FIGUERA'S AETHER MAGNETIC FIELDS LINEAR PUMP, REVIVED

[kampen]

Subject Re: Upper View and Elevation Cut – CAD Alignment Method

Hello dear friend Ufopolitics,

Yes, that approach makes complete sense, and I follow the same logic.
Doing the upper view and elevation cut together in the same 1:1 scaled file is the correct way to guarantee alignment.

Using the shared centerlines, bolt patterns, shaft axis, and fastener references ensures that the elevation view is not just visually correct, but mechanically true to the upper view.

Once everything is fully constrained and verified, separating the elevation into its own file is straightforward and avoids cumulative alignment errors later on.

This is especially important with stacked assemblies like yours, where shaft, bearings, slip rings, commutators, and plates all depend on the same reference geometry.

The combined view you have shown is very clear, and it confirms that all plates, fasteners, and rotating elements are properly aligned.

Good and solid CAD practice.

Regards, Alex

[Ufopolitics]

Hello to All,

Ok, so, before I get involved physically on starting the build itself, and still working on CAD... I like to 'go back in time' and start reviewing all the previous posts shown here, yes, from Page #1 on...

Plus, I also go over all the old files on my PC's...related to 16 AND 32 Commutator Builds, CAD's, Circuit Connections Diagrams, etc.

And after carefully reviewing it all...I have come to the conclusion that we will be better off with a 32 Segments Commutator, same arrangement, Bipolar as same build measurements.

It is just at the time of cutting segments and making connecting tabs...that if I cut it on a 16 segments I simply can not go back to a 32 segments.

A 32 Segments Commutator can always be reversed/converted to a 16 segments...simply by jumping each Two Segments together.

However, a 32 Segments Commutator offer MANY ADVANTAGES over a 16 Segments Commutator

Now, the first Member here who introduced a 32 Commutator variant, was @Cadman and it was back on Page #9 Post 41 (direct link below)

https://overunitymachines.com/index.php/topic,5.msg133.html#msg133

Cadman CAD Drawing:

Cadman was offering this setup to obtain a Full Sine wave at Output, as he compares on his previous posts there [Page 8]

At that time I was testing the Original Figuera Switching Method (One Positive Brush and other ends of coils to Fixed Negative Input) however, I was using the extended coils over Main Secondary I called Method 2

So, I also conceived at that time - in CAD only- a 32 Segments Cylindrical Commutator with just one Positive Brush and its Positive Slip Ring Input (shown on image below with the Jumpers Installed).:

And the Connections to Coils ( Remember, only one positive brush, so it was only Eight [8] Connectors:

ADVANTAGES OF A 32 SEGMENTS COMMUTATOR (VERSUS A 16 SEGMENTS)

• A 32 Segments Commutator get us (4) Strokes per 360º or every One (1) Cycle. (16 segmants have only Two Strokes per Cycle' [2X180º])
• With Four Strokes we can reach desired 50-60 Hertz with only 1500-1800 RPM´s and Max 2000 RPM´s (Half of a Two Strokes or 3000-3600 RPM´s, Max 4000).
• That is 50% Speed Reduction to small DC Motor, which will work very relaxed and less consumption than running it at 3600 RPM´s, Max 4000.

ON THE BIPOLAR 32 SEGMENTS COMMUTATOR

On above image (Real Scale 1:1) I have kept same Numerical Convention as previous Image:

• That is ( 1 to 8 > 8 to 1) then repeat 1 to 8 and back 8 to 1.
• I named the 1 to 8 Sequences as 'Forward Stage' and the 8 to 1 'Return Stage' , then we have Two (2) Forward and Two Return Stages.
• There are Four Quadrants as I, II, III & IV where All Four Stages develop.

So, now we have Two Forward and Two Return Stages, as Forward 1 and Forward 2, and Return 1 and Return 2. respectively.
Another advantage I see here are the Brush width (6.0 mm) fits exactly the size on a segment only by 0.32 mm more on the segment, this will have better overlap between Groups, as lesser time having the brush contacting each segment (the 16 segments having the Groups ON for longer time ratio, of course requiring higher speeds which equalizes timing.

