FIGUERA'S AETHER MAGNETIC FIELDS LINEAR PUMP, REVIVED

[Ufopolitics]

Hello dear friend Alex,

Now you have this concept pretty well understood, which is great!

This Concept was based on the Original Figuera's Design, except it was with only one positive brush, however, according to the Diagram plus Description shown on the 1908 Patent.
It is very obvious that his Positive Brush sweeps from INSIDE-OUT, or the center of inner rotor towards the outer annular ring (where he set all the Sixteen (16) Commutator bars).

I just 'improved' Figuera's design...

Now, some Comments and Opinions related to the New Design that We are posting here...

[...]
One aspect that particularly caught my attention is the elegant way the rotating brush carrier is integrated into the rotor assembly.

I made this design to be built AS SIMPLE AS POSSIBLE...So, that Brush Housing is simplified in order that it could be reproduced from a thin Brass or Copper Plate, which could be bent and shaped to hold the brush and have -at the same time- the two side base, then drilling the Four Holes to be mounted on the Fiberglass Disc on Rotor.
This minimizes the Rotor weight considerably...and that is exactly what our goal is here.

Now, your rendered Diagrams are awesome!!, very realistic and very well presented, very professional!!

You have shown very well detailed info, detailed explanations about every single detail plus Exploded Views to have a Full idea of how the whole Assembly Decompose...

Your Brushes Closeups (on Both, Rotated and Fixed) are very well detailed and show excellent and vivid rendering resolution!

I would try to show NOT TOO Crowded Designs with unnecessary"holes' plus loaded with too many bolts and nuts, and springs...

However, I see some small issues on your rendered images...

1- Your Commutators Inner Diameter seems too SMALL, related to Your very Robust and Voluptuous  Brushes Assembly.

2- On ALL Conventional Commutators, their Copper Elements are mounted (Fused) on a Mica Ring Insulator.
On mi design, I have just used the Polycarbonate Rings...BUT many other Materials could be used, like POM or Bakelite for a more lasting and Industrial  Model.
Your Copper Comm Elements (bars) seem to be bolted on individually and only on one end on a plate.

3- On the Exploded view (as in others) I see like Two Bearings together on the same side, so, 4 in total.

[color=var(--body-txt-color)]My idea and goal here is to have AS MANY PEOPLE REPLICATING this Design as possible.[/color]
In order that with MINIMAL Engineering Skills, ANYONE could put it together on their garage or small shop...

However, and unfortunately no one BUT YOU, have been putting so much effort and work into this project, and I really appreciate that!!

The idea is like let's say a John Doe, who does not have any Engineering Schooling, much less any CAD Design Schooling on how to Interpret and Understand a REAL ENGINEERING BLUE PRINT.
Then, when John Doe looks at your EXCELLENT RENDERED Drawings...He will get literally AMAZED...BUT SCARED!!...those images would be VERY DIFFICULT for him to think he could REPRODUCE them, because He will be looking at ALL this COMPLEX Drawings...and believes He needs to be a Professional in order to be able to build that!!!

So, my recommendations for your future rendering Drawings is:

Please TRY TO KEEP these drawings AS SIMPLE AND MINIMALIST AS POSSIBLE !!
Try NOT to load too many unnecesary components, as much lesser detail that does not add any Mounting or any function means.

Just the Basic Structural Items...cleaner, minimalist build.

To be understood easier by ANYONE viewing your VERY BEAUTIFUL RENDERINGS!!

Regards my dear friend

Ufopolitics

[Ufopolitics]

Hello All,

Ok, so I mounted BOTH BOARDS on each Commutator Assembly...:

POSITIVE COMMUTATOR UP

NEGATIVE COMMUTATOR UP

And below, a short Video where I am showing this 3D Model as seen on images above, plus explaining all details...

