VAWT, PMA, stacked stators and rotors

[GoVertical]

Hi, I completed fabrication of PMA proto type. The configuration does work. I am now working on bench test fixture so I can measure values at different RPMs using a 12 volt deep cell as a test load.

Posting known specs of newest PMA configuration:
3 phase Star/ 12 coils per phase
coils: 2 in hand 22 AWG, 100 turns
phase resistance 9.7 ohms, measured at output to output
16 magnets/rotor, 1 x ¾ N42 cylinders

Measured using 12 volt battery as test load:
190 RPM
Vdc = 11.5
Amps dc = 3.5

I will post more measured values when available.

The largest stator I can fabricate has a 10 inch diameter. I am interested in generating a high output at a low RPM. Is it better to use more small magnet or fewer large magnets for the given stator size?
Comments welcome.




 

[jlt]

       What Is the diameter ,and are all three stators connected in parallel or series.looks like neat work .But very complicated. 

[GoVertical]

Hi, max size of stator diameter that I can fabricate is 10 inch diameter.  I have established the configuration does work. I am trying to do some fine tuning. Search for info for best magnet size and wire size to produce best output performance for 100 to 500 RPM range.  Each stator has input and output connections so I can add more stators and rotors if required. This also allows for different wiring schemes, which may be useful when trying to match PMA to different blade designs.  I am just starting to explore the limits of the configuration.
Assembly is very user friendly. One down side it does require a lot of shaft collars to position the stators. I was unable to purchase the required amount locally, which forced me to fabricate most of them.  The configuration does require less magnets than a dual rotor and does lower the needed RPM for cut in. I will post more number when available. Comments welcome.   

[scoraigwind]

Hi, max size of stator diameter that I can fabricate is 10 inch diameter.

The biggest improvement you could make to this is to break free from your self-imposed diameter limit.  Then you could put all of the magnets on 2 disks and all of the coils in one stator and get 2-3 times the output, from the same amount of material.  Magnets are expensive.  Make them work harder by using all of the magnets on each coil.  You gain nothing by dividing them up into smaller rotors like this.

[electrondady1]

Oden, i admire your building skills and  tenacity in making your layout work.
but what Hugh has posted is absolutely true.

i build drag mills as well.
ihave  even built a four rotor, three stator stacked alternator like you have done.
although mine was not so pretty as yours and used ceramic mags.
it never got off the bench.

i abandoned the idea of completely encapsulating the alternator within plastic pipe.
in favour of a  simple weather shield over the alternator wide enough to keep the stator dry.

 

[GoVertical]

Hi, max size of stator diameter that I can fabricate is 10 inch diameter.

The biggest improvement you could make to this is to break free from your self-imposed diameter limit.  Then you could put all of the magnets on 2 disks and all of the coils in one stator and get 2-3 times the output, from the same amount of material.  Magnets are expensive.  Make them work harder by using all of the magnets on each coil.  You gain nothing by dividing them up into smaller rotors like this.

Hi, this is just a experiment. Can you provide test data from a 3 phase Star/12 coil per phase using the same size coils and magnets as a dual rotor to support your statement?

Project update:
I fabricated a test bench fixture to test outputs at different RPMs. I have only been able to gather data at 2 speeds using a battery as a load.
RPM          Output  Voltage   DC amps
100                          12            1.0
200                          12            3.8


The lathe motor I am using is unable to power the PMA above 200 RPM. At 200 RPMs the lathe motor shuts off and starts again and the cycle repeats. Is there a better motor that I can use for a bench test?  I am increasing the wire size to the load to see if performance will improve. Comments welcome?









Using the PVC fittings was to difficult to work with. Thanks

[lohearth]

Oden
 you may need to use some sort of switching scheme. Try using just 2 of the 3 stators in parallel instead of series at 200 rpm

[ghurd]

Hi, this is just a experiment. Can you provide test data from a 3 phase Star/12 coil per phase using the same size coils and magnets as a dual rotor to support your statement?

It takes a calculator.

About 150 years of test data, mathematically, come down to:
(a bunch of magnet and wire stuff) X Frequency = Power

Placing the magnets over a whole bunch of different disks makes less frequency on each coil, so lower power output.

It seems you want to look at each individual 'layer'.
Each layer has less magnet surface area, included in "(a bunch of magnet and wire stuff)" above, and less frequency.

