Agentgates´s TPU setup with strange wavehill hump

[agentgates]

@Magluvin

Yes, when I was experimenting with that device I didn't observe this carefully, it was only a clue that I though logical. I think I also mentioned that.

It looks like I can't finish with the winding on the nice acrylic today due to the new glue bonds slower than I expected and I can't glue every sides in the same time cause I have to be careful to keep the same spacing on every sides. (Don't get confused on this. It is because this device will be in the presentation and I dont want to make an ugly job)

When I am ready with it I will do some more research on the mentioned polarity and upside down things to make their results an absolute fact. Until then I am improving the software side to make myself useful.

[agentgates]

some updates on previous page to your suggestion guys. Thanks.
Maybe I'll do some birdwatching first 

Ok, that is the best thing when you get stuck. It always helped me.

[broli]

Tony while working on the software I was asking myself. If you use a pulse of 1ms on and 1ms off then you cannot have overlapping phases. The best you can get is having to start the next phase when the previous phase hits low, this would be a delay of half a period. Arduino does support microsecond (accuracy of 4µs) delays but is that high frequency really needed?

[Magluvin]

Ok Gates. We have to accept that.

tes   very good then.

But man if there is a formula to this and it works, I should be able to severely increase the range on my ebike. I get 20 mi per charge now, but if this can make 48v at 20a for very little input, Yippy!  lol  I have 3 ebikes one is a Wavecrest tidalforce, another is a trek with an ebay kit, spins the tire, and its a brushless hub motor. The 3rd is a 160 Target bike with a brushed hub motor, it is inefficient and low torque, but 27 mph top speed all motor. lol
Just dreaming.
Im still rooting for ya Gates. Even though my faith is lackluster, but we cannot say I am the only one.

Seems so unlikely that so little wire could do so much. But these are the days of our lives.

Carry on Gates.
Magluvin 0 7

[altair]

Thanks agentgates,
it's much clearer now.
You mention driving the coils with 50% duty cycle, so with 3-phases that will have the effect of sometimes having two phases simultaneously ON, because of the overlapping. See attached drawing.  Is this what you planned ?
If instead we were to pulse with 33% duty cycle, there would always be only one phase ON at a time...

By the way, why don't you use cyanoacrylate glue for your coils, it's instant ! 

Altair