Muller Dynamo for experimentalists

[oscar]

Hi i_ron,
I am sure that due to the precision of your build, both coils of one pair are precisely aligned in one axis. You may want to try and shift/offset them a tiny bit, or alternatively twist the top and bottom stator-plates of your rig a few degrees against each other to offset the alignment.

Yes, if you do sin(x)-sin(x) you get Zero,
same doing sin(x)-sin(x+2pi)
But when you do sin(x)-sin(x+pi) you get good output.
And when the wave form is very narrow/spiky, the additive effect can be very pronounced, even when the geometric offset is quite small.
Yes, right, then both stator coils of a pair are not directly opposing each other any more ....
Don't know whether this is good or bad ...

Could it be - given spiky waves and a tiny mis-alignment of a coil pair's axis - that the electrical wave shapes of the coils add up while the magnetic forces (Lenz) get subtracted from each other (almost canceling) ?  Yes, sounds crazy

[i_ron]

Hi i_ron,
I am sure that due to the precision of your build, both coils of one pair are precisely aligned in one axis. You may want to try and shift/offset them a tiny bit, or alternatively twist top and bottom stator-plates of a full setup a few degrees against each other.

Yes, if you do sin(x)-sin(x) you get Zero,
same doing sin(x)-sin(x+2pi)
But when you do sin(x)-sin(x+pi) you get good output.
And when the wave form is very narrow/spikey, the additive effect can be very pronounced even when the geometric offset is quite small.
Yes, right, then both stator coils of a pair are not directly opposing each other any more ....
Don't know whether this is good or bad ...

Could it be that - given spikey waves - the electrical wave shapes of the coils add up while the magnetic forces (Lenz) get subtracted from each other (almost canceling) ? Sounds crazy

Oscar,

Shifting the coil axis only distorts one half of the sine... still no romero type wave form. It is a puzzle to me how woopy and null are getting an output with reversed phase.

Thanks for the reply.

Ron

[EMdevices]

@null-points,

if your coil pairs are in opposition, and everything is symmetrical, you should get zero volts, as mentioned by Ron.

Are you using any biasing magnets like Romero?

@all

Here's an analysis of RomeroUK's waveform.   In a little while I want to also INTEGRATE this voltage waveform to determine the FLUX, since V = -N d phi/dt

Now take a look at the waveform:

1)  The first thing to note is that there is a slight slope to the whole waveform.   I show an angle theta between the grid and the bottom red line I drew.    This could be due to the fact that he left the ground clip on his probe hanging and is picking up 60 Hz, so his waveform is superposing on this slower frequency.   But if he did connect the ground clip, then this is very interesting.

2) Also notice that the ratio of a/b is changing from pulse #1 to pulse #3, and then back for pulse 4 and 5.   This could be due to different strengths of the magnets and the fine magnetic balance he has achieved.  (a/b = 1.7 at pulse #1  and  a/b=0.7 at pulse #3, approximately)

3) Notice the 2 short yellow vertical lines I drew.  If we cut out this segment and reconnect the waveform, then we should have a waveform closer to what null-points showed, but why are these segments there?  Again, I believe they are there because of the biasing magnets he is using, they create a phase shift and compression of the waveform. i.e.,  the magnet has to approach a lot closer to the ferrite before it snaps it’s magnetic domains in the other direction.

Stand by for more…..

EM

PS,  forgot to mention,  this waveform was most likely taken while the dynamo was under load, because the peaks are flat, indicating the voltage would have risen much higher but it got clamped by the ouput capacitor voltage, as the FWBR conducts.

[Thaelin]

    Would this then mean that his coils were off just a bit and one starts generating and as it is falling off going the other way, then the other sort of helps it along? I would have issue with romero making that kind of mistake. Too good of a builder. Then, I look at mine sitting here and can't say they are. If they are off a minute amount, then it would make that effect "very" speed dependent. Hmmmmm?

thay

[slapper]

if there is a greater amplitude of one polarity over the other polarity across the genset coils
then i can see how the pulses are riding on a DC offset.

with the Muller config (NSNS) you would not see much DC offset.

its like sending a pattern of 01010101.... down a copper pair line.
as long as the line is balanced you won't see a DC offset.
but once more 1's than 0's are sent down the line or more 0's than 1's you'll see the dc offset.
to compensate the industry uses line coding methods like AMI, B8ZS, or Manchester but still can have issues with DC offset.

i suspect that if romerouk had evenly balanced number of the same magnets the pulses coming out of the genset coils would
end up riding on a DC voltage.

some of this can be seen on his scope shot, although the pulses are not riding on a steady DC voltage, as the offset shows a slope.

i think this slope is part of a changing wave that the pulses from the genset coils are riding on.
the reason for the wave is how he arranged his remaining back bias magnets.

couldn't tell how they're arranged on the bottom but on the top it appears there is a section where he started with a thinner magnet,
then up a size then after a bigger size back down to a thinner size. this is where the generated pulses get caught riding a wave.

lots of variables.

take care.

nap