Pierre's 170W in 1600W out Looped Very impressive Build continued & moderated

[listener192]

Attached scope shots showing stepped waveform (blue) synchronized to line sine (yellow).
Zero cross circuit was modified to add a schmitt trigger prior to output, to remove jitter.
Falling edge interrupt was used and code adjusted to ensure correct step start point.
I should mention that the line supply has slightly flat top sine, which is due to supply harmonics from
large numbers of switched mode power supplies that are in use today.


L192

[listener192]

I have been experimenting with different overlapping composite waveforms with little output power. As side note, I found during this exercise, I had my coil 10 and 9 reversed, so I corrected this.

I subsequently went back to my linked coil arrangement that uses 30 coils, 15 in reverse parallel with their 180 degree counterparts.This configuration produces an output about 35% of input (this equates to about 135W limited by my voltage input ratings) and I have been puzzled as to why this waveform was different.

Firstly I found that maximum output was NOT where I had the main group of of contiguous overlapping coils, but instead it was when the rotor was over the coil 30-1 & 15-16 junction. This I call the transition region. Note: there is no series link between coil  15 & coil 1 in this configuration.

The first attached scope shot shows the unloaded output voltage. The dip in the waveform over the positive peak is the transition region from coil 30 to coil 1.
The coil switch pattern follows...
Coils 8 -15 are on and their 180 degree reversed counter part, coils 23 - 30. The coils decrement every step so 8 & 3 are turned off then 9 & 24 and so on down to coils 15 & 30. As each of these coil pairs are turned off, coils 1 to 8 and 16 to 23 are progressively turned on (initially inverted by the H bridges to maintain same polarity), so there are always 8 coils on per pole.
As the sequence progresses, the poles move away from the  transition region into the middle region of the series linked coil group. If the rotor is aligned in this area, there is only a very small output, although of the face of it, the stepped waveform is identical.

So ... what is the difference?
The middle region coils always have one coil turning on and one coil turning off in the pole group. So essentially the current through the pole group is not changing.
The pole group over the transition region comprises of two sets of coils that add up to 8 coils, these also have one coil turning on and one coil turning off however, they are not  series linked so the the number of coils in series is changing i.e 8 to 0 and 0 to 8.

In the decrementing coil group, inductance is reducing and current is increasing.
Note: The recovery current is progressively greater as the number of coils reduce.
In the incrementing coil group, inductance is increasing and current is reducing.

The overall current for the complete 8 coil pole remains constant, just as a contiguous coil pole however, as the pole is divided into two sections, each section experiences large changes in current

Note: No recovery current until 9 coils are in series i.e. No 1 coil switches off.                                                                                                                                                                                                                                                                                                                                                                   This change in current applied to the stepped waveform is key to improved induction.
                                                     
The second scope shot shows the input current from the (DC supply in purple).

Now, I have confirmed this by accident rather than design, as it was the use of only 15 H bridges that created this break in the series coil continuity.

The last scope shot shows the output power, which is only limited by input ratings.

So how to improve this? 
1. I have tested more than 8+8 coils but this is counter productive.
2. A 36 slot stator with 6 slot coil pitch would allow another coil overlap.
3. Reducing coil current in the middle region to stop wasting power as heat.     

L192

[listener192]

Just a small update.
Attached shows the same input current as before but input is now half sine.
As you can see, the middle region where muitiple coils are in series, is now boosted to the same level as the single coils period, which are now reduced due to the switch period being close to zero crossing.
L192

[pmgr]

Looks like Pierre re-surfaced...


He has uploaded two of his old videos to his channel today:


https://www.youtube.com/channel/UCbTYIW_yo65zgWdMlcA3s9A


PmgR

[T-1000]

Looks like Pierre re-surfaced...

It is unfinished project and still not clear method of obtaining magnetic flux change with fraction of the power in a "slow motion" way..  For background information - the "normal" change of magnetic flux is happening too fast for proper induction in other experiments with reactive power driving a change.
Hopefully Pierre could elaborate on that?

Cheers!