Muller Dynamo for experimentalists

[redrichie]

Here is the start to my dynamo.  My coils are hand Litzed by me. 5 strands, almost .9mm total.   I could not fit even 100 wraps on the bobbins I have.  The cores are radio shack choke cores.  with the cores in each coil measures about 150 micro Henry's (yes micro not mili)  in parallel they are about 63uH.  if I leave in parallel and short one of the coils, of a pair, to itself the inductance drops to about 5uH.  So virtually nothing.   series there is virtually no change in Inductance.
  My question is on resistance.  This is a new meter that I am using.  When I try to measure Resistance @ 200 Ohms scale the meter pegs out.  But on 2k scale they all read 1.39 k Ohms.  so is this k Ohms or Ohms? If it is K Ohms how high of a voltage ill I have to use to even get this spinning?  Should I start re-wrapping now? LOL

Also does having such a low inductance hurt?  will this low of inductance generate any electricity whatsoever?

And yes there are only 6 sets of coils.  This is a testbed for the bigger motor in the works.  I had a feeling there was gonna be a big learning curve and no straight replications when I began this so I went a little smaller and dif configuration for learning.  but same principle

[gyulasun]

....
  My question is on resistance.  This is a new meter that I am using.  When I try to measure Resistance @ 200 Ohms scale the meter pegs out.  But on 2k scale they all read 1.39 k Ohms.  so is this k Ohms or Ohms? If it is K Ohms how high of a voltage ill I have to use to even get this spinning?  Should I start re-wrapping now? LOL

Also does having such a low inductance hurt?  will this low of inductance generate any electricity whatsoever?

.....

Hi Redrichie,

I assume your new meter has a separate resistance and inductance measurement input points?  I ask this for I am not sure what exactly you mean on Resistance here: the DC resistance of the coil I suppose? 
Or your new meter is able to measure inductive reactance at a built-in test frequency?

Try to check you meter first in the 200 Ohm resistance scale with a 100-180 Ohm normal resistor to see if it is ok.  If yes, then your coil(s) probably force your meter into spurious oscillations, I have come across such phenomena once or twice with some meters, very rare but rather confusing.
Also try to check your meter in the 2 kOhm scale too by any normal resistor of 1.5 - 1.8 kOhm. If it shows correctly, then again your coil makes your meter oscillate in the Resistance range and display shows an erratic value.

You can surely estimate your coil DC resistance by its wire gauge diameter and wire length, considering the paralleled wires in your Litz winding of course. I guess the paralleled 5 strands in the Litz has got well under 1 Ohm DC resistance (based on your 0.9mm total dia.)

If you really have a coil inductance value in the some hundred microHenry range, then all you may find is the generated output voltage is less than for a coil in the few milliHenry range. Induced voltage, V_i is linearly proportional to L,  V_i=L*dPHI/dt  where dPHI is the fluxchange in the coil and dt is the time under which the change happens.
So you can compensate the lower L by making higher fluxchange, of course there are limits to this (core saturation). Another choice is to find ferrite cores with higher permeability than the present ones

Nice build you  have!

Gyula

[nul-points]

ALL,

I want to give you some insights that I have into this dynamo that might help you understand it better. 
[...]
3)  notice that the flux is ZERO at maximum voltage according to the laws of electrodynamics, and the flux in this region of current pulses is linear (I drew thicker lines in this region)

4) However, the most important thing to note is the short duration of the pulses, approximatley 0.5 ms.   Why is this important?   It's because it is a shorter duration than the TIME CONSTANT of the coils when shorted
[...]
EM

hi EM

appreciate the work you're doing investigating the little amount of data we have from the Romero's actual build!

can i just throw another piece of information from the actual build, into the mix?

when i ran up a trial on my rotor with a single coil, and then a 1-position coil pair, to get a ballpark value for the turns/volt i'd be aiming for, my single coil waveform was the standard 'sine-with-a-gap' (see trace below)

notice that this looks significantly different to Romero's trace

ONE of the transitions from one peak to the opposite should be a smooth line (as the mag passes across the coil) - Romero's trace instead has NO smooth transition between the peaks (ie. there are two discontinuities, or gaps,  present)

i designed my mag diameter to mag gap ratio to be close to Romero's, so the wider, flatter 'back-porch' on Romero's waveform won't be due to a wider mag gap

the new insight into Romero's trace came when i connected my first coil-pair together

my new trace looked just like Romero's - two gaps (or 'porches'), much narrower pulses instead of previous 'sine' section - also that rather 'irregular' look to the whole trace - like each cycle was unique
(i didn't save the trace unfortunately)

AND THEN I REALISED...   I HAD THE TWO COILS IN OPPOSITION!

they still gave a decent output  (even on my 10 ohm resistive load) but they had this very irregular 'pinched' look

Now - the significance of this observation to your work on the induction from each coil pair is this:  in the single trace which Romero showed us the slope of each 'half-sine' section is actually STEEPER than the regular sine slope (because the pulse has been 'squeezed' into a shorter proportion of the cycle)

the result is that the flux rate-of-change is greater and also there is a better match of the pulse width to that section at the peak of the waveform where the coil energy dumps into the buffer cap through the FWBR


SO - the big question is this: "DID ROMERO HAVE ALL HIS COIL PAIRS WIRED IN OPPOSITION, OR WAS THIS JUST A 'ROGUE' PAIR?!?"

if ALL Romero's gen coil pairs were wired in opposition, this would be a significant factor in the total performance of his device


keep up the good work
np


http://docsfreelunch.blogspot.com

[yssuraxu_697]

Very relevant observation in itself, this waveform shape bothered me too. But I was not able to reproduce it just by wiring single coilpair in different ways. Could you please post a pic how exactly was it wired, where was load and what were measurement points?

[nul-points]

Very relevant observation in itself, this waveform shape bothered me too. But I was not able to reproduce it just by wiring single coilpair in different ways. Could you please post a pic how exactly was it wired, where was load and what were measurement points?

hi yssuraxu_697


here's a schematic for the test waveform i posted above
(and a photo of the test rig in action, with single coil)


config details:

each coil is ~ 100 Turns of 7x0.09mm multi-strand solderable wire
(narrower version of Romero's wire, from same supplier)

DC resistance approx 1.4 ohms per coil

core is 20x6mm OD 4B1 ferrite from Farnell

stator mag was 10x10mm OD Neo, fixed to 20mm OD mild steel washer

load was 10 ohm resistor
(5W? encased in ceramic block, probably wirewound)

pk to pk voltage on single coil was ~ 3.2V into 10 ohm

true RMS o/p (single coil) into 10 ohm ~ 90mW
(plugged PC scope trace data into Excel, (average(sum(instantaneous V*i)) * time)

waveform pulse freq: 278Hz

6 mag rotor, so rpm = 278 * 60 / 6 = 2780rpm

intending to use 3W DC motor drive for rep
(unless some of the magic turns out to be in the pulse motor!)


you say you couldn't get a similar w/f - were your test configs on a good amount of load?


anyway, hope this helps
np


http://docsfreelunch.blogspot.com