How to determine the limits of a coil / magnet configuration

[brandnewb]

As I am working on some test coils and test magnet holders. I thought it important to also be able to test for limits

I plan to test a variety of configurations.

The idea is to have a pure epoxy resin cast coil as a base line. And several other variations to see what the effects of those additives are.

And have several variations of magnet configurations spin over them while the coils are shorted.

Would it be ok for me to drive my alternator test bench with a 20KW AC variable speed motor?  if at all affordable and required.
I can supply a continues power of 11 KW. spread over 3 phases. I am assuming this will enable me to supply the power needed to start seeking the limits of a coil magnet configuration. In case that is not enough I can temporary bypass the my circuit breakers in the part of the electrical breaker cabinet that is my responsibility and for short bursts supply a bit below 19.83 KW before the main breakers ( that are not part of my electrical breaker cabinet) trip.

I am hoping that this will enable enough power to start seeking limits.

I am asking for someone to tell me that this is a good way or how to do it even better.

Or I am kind of hoping I do not need to go that large to begin with. I mean if I can also reach conclusions that people would trust with much less power then I am all for that!

[joestue]

The reason to do the test as I suggested is because it doesn't require you to measure the torque on the coil by the magnets. The torque is constant, by gravity. measuring rpm is easy, measuring torque isn't, especially if the torque is fluctuating due to the torque not being constant, because the rectifier either doesn't draw constant power, or its a single phase test coil.

you can do the tests with a single test coil, and the results will correlate well to 3 test coils connected in delta. but they won't when connected in Y. the reason being the third harmonic is canceled out and "missing".

so for example on my 3 phase generator, the phase to neutral voltage was 138 or maybe 142, but the phase to phase was 230. 10-16 volts went missing. however, it did produce a very nice sine wave.

basically all you need is a test coil, a scale to weigh it, then wind up the string around the generator shaft, short the coil, and measure the peak rpm of the rotor. then compare with thinner coils and closer magnet spacing, or wider coils, etc.. or coils with bigger or smaller ID or also the OD.

[brandnewb]

ok brother.

I will first do as you suggest then!.

Thank you.

It will be soo much less hassle than getting out the big guns.

[brandnewb]

Can you please elaborate a bit on what is OD and ID.

And what is it I am testing for again?

In my previous thread you mentioned we are looking for the sweet spot for how much mechanical resistance the coils can provide to the magnets to produce amperage or something to that extend.

Is this still related?

[Bruce S]

ID= Inside dimension
OD= Outside dimension

When winding coils (even if it's only for a joule thief, though for a joule thief it's usually just a couple turns) I use the metric system, along with calipers with a digital readout.

Cheers
Bruce S

[brandnewb]

sorry guys. I think I am just losing sight of what one meant.


Why not I get out the big guns and starting blowing up some coils ?

Nah but seriously. I am not sure what good it would do if I would weigh my coils.

For one thing I started with a spreadsheet for one of the 0.4mm diam wired coils at 300 winds / 600 turns. it weighs 92 grams.

I already found the material I need to setup the experiment like described but I am not really sure what it is we can learn here. I am sure there is a lot but I am just not sure.

Can I please have some more context?

[JW]

   

 

[MattM]

Temperature plays a role.  You may hit temperature issues before maximum watts or volts.

[joestue]

Can you please elaborate a bit on what is OD and ID.

And what is it I am testing for again?

In my previous thread you mentioned we are looking for the sweet spot for how much mechanical resistance the coils can provide to the magnets to produce amperage or something to that extend.

Is this still related?

The point is to find the optimal inner and outer diameter and thickness of a coil that provides the most resistance.

The problem is, the magnet's radius is also a variable. You cant have the magnets too far apart, because when you connect all the coils together, the current is flowing through other coils which are not producing voltage.. so ideally you need 3 test coils, connected in Y, but shorted. Or perhaps better, rectify the test coils and short the rectifier.

