...
 When I move the magnet in and out of the charged solenoid the voltage reading is steady at 9 volts.  If the battery is not connected I can induce voltage spikes of ~4 Volts. 

Hi Mondrasek,

The 9V battery also has a very low inner resistance like a good power supply so any voltage drop caused by changing a current in the circuit it feeds is also very low, hence you cannot see it with a 3 and a half (or even a 4 and a half) digit resolution your multimeter has.

...   I also see no change in current with any of the experiments, but here is the weakest part of my test set up.  I have no current probe to use with the o-scope.  So I have a Fluke multimeter reading current and that may miss any transients.  But I cannot get it to read anything but the expected 120 mAmps even while pushing the magnet through the solenoid field (which should be generating +- 4 volts).

Well if you have the true RMS type Fluke multimeter like Fluke 87, then you may try it reading AC current in its most sensitive (ACmA) range. In this AC current range the 120mA DC current will not show up of course but you may be able to cause some small mA change when moving quickly the magnet away or towards the solenoid.  You can test this without the battery first, (especially if you do not have the RMS measuring type Fluke) by setting the most sensitive  ACmA range and connecting parallel the tips with the solenoid, then quickly moving the magnet in and out. You will hopefully see some peak mA change...  EDIT: If you see about 4V induced peaks across the unloaded solenoid, then you close the solenoid with the ACmA meter, the AC peak current will approximately be 4Vp / 75 Ohm=53mAp in the solenoid.

Here is some hint on making an AC current probe if you happen to have a higher permeability toroidal ferrit core... :
http://cappels.org/dproj/aciprobe/ACCurrentProbe.html
No need for calibration and the low value resistor termination, only would be useful for indication with your scope...

rgds,  Gyula

That lead me to what I believe is the problem with my test set up and measurements.  I am using an industrial 24 V power supply.  This supply is likely adjusting current/voltage for the changing load and not allowing me to see the real differences between charging the solenoid with and without the permanent magnet in place.

Actually that is the job power supply to keep it's voltage as constant as possible even as it's input current changes.
And yes, current probes will help. Unless you have a "crappy" scope it should be fast enough for these inductor speeds.
Also if you has any choice it would be better to use FET field effect transistors like IRF511 as they have very high off
resistance and can have very low on resistance like .01 Ohms on for specials, much closer to relay contacts. FET are
voltage mode triggered and that makes them easier to understand bipolars. Bipolars (2nxxxx) are current mode controlled
and harder to design with and that control current has a cost - they are often used less efficiently.

 If you want to do power measurements you need to put a resistor across the coil when you disconnect all else - make
 it identical to the DC resistance of inductor.  That way you transfer maximum power, when source and output resistance
(impedance) are equal. <= read this; it answers why you seem to see no effect when you have power connected vs nothing.
A battery's job is to have very low internal impedance.  As Guyla said. Good Luck!

:S:MarkCoffman

Gyula,

I see that I misunderstood your suggestion.  I look forward to trying it correctly.  Unfotunately I must visit a customer tomorrow and will not be able to play with my simple test set up.

When I modified your suggestion to "in series" it was only a guess that you had accidentaly said parallel and truely meant in series. 

I have access to quality ferrous toroids and built a simple AC current probe this morning but of course it did not work on this DC circuit.  I suspected as much.

At least with every failure I learn something new!  Maybe that is what I enjoy the most?

Thanks for all your support.

Any other way to see the current trace of a DC electromagnet when it is initially energized?

M.

If one is to energize any electromagnet (you pick the specs) for 1 msec, what is the power consumed by the electromagnet?
vs.
If one is to energize the same electomagnet for 1 msec, but this time, with a permanent magnet (you pick the specs) in close proximity such that the permanent magnet N pole is facing the electromagnet N pole so as to repel the permanent magnet, what is the power consumed by the electromagnet?

The power consumed is determined by Ohms Law.  See the attached pic I
find useful.
For this calculation:
P=U2/R
so if you apply 10 volts across a 5 ohm coil:
(10x10)/5 = 20 watts

Is this the thinking of your question? :
Will the electromagnet in proximity to a permanent magnet require additional energy because the field of the p.m. is 'sucking' up the field of the e.m. requiring additional input?

Has this been answered? What would be the answer?

At first blush I would have said no - but am now thinking - 'maybe'?
And why/why not?
And the variation due to induction of the moving p.m. near the e.m. windings would be negligable? (see following)

If the electromagnet is energized in proximity to a permanent magnet in repulsion the power consumption should be slightly more, maybe 21 of 22 watts.  When a magnetic force (the permanent magnet) is moved away from a coil of wire(in this case caused by repulsion) it generates electricity, so the power level rises, although only slightly.

