   0   singe/array   123.728813559322
V   N   top   b   10   141   0   singe/array   147.368421052632
V   N      g   0   ~+0   0      
V   N      g   10   ~+0   0      
V   N      h   0   ~+0   0      
V   N      h   10   ~+0   0      
V   N      I   0   ~+0   0      
V   N      I   10   ~+0   0

Next up is a stacked array and see if I can keep the space larger between magnets disks. You see I am not looking forward to the pull. It will bring engineering issues with it

Shoot, I forgot to send the reference photo.


   

[brandnewb]

is not there a simulator for these kinds of experiments by now in where one can basically virtually do what I am doing but then theoritical which is faster and less risky?

the ones I did find are either outdated with bad software or so non operable one can't call it a simulator.

[brandnewb]

Earlier it was mentioned cleaning and scratching the surface of magnets before glueing them. This was in the context of the mad experiment I am really no longer certain abouit to be interest in ;(.

But it does make me wonder. Could not it be that there is something on the magnets I have got that is in the way of reaching again the 70% increase while stacking. This claim I made back a while ago that I said was unreliable?

I mean going over them one by one with a super fine sanding paper is not too much of a hassle.

[Mary B]

is not there a simulator for these kinds of experiments by now in where one can basically virtually do what I am doing but then theoritical which is faster and less risky?

the ones I did find are either outdated with bad software or so non operable one can't call it a simulator.

This is the one I see mentioned here a lot https://www.femm.info/wiki/HomePage

[Mary B]

Earlier it was mentioned cleaning and scratching the surface of magnets before glueing them. This was in the context of the mad experiment I am really no longer certain abouit to be interest in ;(.

But it does make me wonder. Could not it be that there is something on the magnets I have got that is in the way of reaching again the 70% increase while stacking. This claim I made back a while ago that I said was unreliable?

I mean going over them one by one with a super fine sanding paper is not too much of a hassle.

NOT the magnet surface! At least not on Neo's! They will corrode rapidly! But the metal surface yes!

[joestue]

Agros2d can do simple 2d stuff.

Its a bit hard to use

[MattM]

I have 2 sheets of 5mm steel.

so;
single non stacked magnet
--steel plate--
single non stacked magnet
--steel plate--
single non stacked magnet

if that is what you meant I can do the test later today.

Yes.  The top and bottom using same N-S direction with middle S-N.

[brandnewb]

This is the one I see mentioned here a lot https://www.femm.info/wiki/HomePage

Agros2d can do simple 2d stuff.

Its a bit hard to use

Yes Thank you Guys for the suggestions regarding simulators. I took a quick peek and got intimidated rather quickly. So I will park this part for now as I am almost done collecting data to reach a well informed decision should the data and method hold up to scrutiny.

[brandnewb]

Yes.  The top and bottom using same N-S direction with middle S-N.

Interesting experiment. Please find the data.

Matt test (single in isolation)                  
id   field polarity   surface   measurement point   distance (mm)   milli tesla   remark
X   S   top   a   0   302   
X   S   top   b   0   323   
X   S   top   a   10   58   
X   S   top   b   10   60   
         avg flux measured      185.75   
                  
Matt test (SN-5mm steel-NS)                  
id   field polarity   surface   measurement point   distance (mm)   milli tesla   increase of %
X   S   top   a   0   425   40.7284768211921
X   S   top   b   0   423   30.9597523219814
X   S   top   a   10   89   53.448275862069
X   S   top   b   10   77   28.3333333333333
         avg flux measured      253.5   
                  
Matt test (SN-5mm steel-NS-5mm steel-SN)                  
id   field polarity   surface   measurement point   distance (mm)   milli tesla   increase of %
X   S   top   a   0   435   44.0397350993378
X   S   top   b   0   452   39.938080495356
X   S   top   a   10   87   50
X   S   top   b   10   82   36.6666666666667
         avg flux measured      264   



I will keep this setup for a few days just in case one would like to request some additional readings on it.

[brandnewb]

So I will have to take a short break because I am running into issues of fear I do not dare test the stacked array with its counterpart at a 20mm distance yet.

