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

[listener192]

You don't elaborate, but going by the filenames of the scopeshots you provided, I see an average INPUT POWER of 432 watts and an average OUTPUT POWER (presumably using the 100W bulb as load) of 6.15 watts.

Is that right?




Hopefully, everybody is beginning to understand much better now.

Yes that is correct.
L192

[seaad]

You don't elaborate, but going by the filenames of the scopeshots you provided, I see an average INPUT POWER of 432 watts and an average OUTPUT POWER (presumably using the 100W bulb as load) of 6.15 watts.
Is that right?

Hopefully, everybody is beginning to understand much better now.

https://overunity.com/17609/170-watts-in-1600-watts-out-looped-very-impressive-build-and-video/msg518537/#msg518537

[listener192]

Nicely done. For those not familiar with ratiometric Hall probes we might say that this will give a signal that goes from 0V to 5V, with 2.5 V being the "zero flux" baseline, and lower voltages indicate flux in one direction and higher voltages in the other direction. So just as you say the gap sensor is indicating a sinusoidally varying flux, just as intended.

However.... could you not also achieve this exact same effect with only two stator coils, on opposite ends of the rotor, with two H-bridges and PWM?  You probably couldn't get to 60Hz this way using relays but you certainly could with electronic switching.

And I'll bet it would be a lot more efficient than the full version, too.

Attached are two output scope shots using coils 1-16  so two poles rotating in 30 steps.This is the asymmetric waveform I mentioned.
Notice that at the higher clock rate the current reduces compared to the 6 pole scheme clocked at a much lower rate.
This will be a problem for the higher clock rates required to achieve 50/60Hz. The pulse period is not long enough to achieve maximum current with only 25V. Higher voltages would allow a faster rise time.
In the 6 pole version, each of the 5 series coil  groups is providing 35T x 5 coils x 20A =3500 A/Turns, so x 2 between poles =7000 A/Turns. That is a lot of potential flux, which is obviously not being developed due to overlapped coils of opposite polarity.

Irrespective of a missing element (secret), it is difficult to see how the original relay configuration could have developed such output power, with the flux wastage present in this configuration.

L192

[r2fpl]

Yes, 1 coil produces a much higher output however, it is just flux linkage as per a transformer.
L192

A lot depends on the correct work frequency! therefore, combining the coils will also have the same relationship.

[FixedSys]

However.... could you not also achieve this exact same effect with only two stator coils, on opposite ends of the rotor, with two H-bridges and PWM?

Interesting question.
My understanding (or lack thereof) is that the aim is to move (rotate in this instance) the flux lines to "cut" the armature coils, i.e. move the strongest part of the flux through and across as much of the coil as possible.
A two coil configuration would appear to cut in only two spots. If my understanding is indeed correct, then it implies that resolution of the reproduction of the moving flux is crucial, i.e. the higher the resolution, the more movement can be simulated.

Makes me wonder if Nyquist frequency would apply here? i.e. 50Hz would ideally be achieved with 100 coils.