Reply to Message # 769

Dear friend Ufopolitics,

I reviewed your new bipolar rotary switch design, and I believe it is a significant improvement over the original version. 

By eliminating the two continuous shaft-mounted slip rings and reducing the number of rotating electrical contacts, you have simplified the assembly, reduced drag, lowered the parts count, and made the project much more practical for builders without professional machining equipment.

The main advantages I see are improved manufacturability, reduced alignment requirements, a more compact structure, and fewer potential failure points. 

However, I would still pay close attention to insulation between the positive and negative slip rings and the steel shaft, as well as brush arcing, contact resistance, heat generation, and rotational runout. 

Since this system will be switching inductive coil loads, good arc-suppression measures and careful brush design will be important for long-term reliability.
Overall, I think the redesign is mechanically superior to the first concept and represents a meaningful step toward a more buildable and robust prototype.

Best regards, Alex

QuoteVERY IMPORTANT NOTE HERE:

• I have shown both Slip-rings as WIDER than the Brass Mounting on opposite side, just to differentiate that are Two separate parts.
• BUT, that is NOT the way they go!!...They are about the same size, except by a slight difference on Inner Shaft Area is wider than shaft for Slip-Rings.
• In order that Slip-Rings NEVER make contact with COMMON Shaft (Or it will definitively SHORT OUT!)

The way these TWO Parts are set (Brass Shaft Piece that bolts to Fiberglass Mounting Plate (Light Blue here) PLUS the Slip-Ring are based on a Three Bolt Design each, separated by 120 Degrees.

Where they alternate (are rotated by 60 degrees to mount them) in order NOT to coincide on their mounting holes-bolts.

Hello All,

Ok, here are some 2D CAD drawings, which (I hope) will give you a better idea and understanding of how these Two Parts are designed, built and set:



On above image is shown Part #1, which job is to secure Shaft to Fiberglass Disc-Plate, which also holds the Commutator Brush.
As you can see there is a Securing Bolt (Black) which press against shaft Flank-flat side (Upper View is better seen)
So, at the Brushes LEVELS, we must make a flat side on Shaft (by filing it) in order that Securing Bolt LOCKS into it.
Please NOTE the Inner Diameter (ID) of this Brass Piece #1 is EXACTLY the size of Shaft Outer Diameter (OD)...well, "not exactly the same" but a few mills wider so that shaft can slide , but TIGHT!



And here is the SHAFT MOUNTED Continuous Slip Ring, or Part #2.
Please NOTE the INNER DIAMETER of this piece compared with SHAFT Outer Diameter!! NOTE THE GAP!! (Better viewed on UPPER VIEW/TOP IMAGE)
So, basically these Two Pieces are "around" the same. The Mounting Bolts Holes are EXACTLY SET AT 120ª FOR BOTH.
On this Graphic I made a bit wider the Outer Diameter Area where slip-ring Brush will be riding...compared with Piece #1.


Finally, above image shows the way Both Pieces are attached together...
Here I included the Fiberglass Disc-Plate (Light Blue) which is sandwiched between Both Pieces, on Both Views (Upper and Elevation [side])
Also, on this image I have set the Slip-Ring Piece#2 ABOVE, while the Piece#1 is BELOW. (Like it mounts on Positive Commutator [Top] on my previous Post Image)
Attaching Bolts-Nuts to the Fiberglass Disc-Plate are NOT SHOWN here.

So, like I wrote before on my previous post (quoted on the beginning of this post) :

ALL WE NEED TO DO, (WHEN MOUNTING THESE TWO PIECES) IS ROTATE ONE PIECE 60º RELATED TO THE OTHER ONE.
Then we get the result as shown on UPPER VIEW here.
Where ALL SIX Mounting Bolts-Nuts are separated by 60º.

FINALLY, we can see on this Diagram, that on this Design, the Slip'Ring Piece #2 would NEVER be in contact with Shaft, NOR with Piece·#1.

EDIT 1: To make these Two Pieces on a Lathe they can be CUT OUT TOGETHER (as ONE PIECE) to then cut in half and 'vualá' we got the two pieces.
When drilling center SHAFT HOLE (On Lathe) we drill BOTH Pieces at once, with the Drill Bit of Shaft size.
And once SEPARATED, we just run the Slip Ring Piece with a bigger drill bit...
Then we make all the 120º TABS Cuts (on each separate piece) and drill mounting bolts holes...(BUT Before we need to run -on Lathe- a light circle guiding MARK, on both WIDER ends, where the diameter is the same for both pieces!!)
This is VERY IMPORTANT!...because once we drill the ID on Slip Ring bigger, then we no longer have the Shaft as a Guide to MOUNT it on Fiberglass Disc-Plate and set the three bolts Holes!!
On the Fiberglass Disc-Plate, we also need to run-mark, this guiding circle (SAME Diameter) where ALL Bolts would be lined up.