So you want to keep your magnet and coil ratio, and magnet spacing (both are variables you can change) such that you ether get a clean sine wave, or a trapezoid, but if your magnets are too far apart, you get a sine wave ish pulse, followed by dead space, then a pulse of the opposite polarity, and this will be worse.. unless you rectify every coil separately.

 

[brandnewb]

understood sir!

Yet I am 99.999% certain I can generate a clean sine wave or a clean trapezoid. Early tests with oscliscopes suggest I can do at least the sine wave.  Going for trapezoid would mean going stacking sideways like Mary suggested or some other exotic configuration one can think of. I like sine waves though

So the goal is to have weight on a tether pulling the test bench and measure the time it takes for it to reach the ground? the longer it takes the better it is?

Or is the point to have the weight on the tether pull a configuration along and see what generates the highest amperage? Is that even possible with shorted coils? I guess so as what is the difference with a load like a resistor?

Where does the weight of the coil come into play again?

[JW]

 
 

Yet I am 99.999% certain I can generate a clean sine wave or a clean trapezoid. Early tests with oscliscopes suggest I can do at least the sine wave.  Going for trapezoid would mean going stacking sideways like Mary suggested or some other exotic configuration one can think of. I like sine waves though

" A circuit generates three-phase sine waves with excellent amplitude and phase symmetry. A variable-frequency, three-phase, sine-wave generator circuit has been designed for use as a source of polyphase excitation in studies of the propagation of traveling waves in plasmas.Mar 27, 2020

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

3 phase poly wave sine wave.

 

[JW]

https://www.fieldlines.com/index.php?topic=146386.0

[brandnewb]

Ok Sir,

Please see if I am following your lead well enough of that I need to make changes.

I have the gravity torque setup ready now. I think.



And this time I will test the failed 6 coil stacked 3 phase arrangement.

It failed because the coils moved during the curing so now there is no longer a perfect 1.25 degree of separation between each phase.
Also this test is too thick and too heavy and also just not homogeneous enough.

Nah I will do it much better next try.

{1}Also a single coil of 300 winds at 0.4mm is 92 grams net. So excluding any resin around it.{1}

Can you please elaborate a bit on what is OD and ID.

And what is it I am testing for again?

In my previous thread you mentioned we are looking for the sweet spot for how much mechanical resistance the coils can provide to the magnets to produce amperage or something to that extend.

Is this still related?

The point is to find the optimal inner and outer diameter and thickness of a coil that provides the most resistance.

The problem is, the magnet's radius is also a variable. You cant have the magnets too far apart, because when you connect all the coils together, the current is flowing through other coils which are not producing voltage.. so ideally you need 3 test coils, connected in Y, but shorted. Or perhaps better, rectify the test coils and short the rectifier.

So you want to keep your magnet and coil ratio, and magnet spacing (both are variables you can change) such that you ether get a clean sine wave, or a trapezoid, but if your magnets are too far apart, you get a sine wave ish pulse, followed by dead space, then a pulse of the opposite polarity, and this will be worse.. unless you rectify every coil separately.

 

[brandnewb]

it is confirmed. having this many wires so close together in a coil set negates the whole thermal dissipation ability.

Now I am only able to have the heat stabilized at 70C when running 9 VAC at 2.6 amps through all 6 coils wired in parallel.

I will be trying again this time with half the amount of wires.

{1}I was referring to earlier tests like this on single coils. One traditional and one with iron powder. The iron powder one showed promise in both heat dissipation ability and voltage generation ability. It is outlined in my diary about the VAWT backyard experiments.{1}

[brandnewb]

I could not resist testing this failed coils anyway for voltage.

Now I am not really experienced with an oscope so if someone can help figure out what is says here.

I am interested in the frequency which should be really low and the voltage.

please also note that the magnets holder disk parts were spaced apart far more than the intended 28mm due to the thickness of the failed coils casting.