But I will share what I have so far.
If at all readable one will notice the insane increase in average flux of over 500% for the single array with counterpart at 20mm vs the same setup but then at 30mm distance.

 

Reading on single magnet array with counter part 40mm distance (I ran out of spacers so could not simulate current coil width)

Reading on stacked magnet array with counter part 40mm distance (actuallty current coil width is not compressed yet and top and botom layers removed after curing in irone powder filled resin so 40mm distance is good)

id

field polarity

surface

measurement point

distance (mm)

milli tesla

sensor angle

increase over

By %

id

field polarity

surface

measurement point

distance (mm)

milli tesla

sensor angle

increase over

By %

V

N

top

a

0

321

0

singe/array

1.26182965299685

V

N

top

a

0

447

0

singe/array cnt prt 40mm

39.2523364485981

V

N

tp

b

0

338

0

singe/array

3.36391437308868

V

N

tp

b

0

466

0

singe/array cnt prt 40mm

37.8698224852071

V

N

top

a

10

72

0

singe/array

22.0338983050847

V

N

top

a

10

112

0

singe/array cnt prt 40mm

55.5555555555556

V

N

top

b

10

70

0

singe/array

22.8070175438596

V

N

top

b

10

103

0

singe/array cnt prt 40mm

47.1428571428571

V

N

top

a

20

39

0

singe/array

95

V

N

top

a

20

61

0

singe/array cnt prt 40mm

56.4102564102564

V

N

top

b

20

38

0

singe/array

111.111111111111

V

N

top

b

20

60

0

singe/array cnt prt 40mm

57.8947368421053

avg flux in between magnets

146.333333333333

avg flux in between magnets

208.166666666667

singe/array cnt prt 40mm

42.255125284738

(all readings added * 2) / 12

(all readings added * 2) / 12

V

N

g

0

14

90

V

N

g

0

34

90

singe/array cnt prt 40mm

142.857142857143

V

N

g

17

3

90

V

N

g

17

12

90

singe/array cnt prt 40mm

300

V

N

h

0

18

90

V

N

h

0

90

V

N

h

17

8

90

V

N

h

17

90

V

N

I

0

17

90

V

N

I

0

90

V

N

I

17

4

90

V

N

I

17

90

Reading on single magnet array with counter part 30mm distance (with 300 winds / 600 turns this is as narrow  as I can make the coil I think)

Reading on stacked magnet array with counter part 30mm distance (with 300 winds / 600 turns this is as narrow  as I can make the coil I think)

id

field polarity

surface

measurement point

distance (mm)

milli tesla

sensor angle

id

field polarity

surface

measurement point

distance (mm)

milli tesla

sensor angle

V

N

top

a

0

324

0

singe/array

2.20820189274448

V

N

top

a

0

462

0

singe/array cnt prt 30mm

42.5925925925926

V

N

tp

b

0

332

0

singe/array

1.52905198776758

V

N

tp

b

0

468

0

singe/array cnt prt 30mm

40.9638554216867

V

N

top

a

10

80

0

singe/array

35.5932203389831

V

N

top

a

10

127

0

singe/array cnt prt 30mm

58.75

V

N

top

b

10

78

0

singe/array

36.8421052631579

V

N

top

b

10

123

0

singe/array cnt prt 30mm

57.6923076923077

V

N

top

a

15

71

0

singe/array

108.823529411765

V

N

top

a

15

110

0

singe/array cnt prt 30mm

54.9295774647887

V

N

top

b

15

66

0

singe/array

112.903225806452

V

N

top

b

15

106

0

singe/array cnt prt 30mm

60.6060606060606

avg flux in between magnets

158.5

avg flux in between magnets

232.666666666667

singe/array cnt prt 30mm

46.7928496319663

(all readings added * 2) / 12

(all readings added * 2) / 12

increase in avg flux over single 40mm

increase in avg flux over stacked 40mm

8.31435079726651

11.7694155324259

V

N

I

0

17

90

V

N

I

0

41

90

singe/array cnt prt 30mm

141.176470588235

V

N

I

17

2

90

V

N

I

17

7

90

singe/array cnt prt 30mm

250

Reading on single magnet array with counter part 20mm distance (I think the reduced space for coils is well made up and more by this increase in flux)

