About the Multi-Filar Coils...
Hello to All,
I would like to point out a couple of things related to the Multi-Filar wound coils concerning our Motors...
Guys, it is not necessary to go that "accurate" on the litzing/twisting etc...please, I would like you to concentrate first than all, in the Magnetic Fields...rather than in the looks...or the Math to it.
What we are looking for is a simple twist on the wires...just in order that they will not spread out during winding...that's all.
Not too twisted, since this would bring other unwanted electric field effects, like Sir John Stone posted above.
The "Ideal" Multi-filar for this would be perfectly Parallel strands of wires...and, since this is not possible (unless we glue them together... ) because they will tend to spread apart at curves when winding and applying pressure...then we just need a simple twist to achieve this.
And the Math was already posted by Kogs...so that simple.
And yes Sampojo...all strands would be attached/jointed at start and end...cleaning all insulation enamel...and soldering/clipping them together...
The point here is to Increase the Magnetic Field Strength by multiple strands projecting in parallel the same magnetic pole Vectors...while current is forced to split into several wires, weakening its "per strand" value during the coil spiral traveling...no matter if at the end...they will join back again.
In other words...We are energizing several Individual Parallel Coils, following same direction of winding...and each Coil would be projecting its Individual Magnetic Field that will Sum Up in the 3D Space.
Over the years...Our main mistake has been to minimize/disregard or even forget about looking at Electrodynamics from the Magnetic Field perspective and projections in 3D Space...while concentrating in other parameters, like Math...Physical Shape of Coils, Wires etc...and that, is a big time mistake my Dear Friends.
Warm Regards
Ufopolitics
Hello to All,
I would like to point out a couple of things related to the Multi-Filar wound coils concerning our Motors...
Guys, it is not necessary to go that "accurate" on the litzing/twisting etc...please, I would like you to concentrate first than all, in the Magnetic Fields...rather than in the looks...or the Math to it.
What we are looking for is a simple twist on the wires...just in order that they will not spread out during winding...that's all.
Not too twisted, since this would bring other unwanted electric field effects, like Sir John Stone posted above.
The "Ideal" Multi-filar for this would be perfectly Parallel strands of wires...and, since this is not possible (unless we glue them together... ) because they will tend to spread apart at curves when winding and applying pressure...then we just need a simple twist to achieve this.
And the Math was already posted by Kogs...so that simple.
And yes Sampojo...all strands would be attached/jointed at start and end...cleaning all insulation enamel...and soldering/clipping them together...
The point here is to Increase the Magnetic Field Strength by multiple strands projecting in parallel the same magnetic pole Vectors...while current is forced to split into several wires, weakening its "per strand" value during the coil spiral traveling...no matter if at the end...they will join back again.
In other words...We are energizing several Individual Parallel Coils, following same direction of winding...and each Coil would be projecting its Individual Magnetic Field that will Sum Up in the 3D Space.
Over the years...Our main mistake has been to minimize/disregard or even forget about looking at Electrodynamics from the Magnetic Field perspective and projections in 3D Space...while concentrating in other parameters, like Math...Physical Shape of Coils, Wires etc...and that, is a big time mistake my Dear Friends.
Warm Regards
Ufopolitics
Comment