As an example I can compare these 32 Segments Vs the 16 Segments Commutators with the Four Pole Rotary Generators,and the Two Pole Generators, as the Four Poles require only Half Speed as the Two Poles.

Except, we are not making any Magnetic Field / Number of Coils changes on this Machine Virtual Linear "Rotor"...only changes are done to the Driver.

This is a much easier way to drive the Sequential Coils Groups using much less speed, as much less stress to small Motor.

I honestly feel much more comfortable building this 32 Commutator Driver...as I have studied these advantages since a long time ago.

Just have to adapt it here to the BIPOLAR SWITCHING METHOD.

I apologize for this change, specially to @kampen that had worked so hard on the computation/calculation of the 16 segments.

I also had to work more on everything, double the cuts will require higher precision and accuracy.

Regards

Ufopolitics

[kampen]

Subject: Re: Moving to a 32-Segment Bipolar Commutator

Hello dear friend Ufopolitics,

That conclusion is very sound, and I agree with the logic: if you cut and tab a 16-segment ring first, you locked yourself in whereas starting with 32 segments preserves the option to "down-convert" later by jumpering pairs.
Your main advantage is the one that matters most mechanically: more strokes per revolution, which lets you reach the same effective output frequency at much lower RPM.
Cutting the required RPM roughly in half (e.g., targeting 50–60 Hz around ~1500–2000 RPM instead of 3000–4000 RPM) is a big win for:

• lower brush pressure/heat and less wear
• easier commutation at speed less tendency to bounce
• reduced motor demand and better overall stability.

Cadman's earlier 32-segment idea is a good reference point, and your point about using the same bipolar arrangement/measurements while simply increasing segmentation makes practical sense for your build process.
If you proceed with 32, the two things keep "locked" from the start are:

• the matched indexing convention (Pk with Nk), so the sequence remains deterministic and
• the overlap margin (gap/brush width) so make-before-break behavior stays reliable at the new step rate

Overall: starting with 32 is the safer, more flexible path and it reduces stress on every mechanical part of the system.

Regards, Alex

[Ufopolitics]

Hello ALL,

Thanks @kampen , I am glad you agree with my new changes to a 32 Comm. Setup.

Like I mentioned before, I was not only reviewing CAD Diagrams, Images and Posts comments of my old previous posts here...but also looking at my previous Tests Videos. So please, allow me to show you my reviewed videos here, as my final conclusions later on. (All description show the links to referenced posts)

At that time February 2024, I was pulsing Figuera Style on my Method 2, where the Rotary Switch was moving back-forth the Positive Brush through Eight (8)ontacts.
On the Coils Sequence I had only Seven (7) Coils in Series Connection, plus Two (2) End Coils which were NOT Actively as part of the CENTER sequencing, as the Secondary was underneath the seven coils ONLY.

UNDERSTANDING THE VIRTUAL FIELD MOVEMENT/DISPLACEMENT ACHIEVING A CLEAN OUTPUT SINEWAVE AT OUTPUT.

On Post 160/Page 33  I am showing the video below as I am demonstrating the Fields Movement at Slow speeds:

On that video I am adding first an Input of 28V then later to 50V...

And at reduced speed, you can see the Sequential Coils moving back and forth...And we can observe how just by increasing speed I start to obtain Output gain on Secondary underneath.

At low speeds, we can see the Sinewaves are almost "flat" and EXPANDED over Time...then as I increase speed, Sinewave starts to show gain as it "shrinks" at Time Intervals, BUT GAIN ALTITUDE on both Positive and Negative sides (referring to Yellow AC Sine Output).

On the next analyzed Video (INPUT AT 60V) Post 154/Page 31, I express the issue with the tested Mechanical Switch, that I can not reach the higher speeds I need to run system properly, as I show my Copper Tubing as my intentions to migrate to a 32 Segments Commutator, where brushes will be set from inside-out at minute 1:31

ADDING LOADS TO METHOD 2, AND OBSERVING/ANALYZING RESULTS

This post 156 is located at Page 32:  I am adding here Stressing Loads to setup of Method 2:

And we can observe a Voltage Drop from 20 to 30 Volts and even more whenever the loads are connected...