OK.
Think about it this way.
It is not exactly correct concerning all the parameters, but this is the general idea.
How much magnet surface area is there, combined?
That is what you have to work with.
Spreading it out over more disks makes a lower frequency, and a lower frequency makes less output.

=

Or maybe think about it in something more mechanical.
A stupid example, but maybe the concepts will get through.

A riding lawn mower has 10 HP.
Gear it up and it can go really fast.  Therefore it can go really fast.
Fear it down and it can tow 1000 tons.  Therefore it can tow 1000 tons.
Gearing it up 3 times, and gearing it down 3 times, does NOT mean it can tow 1000 tons really fast.

You understand that, Right?

Stacking the stators is like gearing it down.
Seriesing the stators is like gearing it up.

The motor is still 10HP.

The difference is a whole lot of money spent on a whole lot of extra parts (the gear boxes).

Instead of spending your money on gear boxes, you are spending your money on magnets and wire.

The motor is still 10HP.

I could argue the motor output is LESS than 10 HP with all the gear boxes, because of the friction in the gears.


"Electrical Friction" is resistance x current.
Amps X amps X ohms

Shall we say your coils have 10 ohms resistance?
Passing 1A wastes 10W in "the gear box".
Passing 2A wastes 40W in "the gear box".
Passing 3A wastes 90W in "the gear box".
Passing 3.8A wastes 144.4W in "the gear box".

12V at 3.8A is 45.6W.
144.4W is lost in heat.
That makes a total of 190W it takes for the input to get 45.6W out.
24% efficiency.  That sucks.

The gear box is well made, but it is sucking up all the power and turning it to heat.

[GoVertical]

Hi, very will explained.

Project update:  resolved drive belt slipping issue. I connected the three stators as 3 phase star with 4 coils per phase then connected each stator in parallel. Measured dc amps = 4.7 @ 300 RPMs, battery load voltage = 11.7, internal resistance measured from phase output to phase output = 1.4 ohms. PMA running smooth on test fixture.

Thank you for taking the time to comment.  I can add more stators or increase the magnet surface area on the rotors to improve performance? Comments welcome.

[GoVertical]

Hi, I am finding adjusting the belt pulley ratios very confusing. I just measured 6.5 amps @ 360 RPM

[Rover]

How are you measuring RPM?

[GoVertical]

Hi, I am using reflective tape on the top rotor of the PMA and hand held optical RPM meter. I hold the meter very close to the top rotor so there is no reflection from the magnets just the tape.  I will recheck my reading tomorrow and try to provide data from the limited RPM range I can generate with the current test setup.     

[scoraigwind]

Hi, max size of stator diameter that I can fabricate is 10 inch diameter.

The biggest improvement you could make to this is to break free from your self-imposed diameter limit.  Then you could put all of the magnets on 2 disks and all of the coils in one stator and get 2-3 times the output, from the same amount of material.  Magnets are expensive.  Make them work harder by using all of the magnets on each coil.  You gain nothing by dividing them up into smaller rotors like this.

Hi, this is just a experiment. Can you provide test data from a 3 phase Star/12 coil per phase using the same size coils and magnets as a dual rotor to support your statement?

I don't need to do tests, I can predict the output of alternators with great accuracy using a simple calculation based on the size and number of magnets, and this calculation tells me that you will get 3 times as much voltage if you put all 3 alternators together into one with larger diameter.  Whereas the resistance will be exactly the same.  So it is pointless to stack them axially in a series of separate units like this.

[keithturtle]

OK, then, if I get the drift of this thread, I may indeed be on the right track with the attached design.  Forty magnets and thirty coils (3-ph, 10 each in series) and bolted to the bottom of a 6' diameter VAWT.  ( VAWT is not the issue at this point)

Still in the development stage, but the rotors are 19mm thick steel with 13mmx13mm N-45 neos

Turtle, slow

[DanG]

"Relative movement" - coil turns & magnet strength & speed which with magnet - coil interact.

As rotor diameter increases the magnets apparent speed across the coils increases, the value of the induced power can be increased by using a stronger magnet or by moving the wire faster - so there is your limiting variable.

[GoVertical]

Hi, max size of stator diameter that I can fabricate is 10 inch diameter.