Ch1 is coil4 in isolation
Ch2 is coil2 and 5 in series.
Ch3 is coil3 in isolation.

Also the coil 1 shorted out. Could that have happened during the heating test? and if that is the case that would basically mean the end of iron filled resin coils.

I will make sure to properly cure the next stacked coil casting try and when fully cured give the coil an AC beating for like 24 hours and see if coils start failing.

[Mary B]

What volts per division did you have the scope set at? 500mv(.5v)? just count the number of divisions top to bottom to get your peak AC voltage...

[brandnewb]

yes 500mv. but mostly it is the frequency I have difficulties with distilling out of what I see on the screen. I try to rotate the magnets over the coils at a speed that resembles around 3.8 HZ. that is the target frequency

[Mary B]

yes 500mv. but mostly it is the frequency I have difficulties with distilling out of what I see on the screen. I try to rotate the magnets over the coils at a speed that resembles around 3.8 HZ. that is the target frequency

Your peak AC voltage doesn't care about frequency, rectified DC will somewhat with lower frequencies = lower rectified DC... typical bridge rectifier is AC voltage multiplied by 1.4 for a full wave bridge rectifier(electrolytic capacitor filter), there are also voltage doublers that can increase voltage but cut current in half... that is 2.8 x AC volts

[brandnewb]

I understand Mary,
But I was not yet thus far as to deal with rectification.

I am interested to learn how to deal with an oscope as the manual is not getting me there. I have tried please believe that.

The context I am still in is a direct drive. This means that the magnets reverse field at a given Hz at a given wind speed. in this context there is no gear mechanism to change that. Other than doubling the magnets to double the frequency of field reversal.

So in order for me to be able to reach a target voltage at a given wind speed (direct drive PMA) I need to be able to read the frequency of field reversals of the waveform.

Can you help figure that out based on what I showed?

If not that I can respect that as it might be that the interface in not adhering to standards so I will try going the costumer support route if that is any good.

[Mary B]

https://www.dummies.com/article/technology/electronics/general-electronics/measure-electronic-waves-how-to-use-an-oscilloscope-180231/

Flat line on your scope is zero volts, peaks below it are negative, above it positive. Time per division is frequency and info on that along with the math is here https://sparks-activities.concord.org/sparks-content/tutorials/tutorial-25.html

With multiple coils you are going to be well above 60hz so the rectifier formula I gave will be close enough. So you are showing approx 3.5 volts peak to peak with your scope set at 500mv per division(how AC is measured). Rectified that would be 4.9 volts from just 1 coil... now add in all the other coils... say you have 6 coils per phase, or 3.5 volts x 6 = 21volts AC and rectified/filtered 29.4 volts. Enough to charge a 24 volt battery bank!

[JW]

for your application you should use an analog 12vdc meter. The Oscope is giving to much information.

Look for analog voltmeter (with a needle)

[brandnewb]

Wicked

winding narrow is awesome!

I can compress these 300 winds coils within 15mm thickness.
And all read around 9.6 ohms.

Now if only I can bring the heat dissipation ability back then things will be insane by what I have seen yet in my life !

[brandnewb]

BTW, most, if not all, modern smartphones come with a magnetic field sensor these days. probably a hall effect type of thing like on an actual probe.

Now this works rather nice and accurate other than of course there is no probe you can stick into nooks and crannies.

I tried
https://play.google.com/store/apps/details?id=com.appdevgenie.magnetometer
and was rather impressed.
No need for any permissions other than this 1 sensor and also no tracking and/or other spyware detected by the exodus scan.

[brandnewb]

I have it on high authority that these waveforms somewhere fall around the 6hz mark.
And I fully trust this guy's ability to quick scan my photo and sling out an approximate number.

Now as I am getting more able to also try my self I get around the 5Hz mark using
https://www.unitjuggler.com/convert-frequency-from-ms(p)-to-Hz.html?val=200

Now whether it is 5 or 6 Hz really does not matter at this moment in time.
All that matters is that I spun the magnets too fast!!