Regards


Ufopolitics

PD: EDIT 1

On this Setup, when I am running it Clock Wise (CW) the Quadrants are exactly like I described on the video.
HOWEVER, when am running it Counter Clock Wise (CCW), ALL QUADRANTS REVERSE.
SO, the Quadrant which was Positive Rotational Sequence  [1,2,3,4,5,6,7.8] at CW...
On CCW that same Quadrant becomes now Negative Rotational Sequence [8,7,6,5,4,3,2,1].
BUT as I said on video: Changing the Direction of Rotation will NOT AFFECT the Sequential order.
All that changes is the Sequential order on each Quadrant, but the FINAL, OUTPUT RESULT is exactly the same.
We will still have Two Quadrants Positive as Two Quadrants Negative ALTERNATED, following on either Rotational Direction (CW or CCW)
Regards

[kampen]

Reply to Message # 781 and #783

Dear friend Ufopolitics, 

This is actually one of the most practical ideas You have proposed so far for the Bipolar Commutator Driver.
After reviewing the three CAD renderings, I see a significant evolution from:

Code Select Expand

32 Individual Segment Outputs

Code Select Expand

          ↓

Code Select Expand

32 External Wires

Code Select Expand

          ↓

Code Select Expand

Manual Interconnections
to:

Code Select Expand

32 Segment Outputs

Code Select Expand

          ↓

Code Select Expand

Annular PCB Interface

Code Select Expand

          ↓

Code Select Expand

Internal Sequencing Network

Code Select Expand

          ↓

Code Select Expand

8 Final Outputs
What the New PCB Does
The annular PCB acts as an integrated "wiring harness."
Instead of:

• 32 separate wires
• 32 terminals
• 32 opportunities for wiring errors

The PCB performs the sequencing internally.
The result is:

• Cleaner assembly
• Reduced wiring complexity
• Easier maintenance
• Better repeatability

What I Like
1. Direct Mounting to Commutator Bolts
The board is bolted directly to the commutator terminals.
Advantages:
✅ Very short electrical path
✅ Lower contact resistance
✅ Less wiring clutter
✅ Mechanically robust
The CAD appears to use each segment bolt as both:

• Mechanical support
• Electrical connection point

Which is a smart solution.

2. Eight Final Outputs
This matches Ufopolitics' grouping concept.
Instead of routing:

Code Select Expand

1

Code Select Expand

2

Code Select Expand

3

Code Select Expand

...

Code Select Expand

32
individually,
The board internally combines the segments into:

Code Select Expand

Output 1

Code Select Expand

Output 2

Code Select Expand

Output 3

Code Select Expand

Output 4

Output

Code Select Expand

5

Code Select Expand

Output 6

Output

Code Select Expand

7

Code Select Expand

Output 8
This greatly simplifies connection to:

• Sequential coils
• SSD systems
• Future driver electronics

3. Two-Layer PCB Architecture
The transparent rendering clearly shows why a single-layer board would become difficult.
The traces must cross.
Using:

Code Select Expand

Top Layer

Code Select Expand

+

Code Select Expand

Bottom Layer

Code Select Expand

+

Code Select Expand

Vias
is exactly how a professional PCB would be designed.
This avoids:

• Jumpers
• Flying wires
• Solder bridges

One Engineering Concern
The traces near the annular ring appear fairly narrow.
If the system is expected to carry significant current, I would recommend:
Heavy Copper
Prefer:

Code Select Expand

2 oz copper

or even: 

Code Select Expand

4 oz copper
Instead of standard 1 oz PCB copper.

Wider Traces
Particularly near:

• Segment outputs
• Grouped output buses

The current may be concentrated into only a few final outputs.

What I Would Add
Numbering Layer
I would permanently label:

Code Select Expand

1-32

around the ring.

And:

Code Select Expand

OUT1

Code Select Expand

OUT2

Code Select Expand

OUT3

Code Select Expand

...

Code Select Expand

OUT8
at the connector side.
This will greatly help assembly and troubleshooting.

Mounting Isolation
One thing that needs verification:
Are the commutator bolts electrically isolated from adjacent segments?
The PCB assumes:

Code Select Expand

Segment 1

Code Select Expand

Segment 2

Code Select Expand

Segment 3

Code Select Expand

...
Remain completely isolated.
If metal washers or support hardware bridge segments, the sequencing could be compromised.