Now here you can see on video the "STRUGGLE" my Rotary Switch and Motor goes through to just reach not even close to 3000 RPM's (2500)...but also to start losing contacts.

First I get Generator to reach 125VAC Output,at 60VDC Input, NOT LOADED, and as soon as I turn on the small hand tool the VD (Voltage Drop) is 20V falls from 125 to 105VAC.

Now, as I start accelerating hand tool, the VD Voltage Drop is 40VAC, (from 125 to 85VAC)...NOW, observe at min 2:58 how the Generator -by itself- not adding any speed increase goes back to 105 VAC still with hand tool at High RPM's...as when I put stress on tool it keeps around same VD droping at 95-90 VAC.

CONCLUSIONS:

The Method 2, even showing 'promising results', I've found the different 'Flaws' it had:

• Only Seven Coils in Series where the Positive Brush drives a 'SPLIT/DIVISION' between them, generates too High Amperage consumption at Input, even with 23 gauge coils.
• But mainly, the fact that everytime I 'Expand Field' I achieve this by adding more coils to the sequence, therefore, more resistance, and consequently Amperage drops.
• Above point is the main failure of this Method 2, having 'Variable Resistance' at the 'non-desired' Field Volumes (Expansion-Retraction)
• Adding the issue that Rotary Mechanical Switch can not keep up with operating speed of 3600 RPM's, NOT even close to 3000, without starting to lose contacts.

This is the reason why, I designed the New LINEAR-SERIES METHOD, where:

• There will always be Eight [8] Coils in Series ON per each Group.
• Input Resistance will always be CONSTANT, therefore Amperage will also be CONSTANT.
• Calculated Resistance per Group of Sequential Coils Model will be in the range of 55-60 Ohms, forcing Input Amperage to 1.0 Amp at 60VDC, and 2.0 Amps at 100VDC.

And, of course, we will need a ROBUST DRIVER this time to run this awesome Updated System!!...

My Bad: I don't even now, why? I wasted my time making all CAD's with the Old failed 16 Segments Commutator that requires 3600 RPM's to reach the Operating Speeds.
When I can do it with HALF of that speed with a 32 Comm Segments!!

****************************************

Electromagnetic Induction Generation of power/energy has NOT changed since Faraday discovered it on the 1800's...As this Formula basically "bolts down to":

SOURCED INDUCING POWER VERSUS DEMANDED INDUCED POWER TO OUTPUT

Now depending on the "Robustness" of the Sourced Inducing Power it will depend entirely our collected Output Power.

This Robustness consists on a Condensed, Solid and Efective Power Inducing Source structure, mainly braking it in Two Parameters at play:

• Voltage
• Amperage

Therefore:

• IF just one (1) of the two parameters above fails, weakens or drops...this reduces the Inducing Power Source considerably.
• Directly implying also a Drop at our Induced Power Sourced at Output.

Once we Load Output, this generates an INSTANT DEMAND to INCREASE our Inducing Sourced Power.
And IF this SOURCED INDUCING POWER is not a Solid, Robust and Condensed Power, it will struggle to reach the DEMANDED requests.

Regards

Ufopolitics

[kampen]

Subject Re: 32-Segment Upgrade to the Lower RPM and Constant Coil Groups

Hello dear friend Ufopolitics,

Yes, this is another strong point in favor of the 32-segment approach. 
Your sequential coil groups remain in the same resistance range (about 55–60 Ω), so the Current stays predictable ≈1 A at 60 VDC, ≈2 A around 100–120 VDC. 

That stability makes testing and load comparisons much cleaner.

And I do agree with your main conclusion: if the same commutation logic can be achieved at roughly half the RPM, then it is simply a better engineering choice. 
Lower speed means: 
less brush stress, less heating, less mechanical vibration risk, and an easier job for the drive-motor.

Also, do not see the 16-segment CAD work as wasted at all. 
Those drawings helped validate the geometry, the brush concept, and the assembly stack. 

Now you are just applying the same proven structure to a more efficient commutation scheme.

Regarding the driver: 
Yes, it must be robust and designed for inductive switching behavior. 
With your Current and Voltage targets, you want something that can handle transients cleanly and what does not collapse under load changes.

This updated direction looks like the right path to follow 👍 

Regards, Alex