The biggest improvement you could make to this is to break free from your self-imposed diameter limit.  Then you could put all of the magnets on 2 disks and all of the coils in one stator and get 2-3 times the output, from the same amount of material.  Magnets are expensive.  Make them work harder by using all of the magnets on each coil.  You gain nothing by dividing them up into smaller rotors like this.

Hi, this is just a experiment. Can you provide test data from a 3 phase Star/12 coil per phase using the same size coils and magnets as a dual rotor to support your statement?

I don't need to do tests, I can predict the output of alternators with great accuracy using a simple calculation based on the size and number of magnets, and this calculation tells me that you will get 3 times as much voltage if you put all 3 alternators together into one with larger diameter.  Whereas the resistance will be exactly the same.  So it is pointless to stack them axially in a series of separate units like this.

Thanks, I tried connecting them in parallel and the internal resistance drops every time I add a stator. Does your formulation still project the same results?

[GoVertical]

New connections

[Dave B]

Oden,

  This is great work and it certainly reflects your passion for this design. Others may have valid points regarding the efficiency and best use of magnets and coil material etc. but this is comparing your project to a different design. Others so quick to ask (why ?) would you want to do it this way if this other way is so much better I think maybe are missing the point of your own passion for  your own particular project. You are enjoying building this design and lest we not forget that this is exactly how this discussion board got it's start. Do what you love and love what you do, great project and excellent work.  Dave B.

[GoVertical]

OK, then, if I get the drift of this thread, I may indeed be on the right track with the attached design.  Forty magnets and thirty coils (3-ph, 10 each in series) and bolted to the bottom of a 6' diameter VAWT.  ( VAWT is not the issue at this point) (Attachment Link)

Still in the development stage, but the rotors are 19mm thick steel with 13mmx13mm N-45 neos

Turtle, slow

Hi, I hope to see some photos of your project. How are dealing with the internal resistance problem of have so many coils in series? Ghurd made a very valid point regarding the issue and when I had my PMA connected as a 12 coil/phase I was measuring very poor results. Because of my stator and rotor configuration I was able to connect each stator in parallel which appears to have solved the problem. The internal resistance is much lower and the output is much higher.

[ChrisOlson]

Thanks, I tried connecting them in parallel and the internal resistance drops every time I add a stator. Does your formulation still project the same results?

Hi GoVertical - first of all, nice work on that generator!

I have built several dual stator 12 volt generators.  I can't see how you got your magnets arranged, but when I built mine I did not put magnets on both sides of a center rotor because it would totally saturate the plate.  I had an air gap of about 1/2" between the center rotors, with magnets only on one side of each.  So basically, I had two stators, four rotors, all stacked on the same shaft.

I think looking at internal resistance doesn't really mean a lot.  Each generator is its own unit when they're connected in parallel (which is what I did).  The basic rule of thumb I came up with is that if you have one generator and it puts out 500 watts, put two on the shaft and you get 1,000 watts. Period.  As long as the blades got enough power to drive it, that is.  I had a 12 volt dual stator generator on one of my turbines that would deliver 120 amps sustained on a 12 volt system.  Try to do that in one stator and you'll burn it up unless you wind it with rebar.  The dual stator unit handled it with no problem.  The rectifiers were another issue.  But that's a different topic.

Nice work!
--
Chris

[GoVertical]

Thanks, I tried connecting them in parallel and the internal resistance drops every time I add a stator. Does your formulation still project the same results?

Hi GoVertical - first of all, nice work on that generator!

I have built several dual stator 12 volt generators.  I can't see how you got your magnets arranged, but when I built mine I did not put magnets on both sides of a center rotor because it would totally saturate the plate.  I had an air gap of about 1/2" between the center rotors, with magnets only on one side of each.  So basically, I had two stators, four rotors, all stacked on the same shaft.

I think looking at internal resistance doesn't really mean a lot.  Each generator is its own unit when they're connected in parallel (which is what I did).  The basic rule of thumb I came up with is that if you have one generator and it puts out 500 watts, put two on the shaft and you get 1,000 watts. Period.  As long as the blades got enough power to drive it, that is.  I had a 12 volt dual stator generator on one of my turbines that would deliver 120 amps sustained on a 12 volt system.  Try to do that in one stator and you'll burn it up unless you wind it with rebar.  The dual stator unit handled it with no problem.  The rectifiers were another issue.  But that's a different topic.