Manufacturing Assessment
Compared with the original wiring concept:

Feature	Original	PCB Version
Wires	32	8
Assembly time	High	Low
Wiring errors	High risk	Low risk
Troubleshooting	Difficult	Easier
Appearance	Complex	Professional
Scalability	Moderate	Excellent

My Recommendation
I would strongly encourage you to continue with this PCB approach.
In fact, if the final dimensions become available, the next logical step would be:
Rev 1 Engineering PCB
Including:

• 32 segment inputs
• 8 grouped outputs
• Two copper layers
• Through-hole vias
• Silkscreen numbering
• Mounting holes matching the commutator bolt circle

At that point, the Bipolar Commutator Driver starts transitioning from a hand-wired prototype into something much closer to a reproducible engineering assembly.
This PCB concept may end up being one of the most important improvements to the entire project because it removes a large portion of the wiring complexity while preserving the commutation logic exactly as Ufopolitics intended.
 
Reply to Message # 783
Ufopolitics is correct: this routing cannot be cleanly implemented on a single-layer PCB because the positive and negative quadrant wiring creates many unavoidable crossings.
My advice:
The annular PCB idea is good, but it should be treated as a proper 2-layer or preferably 4-layer current-distribution board, not just a decorative wiring simplification.
For a practical build:
Use at least 2 copper layers

• Top layer: one polarity routing or one set of quadrant paths.
• Bottom layer: crossed return/grouping routes.
• Use vias or plated through-holes at controlled points only.

Consider 4 layers if the current is high

• Layer 1: positive routing.
• Layer 2: positive bus reinforcement.
• Layer 3: negative bus reinforcement.
• Layer 4: negative routing.
   This gives lower resistance, less heating, and cleaner separation.

Do not use thin signal-PCB traces
 If coils draw meaningful current, use:

• 2 oz or 4 oz copper.
• Wide traces or copper pours.
• Large annular pads at all 32 commutator bolts.
• Multiple vias in parallel where layers connect.

Must add clear labeling
 The board should have silk-screen labels:

• Segment 1–32.
• Quadrants I–IV.
• Outputs 1–8.
• Positive/negative board orientation marks.
• Brush rotation direction.

Use two separate boards if needed
 One PCB for the positive commutator and one for the negative commutator may be cleaner than trying to combine everything into one dense board.
Watch insulation and creepage
 Because this is switching inductive coils, arcing and voltage spikes are possible. 
Keep good spacing between unrelated traces and avoid routing high-voltage pulse paths too close together.

REMARK: 
Ufopolitics' PCB concept is a major improvement over loose wiring. 
It reduces 32 connections down to 8 outputs and makes the driver far more buildable. 
But it must be designed like a real power PCB, with heavy copper, wide traces, proper vias, good spacing, and clear labeling.

Best regards, Alex

[kampen]

Reply to Message # 785  >> I will create a much simpler "builder-friendly" rendering style.<<

Hello, dear friend Ufopolitics,

Thank you very much for your detailed feedback.
I completely understand your point, and I agree with you.

The previous renderings were made more as engineering explanation diagrams, so they included many details, bolts, springs, exploded views, labels, and close-ups to help me understand the complete operating principle.

However, I now see that for builders and replicators, this can make the project look more complicated than it really needs to be.
Next step, going forward, I will create a much simpler "builder-friendly" rendering style.

The next version should show only the essential structural and functional parts:
• The central shaft.
 • Upper and lower support plates.
 • One upper bearing and one lower bearing.
 • Positive and negative slip rings.
 • Simple thin bent-plate brush holders.
 • Fiberglass/polycarbonate rotor discs.
 • Stationary commutator rings.
 • Copper commutator bars embedded into an insulating ring.
 • Basic fixed input brushes.
 • The 180° brush relationship.
 • The 32-to-8 output board or wiring concept.