Regards to All

Ufopolitics

Reply to Message # 769

Dear friend Ufopolitics,

Here is my engineering assessment.

The new design is mechanically better than the first one, mainly because it reduces the rotating contact system from four slip/brush interfaces down to two fixed input brushes feeding two rotating slip rings. 

That lowers drag, parts count, height, alignment burden, and failure probability. 
Ufopolitics is right about that part.

However, the project is still mechanically and electrically demanding. 
The weak points are not fully solved; they are moved into a more compact assembly.

Main verdict
This commutator is buildable as a prototype, but I would not expect it to be reliable at high speed or high current unless it is machined very accurately and operated conservatively.
The biggest risks are:

• Slip-ring shorting to the steel shaft
   The positive and negative slip rings must be perfectly insulated from the common metal shaft. Any metal dust, carbon dust, loose washer, burr, deformation, or heat damage could create a short.
• Brush arcing
   If this is switching inductive coils, arcing will be significant. 
• Mechanical commutators hate inductive loads unless you add suppression: flyback paths, snubbers, MOVs/TVS, capacitors, or controlled timing gaps.
• Contact resistance and heating
   Carbon/copper brush contact is not ideal for large pulsed currents. 
• The heat-sink/fan-fin idea helps cooling, but it does not solve poor contact pressure, bounce, arcing, or pitting.
• Rotational balance
   The slip rings, brass hubs, bolts, fins, wires, and brush arms must be balanced. 
• At speed, even a small asymmetry becomes vibration, brush bounce, bearing load, and arcing.
• Axial/radial runout

The slip rings must run true. If they wobble, brush pressure varies cyclically, causing noise, heating, sparking, and contact dropout.[/list]
Good improvements in the new design
The redesign has real advantages:

• Fewer brush contacts.
• Lower drag on the drive motor.
• Shorter vertical assembly.
• Less copper tubing and fewer stacked plates.
• Easier assembly than the original "six-story" design.
• Fixed external input brushes are much easier to adjust than shaft-mounted supply brushes.
• Separating supply slip rings from the switching commutator is a cleaner architecture.

So conceptually, yes: the new design is superior to the first version.
But I would change several things
1. Do not rely on fiberglass alone near arcing zones
The fiberglass plate is okay structurally, but carbon dust, voltage plus humidity can create leakage paths. 
Use insulating barriers, grooves, and creepage distance between positive, negative, and shaft metal.
I would add:

• Phenolic, G10/FR4, or glass-filled epoxy insulation.
• Recessed insulating washers.
• Shoulder bushings on every bolt.
• No exposed conductive fasteners crossing polarity zones.
• Clearance grooves around slip-ring mounts.

2. Use isolated hubs, not just "rings near a shaft."
The positive and negative slip rings should be mounted on proper insulated hubs/bushings. 
The steel shaft should never be close enough that a loose chip or washer can bridge it.
Best practice: steel shaft → insulating sleeve → brass/copper carrier → slip ring.
3. Avoid internal lathe cutting of large copper tubes if possible
Ufopolitics is correct: boring the inside of a large copper cylinder on a mini-lathe is risky.
Better options:

• Use commercially available copper/brass rings.
• Waterjet/CNC-cut ring plates.
• Use stacked copper segments bolted to an insulating drum.
• Use a PCB-style commutator disc for low/moderate current.
• Use purchased slip rings for the continuous supply function.

4. Add arc suppression immediately
If this switches coils, the commutator will spark badly without protection.
Minimum:

• RC snubber across each coil or brush segment.
• Fast diode/TVS path if DC polarity allows it.
• Adequate dead zone between segments.
• Wider air gaps between adjacent live segments.
• Test first at low voltage/current.

5. Brush material matters
For the slip rings, silver graphite or copper graphite brushes are better than random carbon blocks. 
For high current, use multiple smaller brushes in parallel rather than one hard brush with high spring force.
Too much spring pressure increases drag and wear. Too little causes bounce and arcing.
6. Do not underestimate timing accuracy
The commutator segments must energize the coils at the correct angular position. 
If timing is off, the device will waste power, heat coils, arc more, and possibly oppose rotation.
You need:

• Adjustable brush holders.
• Angular timing marks.
• Ability to rotate/advance brush position.
• Oscilloscope measurement of coil voltage/current pulses.