So the power would rise because it's in repulsion resulting in an 'opposite flow' of charge due to induction of the p.m./windings to the e.m. requiring additional energy from the power supply?  And 1 or 2 watts?  Would that be a HUGE magnet?  Or would it be more like .1 or .2 watts (or less) for a small neo (.75x.5x.25)?  The p.m. is not moving ACROSS the windings but away, so even smaller?

Interesting thread - and I have alot of questions/thoughts to follow

tx

@capthook, thanks for all the thought and input.  I'm glad you are interested by these questions.  When I first asked them I thought I would receive a simple textbook answer or reference to experimental data that would make everything clear.  I didn't imagine this was going to be a testing/theory brain teaser.

@all

So after looking further into Bedini for a BEMF capture circuit I went ahead and made Imhotec's Bedini fan this morning.  Destroyed quite a few good fans trying to find one that had the correct coil wrap configuration as well as trying to understand the various instructions.  But it works and I'm trying it out now to recharge and old NiMH 9V that has been in a box for 6 or so years since I was last into micro electric RC planes.  That circuit is what I wanted to understand so I could apply my ideas *if* the permanent magnet does not consume an equal amount of power from the electromagnet when it is fired vertically.  I still look forward to trying everyone's ideas back at my deck at work with the test setup on Monday.

I've coorresponded with mscoffman a bit in the background as well.  He said he has some more info to forward once he has the time to collect it and pass it along.  I'm interested in his response to my questions about the Bedini SSG circuit as well.  I'm trying to understand it from a clasical EE point of view (ie without the "radiant energy" theories) for now.  I'd like to hear your opinions on that as well if you have any.

Thanks,

M.

I put a 1.6 Ohm resistor in series with the solenoid per Gyula to read the votage drop across it as an indirect way of measuring current in the solenoid circuit.  I'm not really interested in the absolute values, just the difference in the current over time as the permanent magnet is accelerated away from it's starting position inside the solenoid.  Again, I did not see the results I expected.

Without the permanent magnet inside the solenoid the voltage measured across the solenoid and the in-line resistor both rise to their steady state values and hold when I press the switch connecting them to the 9V battery.  The o-scope traces look exactly like a square wave.

With the permanent magnet inside the solenoid the voltage measured across the solenoid and the in-line resistor both rise to the same values initially.  But the voltage across the in-line resistor then drops slightly before returning back to the expected steady state value.  The shape of the curve of this voltage drop and it's return to normal appears to be a nice parabola.  I assume the drop in the voltage curve represents is the permanent magnet accelerating and the return icurve representss the two magnetic cores increasing in distance.

So does this mean the current to the solenoid actually decreased as the permanent magnet was accelerated away?  Did the power consumed by the solenoid actually decrease as it did work upon the permanent magnet?  Or where (oh where) is the measurment error now?

Thanks,

M.

@modrasek;

If that is what you measure then that is probably it. One thing to think about is the
fact that the magnet has two poles and what does the solenoid see as it accelerates
the magnet through itself. It could be that the solenoid's stronger field actually forms
lenz current in the magnet. That bounce could also be from the power supply recovering.
See a better method below.

So if you see a decrease in current it maybe true. Don't forget that energy is measured
in "milliwatt seconds" and time plays a factor, if you have to keep the solenoid energised
longer to fire the magnet and increase the field then it has potentially used more total
energy.
 
IMHO you will not see overunity when an electromagnet accelerates a magnet. You will
see overunity if one PM lifts another. (but then lose momentum of the wheel trying to
pull them away from one another). There is also Smot runner gain, but one hasn't (yet)
extricated the runner to get it to the beginning of the track array.


@ALL

If you want to see overunity energy production from the Bedini
Fan I recommend use of acid/lead storage batteries or Gel Cells
for both the source and the charging battery as the overunity
part occurs due to battery chemistry. Don't use strange battery
chemistry ..and then say that the overunity part doesn't work,
please.