I will also correct the proportions so that the commutator inner diameter is larger and the rotating brush holders are much smaller and lighter, more like the simple bent brass/copper plate construction you described.
I also understand your point about the commutator bars.
They should not appear as individually bolted pieces on one side only.

They should look more like conventional commutator bars set into an insulating support ring, except reversed for your inside-out brush operation.
And I will correct the bearing layout so the assembly does not look like it has unnecessary double bearings on each end.

So from now on, I think we should have two different rendering types:

• A simple builder-friendly version, meant to show that the project can be reproduced by ordinary builders with basic tools.
• A technical engineering version, used only when we need to explain current paths, exploded assemblies, quadrant sequencing, and internal operation.

Your main goal is clear: make the design understandable and reproducible, not intimidating.
Thank you again for the correction. I will simplify the next renderings accordingly.

Best regards, Alex

[kampen]

Reply to Message #786 Today at 03:33 AM from Ufopolitics:

This is actually a very significant development in the project.

After studying both new renderings and Ufopolitics' explanation, my engineering assessment is that the annular PCB concept solves one of the biggest practical problems of the original design:
Before
The original system required:

• 32 commutator segments
• 32 external wires
• Complex cross-connections
• Large wiring harnesses
• Significant assembly labor
• High risk of wiring errors

As shown in the original 32-segment diagram, there are many crossing connections that cannot realistically be implemented on a single-layer board.
Now
The new annular PCB effectively becomes a:
32-to-8 Sequential Routing Matrix
The board performs all internal interconnections.
Instead of:
32 outputs per commutator
You obtain:
8 final outputs per commutator
This dramatically simplifies construction.
What I Like Most
1. The commutator becomes modular
The builder can assemble:

• Commutator
• PCB ring
• Output connector strip

as one module.
A future builder could simply bolt the PCB directly onto the commutator ring.
2. Manufacturing becomes easier
Instead of manually wiring:
32 copper bars
You simply design:

• Top copper layer
• Bottom copper layer
• Through-hole vias

And send the Gerber file to a PCB manufacturer.
3. Maintenance improves
If a wiring fault develops:

• Replace PCB module

Instead of tracing dozens of wires.
4. Easier replication
This is exactly what Ufopolitics is trying to achieve.
A builder no longer needs to understand the entire quadrant sequencing theory.
They simply:

• Install commutator
• Install PCB
• Connect terminals 1-8

Done.
My Suggested Improvements
Version 2 PCB
Instead of individual screw terminals around the ring:
Use:

• copper pads
• soldered brass pins
• threaded inserts

Only at the final 8 outputs.
This reduces clutter considerably.
Must add quadrant markings
I would add:
Q1
 Q2
 Q3
 Q4
directly on the PCB silkscreen. This helps builders align brushes correctly.
Add brush reference marks
I would print:
POS BRUSH
and
NEG BRUSH
Locations on the board. This prevents installation mistakes.
Concerning the CW / CCW Note
Ufopolitics' explanation is correct.
When rotation reverses:
CW:
Q1 = 1→8
becomes
CCW:
Q1 = 8→1
However:

• two quadrants remain ascending
• two quadrants remain descending
• positive and negative remain alternated

Therefore, the sequential transfer principle is preserved.
The sequence orientation changes, but the operating logic remains symmetrical.
Impact on Future Renderings
Because of this PCB development, I would now modify the rendering roadmap:
Builder-Friendly Version
Show:

• Stationary commutator
• PCB ring mounted directly above it
• Only 8 output terminals visible

And hide all internal routing. Keep it simple.
 
Engineering Version
Show:

• Transparent PCB
• Top layer traces
• Bottom layer traces
• Via connections
• Current paths
• Quadrant labels

This is where the transparency views Ufopolitics posted become extremely useful.

Overall Assessment
I would rate this PCB-ring development as one of the most important practical improvements introduced so far because it transforms the concept from a difficult hand-wired 32-segment machine into something that could realistically be reproduced using standard PCB manufacturing processes.

It moves the project significantly closer to a repeatable and builder-friendly design while preserving the original sequential commutation logic.

Best regards. Alex