Reading on stacked magnet array with counter part 20mm distance (I do not dare to take these readings yet as these magnets holders were and early and weak version unable to fine tune the distance in a controlled manner)

id

field polarity

surface

measurement point

distance (mm)

milli tesla

sensor angle

id

field polarity

surface

measurement point

distance (mm)

sensor angle

V

N

top

a

0

337

0

singe/array

6.30914826498423

V

N

top

a

0

0

singe/array

V

N

top

b

0

346

0

singe/array

5.81039755351682

V

N

top

b

0

0

singe/array

V

N

top

a

10

132

0

singe/array

123.728813559322

V

N

top

a

10

0

singe/array

V

N

top

b

10

141

0

singe/array

147.368421052632

V

N

top

b

10

0

singe/array

avg flux in between magnets

956

(all readings added * 2) / 8

increase in avg flux over single 30mm

503.154574132492

V

N

g

0

~+0

90

V

N

g

0

90

V

N

g

10

~+0

90

V

N

g

10

90

V

N

h

0

~+0

90

V

N

h

0

90

V

N

h

10

~+0

90

V

N

h

10

90

V

N

I

0

~+0

90

V

N

I

0

90

V

N

I

10

~+0

90

V

N

I

10

90

[brandnewb]

Correction!!!. I found a blunder in a formula.

It is not over a 500% increase for the avg flux between single magnet in array with counterpart at 20 and 30mm distance. but a 50.788643533123.

I will keep double checking as blunders can and will be made I am afraid. I am now referring to the formulas and the mistakes that creep in coping cells around in a spreadsheet in where references can start become misaligned. So the readings them selfs should be top notch now.

Btw Adriaan. My offer for you to have you confirm or invalidate the findings using the setup I used still stands. Or anyone else for that matter that has the skill and is also a fellow lowlander.

I am willing to drive around the country and leave my stuff there for a few days and then pick it all up again once we know if we are on the correct track or not.

But before I do that I want to make sure I did all I can to spot obvious blunders first my self.

[brandnewb]

Btw my weather station is now reading at this very moment 14.9 m/s


I have some homework to do to make sure things do not take off ;(

[brandnewb]

wow 19.1 m/s wind speed at this time. I am getting slightly nervous.

Also regarding the results of the data. I just cant demonstrate favorable results for stacked magnets ;(

I can try one more thing inspired by the horizontal stacking suggestion I got here and hallbach. So this will be my last attempt at trying something before starting without stacking and see how far that brings me. Unless new information is brought up of course.



You see now the fight to keep same poles close to each other is much less. Something I still deem save on my lawn.

OR!! and someone please hold my hand here. Should I not be doubling those percentages increase for individual readings  as I am really only measuring up to 50% of a paired array.

But then again, if that were correct then there should have been a huge difference between the average flux over individual readings.

I am rather sad at the moment ;(

[TechAdmin]

Technical off-topic, if you want to add something right after you posted, please do try and edit your post rather than posting a new message. If it's several hours or a day apart it's fine, but if it's some 10 minutes or similar not as much because it completely clogs the recent posts area. Thank you
(Also I can't understand anything about the discussion but damn it's interesting! Keep it up!)

[brandnewb]

Sometimes I can manage  that just like now but some times I am really trying to make sure I get to complete the task at hand. And looking for the modify button seemed just so more daunting to me than reply

But understood sir !!

[Mary B]

Analysis paralysis... it is a real thing... sometimes you just have to leap in and build!

[brandnewb]

wise words Mary, Thank you.

If a single magnet array proves disappointing there is nothing stopping me putting another one on top of it

You see. that is why I value sharing my work. At times you get just the nudge you need else risk stagnation.

[MattM]

I posted about this strange ceramic magnetic phenomenon in January of 2022.  Where two magnetic fields overlap to combine effects.  Its worth exploring.

[brandnewb]

Analysis paralysis... it is a real thing... sometimes you just have to leap in and build!

I would like to elaborate a bit so that people get to know me a little better.