Nice work!
--
Chris

Hi, no metal in the middle rotors.  The inner smaller magnet holds the larger ones in place with no glue. They stay in place really will, tested at over 1000RPMs with no problems.  I did find lowering the internal resistance improved the output for this configuration dramatically. 12 coils in series per phase measured over 9 ohms and the highest I could measure using my test fixture was 3.6 amps dc. Because the stators have inputs and outputs I wired each stator as a 3 phase with 4 coils per phase and then connected each stator in parallel. This lowered the internal resistance to 1.4 ohms and I was able to measure 6.5 amps without changing the belts on the lathe. Do you think your chain drive could be adapted for a vertical application? The blade design I am testing is a low RPM, high torque configuration and adding a chain drive would solve my low RPM problem. 

[vawtwindy]

congrats govertical.

[ghurd]

From sketch in Reply #16, "Center tap and outputs are then connected in parallel".

Not a good idea.
That configuration allows circulating currents between stators.  Basically, in mechanical terms, 'it leaks'.

Better to rectify each individual stator, then parallel the DC outputs.
That will stop the 'leaks', and should increase efficiency a bit more.


Now the resistance is lower, the efficiency is higher, and the same lathe motor is able to turn it without shutting down from over load, and make about double the output power.
Now the resistance is lower, there is less heat in the stators.
G-

[GoVertical]

Oden,

  This is great work and it certainly reflects your passion for this design. Others may have valid points regarding the efficiency and best use of magnets and coil material etc. but this is comparing your project to a different design. Others so quick to ask (why ?) would you want to do it this way if this other way is so much better I think maybe are missing the point of your own passion for  your own particular project. You are enjoying building this design and lest we not forget that this is exactly how this discussion board got it's start. Do what you love and love what you do, great project and excellent work.  Dave B.

Hi, yes I am obsessed with this I idea. I am trying not to let my personal biases or passion for this project mislead test results.  It has taken sometime to just to gather the resources just to able to fabricate anything.
If the idea is a bad idea I will still have the resources to fabricate a dual rotor.  To be able to fabricate a device that converts free wind energy to electricity, I am unable to see a down side. Thank you for all the help I am receiving. 

[electrondady1]

i think we are all obsessed with these machines Oden.
 otherwise, we would just give up and do solar )

when your finished, i recommend you leave this machine up and working rather than tear it up and scavenge the parts.
and if you want to try another configuration, start it from scratch.

i'm just a romantic at heart, to me windmills are a bit like boats.
you build it ,give it a name, launch it , and then it takes on a life of it's own.

[Volvo farmer]

i'm just a romantic at heart, to me windmills are a bit like boats.
you build it ,give it a name, launch it , and then it takes on a life of it's own.

You mean that a wind turbine is a hole in the air that you throw money into?

[ChrisOlson]

I see how you did the center rotors now.  Quite interesting.

Ghurd makes a good point about paralleling multiple stators - I found that paralleling the AC output did not work that well because I could not get them precisely in time.  So it's easiest to do the paralleling at the DC output.  That's what worked for me.

As to using my drive system on a vertical machine, yes, it could be done.  For the same level of reliability I have gotten in the horizontal version, you would have to incorporate a guide on the slack side of the chain to prevent gravity from making it sag after the links and pins break in, which would cause excessive side wear to the high speed pinion.  You would also have to design an oil slinger for the high speed shaft that would pull oil out of the sump and sling it up on the chain.  Even on my horizontal machines I use the slinger in the last three I have built, rather than "dipping" the chain in the sump oil.  So yes, it would work fine.

When I build my generators, keeping the diameter small is what I strive for.  Since trying geared drives I have found direct drives to be boring and have found ways to get more power from geared systems than is possible from direct drives without spending a LOT of money on the direct drive generator and ending up with a monstrosity as big as a truck tire.  With your diameter under 10" you have a very good system there that would work very nicely with gearing I think.