But now at least I can start aiming much better.

60Hz is what one would get from a mains in the states. Over here it would be 50 Hz.

This is also why I am reluctant to abandon iron powder coils just yet. I mean at around 6 Hz there should not be enough energy for the 150 um iron particles to cause damage to the enamel of the copper wires when fully encased in a properly post cured casting.

Well that is my hope at least.
Lets see how things turn out in practice.

[brandnewb]

for your application you should use an analog 12vdc meter. The Oscope is giving to much information.

Look for analog voltmeter (with a needle)

I am worried JW that a 12 VDC meter is not going to be of use when dealing with around 52 VAC at around 3.8Hz.
Or if/when I decide to go full force. Then with ease more than 200 VAC at around 3.8Hz. I also think this oscope is not designed to deal with those rather high numbers

[JW]

You need to use the 12volt because it will give you the best response time, you can always ad a resister in series on one side of the circuit.

Yes the Oscope measures time into the equation. However many things have to have a "feel". Regardless what you do, to not use an inverter? power in watts is what counts. Some of the things you describe don't account for ohms law while considering power in watts.     

[JW]

virtually every system we describe here uses a battery bank since the power from the mill is variable. It doesn't have the power continuous. IE it takes time to recharge the battery bank with the windmill.

https://www.electrical4u.com/half-wave-rectifiers

https://www.electronics-tutorials.ws/diode/diode_6.html

[Mary B]

for your application you should use an analog 12vdc meter. The Oscope is giving to much information.

Look for analog voltmeter (with a needle)

I am worried JW that a 12 VDC meter is not going to be of use when dealing with around 52 VAC at around 3.8Hz.
Or if/when I decide to go full force. Then with ease more than 200 VAC at around 3.8Hz. I also think this oscope is not designed to deal with those rather high numbers

What JW is getting at is an analog meter scale is easier to read... any old analog voltmeter will do a couple hundred volts on the top scale, and as little as a volt on the bottom scale... I learned on a Simpson 260!

[brandnewb]

I am reading this over and over again yet still am a little worried that my current gravity torque setup is not as intended.

For example; the weight I put on the string. Should it be similar to the coil weight?
Now although I did not measure the weight of the new coils yet we can deduce it based on that an 11 ohms coil is 92 grams. The new coils are 9.6 ohms.
But lets avoid math for now I will measure the actual weight later on.

And I think I should be testing not a single coil but at least 6 of them 3 phase wired in series yes?

The reason to do the test as I suggested is because it doesn't require you to measure the torque on the coil by the magnets. The torque is constant, by gravity. measuring rpm is easy, measuring torque isn't, especially if the torque is fluctuating due to the torque not being constant, because the rectifier either doesn't draw constant power, or its a single phase test coil.

you can do the tests with a single test coil, and the results will correlate well to 3 test coils connected in delta. but they won't when connected in Y. the reason being the third harmonic is canceled out and "missing".

so for example on my 3 phase generator, the phase to neutral voltage was 138 or maybe 142, but the phase to phase was 230. 10-16 volts went missing. however, it did produce a very nice sine wave.

basically all you need is a test coil, a scale to weigh it, then wind up the string around the generator shaft, short the coil, and measure the peak rpm of the rotor. then compare with thinner coils and closer magnet spacing, or wider coils, etc.. or coils with bigger or smaller ID or also the OD.

[brandnewb]

84 grams the new coils weigh

[JW]

And I think I should be testing not a single coil but at least 6 of them 3 phase wired in series yes?

No there is star and delta for 3 phase

[brandnewb]

And I think I should be testing not a single coil but at least 6 of them 3 phase wired in series yes?

No there is star and delta for 3 phase

understood sir. Thank you.

I will also play around with star and delta.

[JW]

with delta there is more power, with star the impedance is higher but at the nexus point of star is a neutral (4th wire) not possible with delta.