Next to my other unusual features is that I might get diagnosed with, if I ever went for a diagnose, OCD but then in a super mild manner that can be broken by simple mental gymnastics. But since I can't start on the new version of the turbine structure yet before the 30th of this month I just want to see what happens when I try a few different unusual magnet configurations.

I realize that things can be faster with simulation but I am both intimidated by the software and it would mean I have less experience running field tests in the end. Which can be considered a value when having it

Anyway I realize that my build quality when prototyping is not great but so please bear with me.

This will be my test bench for trying out the runners up magnet configurations with an oscilloscope.
\
Basically what you see there is something to hold the test coil in place and disk with a 1.34m diameter on where to put the magnet holder prints on the bottom. And to connect to the magnet holder prints for the top.

Let's see how far this gets me. I am really interested to learn how to use an oscilloscope.

Also Mary, You said I should be looking for/aiming for maximum width and height when reading on an oscilloscope yes?

That tells me that frequency of field reversal is only a lesser part of the equation.  It certainly is not in the simplified equation I brought under attention here.
e(voltage) = B(field (not even know in what unit it is expressed) * L (length of the conductor (probably expressed in metric units?)) * v (velocity of the coil moving through the field (probably expressed in metric units?))

But now I think that is not correct as I have already had a brain fart once in where I configured many magnet, if not all, wrong same pole up. I am imagining that on an oscilloscope that would look like a pulse with a width of infinity.

The voltage readings were definitely non 0 but just so much less that I had expected.

So my am now holding on again to what I was told once by someone I believe to have an excellent understanding of these matters.

An important factor is the frequency of field reversal. Yet another once told me that since I could not find a frequency component in this formula I keep bringing up that velocity basically means field reversal in this scenario.
I now no longer believe that to be correct.

So if someone has some easy to understand insights then please share. It might help prevent me from running some te4sts that are not going to lead anywhere.

EDIT: I chanced my mind  I again think this second someone is correct. Because the velocity with which a coil moves through the field is directly correlated to the frequency because this rotating coil in the video I shared earlier does crosses both fields.

There are soo many intuitions that are slowly developing in my mind. But it is a slow process

[Mary B]

On an Oscope the vertical amplitude is your true voltage, wider the pulse the more you have to put into your battery once rectified because really narrow pulses will average into lower voltage and amps...

Others on here have shown oscope readings of their coils... a quick search didn't turn it up but they are there... 

[brandnewb]

In honor of the name of this forum I have a question about fieldlines. And also about if there is a general consensus regarding what to focus on more. The length of the conductor or the strength of the field.

Please see 2 parts of a 3 part hamburger
[ Specified attachment is not available ]
The middle rotor segment is configured in a repelling setup.
This setup will allow for a 400% EDIT (100% but I have this itch that tells me that that will mean an increase of 200% well used coil length) increase in coil real estate for a 50% of a magnet increase.
The top 3rd layer of the rotor (that is not on display) will be similar to the bottom part. So not its counter part

The coil will be a U shaped one going around the middle rotor. EDIT: there is no need for a difficult U shape coil. Just 2 separate coils should also work I think.

The fieldlines it creates are roughly as outlined in yellow as far as I think I understand my readings.

While I am making modifications to my test bench to start using my oscope I am already worried that the shape on the display will be all over the place. Should I be concerned about that?

But anyway is there a general rule of thumb to what aspect to have weigh more while building alternators? Does one better focus on a stronger field as there is no down side to that other than engineering challenges and the increasing cost of the magnets.
Or does one focus on the length of the conductor which is cheap yet does bring resistance considerations ultimately affecting the amperage ability.

EDIT: further insights tell me that there is a tight curve in the fieldlines at the bottom of a repelled magnet.

EDIT: further insights tell me that there is also still a curve in the fieldlines between magnets at the bottom disk that are traditionally configured.
Which we can do away with if we were to get things even closer than 30mm. But I am just not sure I like taking on the challenges that come with that.

[brandnewb]

Just to make sure I'd like to remind everyone that there is an actual goal here.

please see the +7m diameter cup holders.





There are soo many advantages to having the arm beams that I can only imagine they outweight the disadvantages.