Overall, I like your design.  Engineers have for years used parallel systems to increase power output; instead of using one big locomotive to pull the train, several smaller ones are hooked on.  Instead of the airplane having one huge engine, it has two or more smaller ones.  Instead of using one huge battery in your battery bank, you use several smaller ones.  Using parallel systems to increase power has merit and it's rather disappointing to read the negative comments from folks who insist that it can't be done, it won't work, or you have to use one big system to make it efficient.  If it weren't for the experimenters and inventors trying things that consensus says won't work, the wheels of progress would be seized up.

Thanks for posting your project - it's one of the better ones I've seen on the board!
--
Chris

[GoVertical]

Today's test result:

1 Stator      Phase output to phase output = 4.3 ohms

load voltage   dc amps      PMA RPM
11.67      0.38             190
11.75      2.06             310
11.86      3.8             440      

2 Stators      Phase output to phase output = 2.4 ohms
11.61      0.51             189
11.74      2.6             288
11.85      3.8             340

3 stators      Phase output to phase output = 1.9 ohms
12.0             0.03              150
11.41          1.56              186
12.6             2.95              270

Each stator was configured as a star 3 phase.
I only changed the drive belt on the lathe to create different speeds at the PMA.
Data was recorded adding stators in a parallel configuration at the different speed produced by the test fixture. 
 

[GoVertical]

Ghurd makes a good point about paralleling multiple stators - I found that paralleling the AC output did not work that well because I could not get them precisely in time.  So it's easiest to do the paralleling at the DC output.  That's what worked for me.
--
Chris

Hi, I was able to configure the mold with shaft locators and coil center air holes. The coils are vertically aligned with minimal offset. I am trying to avoid the cost of the extra cable that would be needed.  The stators I first made were just round and spun in the opposite direction from the rotor rotation.  The new castings keep the coils lined up. There is a lot counter rotational force experienced by the stators. From the data I collected today it looks like I the unit may require a gear box. Thanks for information. 

[ChrisOlson]

Hi, I was able to configure the mold with shaft locators and coil center air holes. The coils are vertically aligned with minimal offset. I am trying to avoid the cost of the extra cable that would be needed

What I found was, that even using a scope to try to time the phases, every phase was not perfectly identical stator to stator and some performance was lost because of it.  Paralleling on the DC side eliminates that problem.  What I would do is rectify at the generator and transmit the DC power on the long run to the batteries, which only requires two wires.  Transmitting DC is ultimately more efficient than transmitting AC anyway.

--
Chris

[Watt]

Hi, I was able to configure the mold with shaft locators and coil center air holes. The coils are vertically aligned with minimal offset. I am trying to avoid the cost of the extra cable that would be needed

What I found was, that even using a scope to try to time the phases, every phase was not perfectly identical stator to stator and some performance was lost because of it.  Paralleling on the DC side eliminates that problem.  What I would do is rectify at the generator and transmit the DC power on the long run to the batteries, which only requires two wires.  Transmitting DC is ultimately more efficient than transmitting AC anyway.

--
Chris

Chris, just curious why you say this.  No argument, as I wouldn't know for sure anyway, but I was always under the impression DC transmission has a higher loss with heat in the wire than AC of the same voltage. Could you elaborate.  Thanks

[ChrisOlson]

Chris, just curious why you say this.  No argument, as I wouldn't know for sure anyway, but I was always under the impression DC transmission has a higher loss with heat in the wire than AC of the same voltage. Could you elaborate.  Thanks

No, it's the other way around.  AC power switches directions every half cycle and AC impedance is always higher than DC resistance for the same amount of power transmitted.  In basic terms the wiring on the run has stray capacitance; so when you put voltage into it, you are essentially charging up a capacitor - the wire itself.  Only after the wire is charged up will you get power out the other end.  With AC power the charging happens every half cycle because the current is constantly switching directions.  This doesn't matter much if it's a very short wire, but if you make it long enough eventually you get to a point where more current is being used to charge the wire than is coming out the other end.

With DC the wire is only charged once.
--
Chris

[silentblue1987]

Honestly that has to be one of the most impressive WORKING stacked PMA I've seen yet.

I'm just concerned about the power output. A similar sized PMA with hughs design at the same rpm's doubles, triples and sometimes quadruples power output. Then again that mad scientist has been doing this stuff for over 20 years all across the world..

can anyone post output of a hughs based design PMA at the same rpm's for comparison?
It may give you the answer you're looking for as to whether or not many smaller mags are better than single larger mags.