A mechanical commutator is useful if the project specifically requires mechanical switching, but as an engineering choice, electronics will be more reliable.


Bottom line:
Ufopolitics' redesign is a meaningful improvement over the original. 
It is simpler, shorter, lower-drag, and more buildable.

But the project still has serious risks: insulation failure, brush arcing, heat, runout, balance, and timing accuracy. 
I would treat it as a delicate experimental commutator, not a robust machine component, unless precision machining and proper arc suppression are used.

Kind regards, Alex

Hello, dear friend Ufopolitics,

After carefully studying your 32-segment bipolar rotary switch redesign, I created a preliminary 3D isometric engineering rendering, Vers. 1. 
Based on the drawings and descriptions you posted in message # 769.


The purpose is to visualize the complete assembly, including the fixed positive and negative slip rings, fiberglass rotor discs, 32-segment bipolar commutator, brush assemblies, bearings, shaft, and optional cooling fins.

The rendering confirms that your NEW architecture is considerably simpler and more compact than the original design.
Reducing the number of rotating electrical contacts while moving the supply brushes to fixed positions appears to be a very practical improvement from both a manufacturing and maintenance standpoint.

I want to take the next step and convert this conceptual rendering into a true parametric CAD model that can eventually be used for detailed design, machining drawings, assembly verification, interference checking, balancing studies, and possibly CNC fabrication.

To do that accurately, I would need the following dimensions and design parameters as requested below.
Please give me all dimensions in (mm). Later, can this be converted to (inches)

• Steel shaft diameter.
• Steel shaft overall length.
• Fiberglass rotor disc diameter and thickness.
• Brass hub/sleeve outer diameter, inner diameter, and length.
• Positive slip ring:
  
  
  • Outer diameter
  • Inner diameter
  • Ring width
  • Thickness
• Negative slip ring:
  
  
  • Outer diameter
  • Inner diameter
  • Ring width
  • Thickness
• Distance between the positive and negative slip rings.
• Commutator outer diameter.
• Commutator inner diameter.
• Commutator axial width.
• Number of segments (currently shown as 32).
• Segment insulation gap width.
• Brush dimensions:

• Width
• Thickness
• Contact length

• Brush spring travel and desired contact pressure.
• Bearing type and dimensions.
• Distance between upper and lower end caps.
• End cap diameter and thickness.
• Three-bolt mounting pattern dimensions (bolt size and bolt circle diameter).
• Cooling fan/heatsink fin dimensions and quantity.
• Maximum intended operating RPM.
• Maximum expected voltage and current.

If exact dimensions are not available yet, even approximate values would allow me to build a first fully parametric CAD assembly. 
Once the base model is completed, all dimensions can be adjusted later as the design evolves.

I believe a proper CAD model will help identify mechanical clearances, insulation requirements, balance issues, brush geometry, and assembly challenges long before fabrication begins.

Looking forward to your thoughts and dimensions.

Best regards, Alex 

Hello dear friend Alex,

Thans so much for all your detailed analysis of my new design!

Also, thanks very much for the late rendering you have uploaded!

You are completely right, only a FULL 3D rendering (or an Isometric or Perspective views) of all these components, from different angles, could display exactly the way this Commutating Driver works.

Now, on your excellent rendering, there is one thing that is NOT properly represented:

• The Two Commutators you show are the ones that we see typically on any Brush Motor.
• Meaning, they rotate with Motor Shaft.
• As their Brushes are stationary, mounted on the Motor Housing.
• Brushes are in charge (on a brush motor) to energize the Armature Coils.

My System is completely 'Reversed Engineered' to the Typical Fashion that any Brushed Motor Operates:

• My Commutating Brushes are attached to the Inner Rotor Assembly.
• Therefore, these brushes sweep 'from inside out'.
• Commutator design is completely reversed from its original and typical concept.
• Where the contact elements are on the ID (Interior Diameter)
• It is the only way, we could run all 32 wires (from the 32 Elements) -of each commutator- to an Exterior Connection.

The Typical Commutator design (like you have shown) is limited and constrained to supply the commutation excitement (power), to only its Inner Rotating Armature Coils.
And we need here the complete opposite operation, and that is to take out of each commutator all 32 wires from each (Positive and Negative) to an EXTERIOR Location.
Therefore, our commutators can NEVER ROTATE, nor be part of the Rotor Assembly, BUT be on the STATOR SIDE.

I am going to make all these components into a 3D CAD...And post it here when I have them ready.

Best regards!

Ufopolitics