For example now, at the expense of slight turbulence increase , we can have only top arms. no more need for bottom ones.

Because now we can hook up the bottom of the blades to the extended arm next to it.

This is what the material is for that i expect to arrive the 30th. And of course I will try and make it as strong as I can make it but truth be told. 19.1 m/s??? I am still nervous if I want to keep it a free standing structure. I am considering cheating a bit and connecting it to the ground but i realise then that it will no longer be technically be free standing ;(

[brandnewb]

I realise now that maybe some more context is needed to help me choose the path at this crossroads.

I can keep the coil wire thin and aim for 250 V at 20 m/s so that the midnight classic can still do its MPPT job.

Or I can keep it low tech and just install bigger wires and aim for 70V at 20 m/s and let the battery pack deal with things being guarded by a cheap charge controller that theoretical will hit the brakes at 60V input.

EDIT" the midnight classic, although good stuff, is probably not applicable to 3 phase alternators. It has a dc input. that is too bad.

[MattM]

You probably want two contact points for each rotor arm or twisting forces from wind oscillation will destroy them quickly.

Did you happen to look at the 'ceramic magnetic phenomenon' I was trying to describe?  I think the steel bar allows two same-polarity magnets attached to it to compound into the same area.  Other ways of trying to compound magnetic force just seem to add distance whereas this keeps them equidistant.  The greatest force seems to concentrate into the center of the bar, and distance between magnets has to be fine-tuned to see best results.  Just a theory, but the phenomenon isn't documented anywhere that I could find.

[brandnewb]

Yes thank you Matt for the heads up regarding the turbine. I will do my best hooking up the blades to the arms next to it and bring stability in that way. But it might turn out to be a dead end and i'll introduce bottom arms again.

Also thank you for this ceramic configuration with steel. I was a little ashamed to tell you that I don't really understand yet ;( I'll let it sink in for a while before going to try and make it with neos.

I am looking forward to the oscope readings It will be educational I am sure.

[MattM]

I had mixed results using neos doing the same trick.  While neos are extremely strong, the distance that their fieldlines induce a pull appears to be noticably shorter.  If you want a compact generator there is no substitute for neos.  Those ceramic magnets are very cheap, often 1/10th the price of good neos.  For such a large radius as you are aiming for it would be a much cheaper cost for prototyping.  Just my two cent, peanut gallery opinion.

As for the arms of your turbine, it is important to keep as much weight at the center of gravity (COG) as possible.  Science dictates you will have less force required for accelleration by keeping mass nearest COG.  As mass is added away from COG it will have more resistance to accelleration.  On the flipside, objects in motion tend to stay in motion, so your rotational rate will vary less to force fluctuations (e.g. wind) as mass is added away from COG.  Slow will stay slow but at least it won't be as temperamental.  So if you are aiming at longevity embrace the lower velocity.

[brandnewb]

thanks Matt, I'll keep it mind.

I have another reason going for neos though. I'll explain in a bit.

First a quick update.

I have decided to go ticker wire lower voltage. The advantages are multifold.
* Less heat build up
* Higher amperage ability before hitting heat issues, although technically I think an MPPT controller should be able to keep the amps away from the generator and produce amps in the controller.
* Can make use of cheaper electronics that are easier to replace.
* and last but not least. Now there is ample space between the coil wires for it to fill up with iron filled resin keeping the wires firmly in place while reducing the flux resistance.
This particular iron powder I am using does not conduct electricity so I am not worried about any negative effects of eddy currents.

Now the reason for sticking to neos is that in case I am running into difficulties reaching my target voltage that I need to set yet.
I still have space to just double the magnet count. So no stacking just traditional. I really believe that is the best bang for the buck over any exotic configuration.
As slowly I am beginning to read between the lines of what is being written (like frequency of field reversal).
You see I, by mistake, configured once all poles same direction and still it produced(albeit a super low) voltage. Certainly not 0. That is because not all magnets are equal strong and or the exact same distance from the coil.
So it is the velocity of a change in magnetic field that is the only important factor to aim for. Hence the frequency of field reversal as going from one pole to the opposite has the greatest delta.
But some experiments will tell if I am arriving at the correct conclusion.
For target voltage the last value I arrived at was 70V at a blade tip speed of 20m/s. But those are just ramblings that are not thought out yet. I do not even know the relation of the blade speed to wind speed yet

And as a last fall back I have is going from a direct drive to a geared one so I can increase the velocity at which the alternator spins. But that would introduce potential points of failure that I am looking to keep to a minimum.

[brandnewb]

The magnets to coils ratio was mentioned here earlier.
This time I searched for "magnets to coils ratio" and voila. I found 2 results with one on this forum.

But those do not mention to the why to have this ratio. But I believe it mentioned in an ancient past that it is to help reduce with what I call cogging. That the magnets are attracted to a coil (because of the steel plating) and then resist moving and when they do they jump to the next. If this is correct I am again at a cross roads.

Since now the coils will be stacked the distance between coils is soo little that this "cogging" should be minimal and if I also fill up the space in between coils with iron filled resin then it should be gone completely. But then the mag field has an easier path with less resistance around coils so I am not sure I want to do that.
Or abandon the idea of iron filled resin completely.

One thing I can't do is have all coil slots filled with wire if I want creating coils to stay practical. And if I end up doubling the magnet count again then that might also help reduce cogging as the coil size will stay the same else end up with no space for coils.

Can anyone please shine a light on why this rule came to be in the first place?

[bigrockcandymountain]

I realise now that maybe some more context is needed to help me choose the path at this crossroads.

I can keep the coil wire thin and aim for 250 V at 20 m/s so that the midnight classic can still do its MPPT job.

Or I can keep it low tech and just install bigger wires and aim for 70V at 20 m/s and let the battery pack deal with things being guarded by a cheap charge controller that theoretical will hit the brakes at 60V input.

EDIT" the midnight classic, although good stuff, is probably not applicable to 3 phase alternators. It has a dc input. that is too bad.

I think 70v at 20m/s might be too low.  Are you 24v or 48v? Generally aim for just over your battery voltage at 5m/s. 

The classic 250 is happy to have input from a 3 phase alternator.  It just has to be rectified to dc first.  Kitestrings and I both have our systems set up like this.

I would encourage you to build it first and do a bunch of testing before you buy a classic, but don't rule it out. 

[brandnewb]

Thank you for this input.

I have a 16S (48V) Lifepo4 battery bank.
So if I aim for 52 V at 5 m/s then I forsee no issues with using loads of 0.4mm diameter wires in the coil.

And then when we have something like last weekend and the wind goes crazy. Does a proper MPPT like the midnight classic 250 happily convert the much higher voltage to a lower one thus keeping the amps away from the alternator?

[MattM]

brandnewb-

To keep loads symmetrical.  If your loads are assymmetrical you create load oscillations.

[bigrockcandymountain]

Does a proper MPPT like the midnight classic 250 happily convert the much higher voltage to a lower one thus keeping the amps away from the alternator?

Yes.  It can run the input voltage up to 250v without damage.  It also has aux terminals that can be set to do a lot of different things.  One of mine switches on a resistor bank on the 3 phase side when the input voltage goes too high.

I limit my dc amps to below 30a on the input of the classic by letting the voltage rise.  That makes the amperage on the 3 phase side even less than that. ( 3 wires vs 2).

[brandnewb]

Thanks guys for the additional info. I will take it into account.

I just ran an experiment to see if iron powder filled coils make sense and I think they do.

this test makes use of 2 iron powder bars connected to each magnet. The powder starts after about 2.5mm of print layer height.

I measured without and with powder bars at roughly 16mm distance.
without is around 35 mt and with around 65 mt
incredible !!

now obviously i can't put as much powder in the coils as most space will be wires but I do think it can help if the field will not start to meander around each coil wire.

This did not happen this time because the 2 bars aren't connected so no where to run for the field

I tried using the oscope but I can't figure it out while I am still looking for a manual ;( I have put it in the photo so perhaps one recognizes it and know where I can find a manual .
EDIT: I found a different unit that looks exactly the same that does have a manual. So never mind.