Small Acid/Lead Batteries are available in those automotive jump-start
units and small grey or orange Gel Cells are available in building
emergency lights that come on when the utility power fails. I like
the small batteries used in ICE motorcycles. They also make 9Volt
size acid/lead's I believe. 9volt batteries are probably too
small to support the fan well but it's worth a try.   

imhotep's Bedini Fan is an excellent experimental device as it doesn't
have an electrical interference footprint much larger then a normal
DC fan. imhotep's Morray's Vibrator Overunity Light unit also is
an excellently simple device, some experimentation will be required
to modify the base unit for other applications however. See
imhotep's youtube.com videos for more details on building these devices.
A Bedini Fan might be a valid manufactured product...overunity included!
Rather then build a Bedini SGS motor which I don't consider a good
experiment why not include *your own* electrically run wheel motor as
an actuator in a custom version of a Morray's Vibrator circuit?

imhotep's videos:

http://www.youtube.com/user/kojsza


The principal being demonstrated;

http://uk.youtube.com/watch?v=qaCk0jK--8s


----

Magnetic Pulse Experiment

The following experiment should let one see then pulse from a magnet
in a straw accelerating away from being fired by a solenoid coil when
you apply power through a switch. This isn't easy to do experiment but
here goes;

Remember that a capacitors store power proportional to it's voltage
while inductors store power proportional it's current. So it is
easier to think about inductor equations in terms of current.

What we are going to do is look at the current flowing through
two coils simultaneously and use one coil as reference and
subtract that from the coil that accelerates the magnet using
functions available on most oscilloscopes. The "invert" and "add"
channels functions.

a) you need;
   two identical solenoid coils and if you don't have scope current
   probes we can use two "current transformers" instead. (see Wikipedia)

http://en.wikipedia.org/wiki/Current_transformer

   A valid current transformer is a 20KHz bandwidth audio transformer
   line-to-load transformer with a 600 ohm impedance primary
   (audio line = 600ohms) and a 1.6 ohm secondary or the lower
   the better. I would say 100Watt audio transformer
   (audio bandwidth = 0->20KH) you then must solder a 600 ohm 5watt
   resistor across the 600 ohm primary or match the primary resistance
   in a way that is won't accidentally disconnect else a current
   transformer can become a shock hazard and damage equipment. Now
   connect the low resistance secondary(s) in series with coil(s)+power
   supply and attach one each primary+resistor to each of two scope voltage
   channels.

b) now flip the "invert" switch on the reference scope channel

c) find the "add" channels button on the scope

d) the "invert"+"add" now equals "subtract" one channel signal from
   the other

e) now one has to adjust the gain of the channels so that the pulse
   visible then the power supply is "fired" into the two solenoids
   then adjust the channels so that the difference is as small as
   possible when there is nothing fired from both solenoids.

f) You can validate that swapping the straw and magnet between
   solenoids should create a plus going pulse when the magnet
   is in one and a minus going pulse when it is the other.
   The magnet uses most energy at the beginning to accelerate
   therefore the pulse.
 
g) you can also watch as a magnet flies-by with an some other inductive
   coil attached to the scope. There will be a bidirectional pulse
   at the moment the magnet flies past, with the zero transition
   at the point of closest approach. The Bedini Motor 2n3055 transistor
   circuit uses a sense coil that works this way.


:S:MarkSCoffman

@mscoffman.  Great information as always.

@Gyula.  I was able to see the correct exponential curve when adjusting the timebase as you suggested.  Unfortuantely I was working with another unfamiliar o-scope since the one I had played with earlier was out in the field with a technician doing actual work (not my desk experiments).  This new scope was taking too much time to learn and the only way to see the current drop due to the magnet on screen was to change the timescale so that the exponential curve was compressed to look like the square wave.

I was very interested by the current drop, but as mscoffman says, this drop is over a much longer time period than I would have expected compared to the saturation rate of the circuit.  Interesting effect, but what could it be good for?

I was curious if I could keep the circuit/solenoid charged only as long as the current drop was occuring.  In a failed attempt I replaced my mechanical switch with a custom design, utilizing the perminant magnet as a switch contact.  I placed the two wires oringally connected to the switch into the bottom of the tube supported by other elements so that they were contacts that would be bridged by the permanent magnet.  With this in place I then was able to move the solenoid up and down the straw to different locations with respect to the permanent magnet, into repelling and attracting, up and down configuarations.  When the solenoid pushed or pulled the magnet upwards it would break the circuit and allow the magnet to fall again, re-connecting the current.  The result was interesting at best.  The resultant pulses of current applied to the solenoid would only raise the magnet a less than noticable amount, though you could hear and see the vibration and arcing.  So the switching frequency was fairly high.  Putting the multimeter across the solenoid in AC mode caused readings from several hundred milliVolts to several Volts above the DC supply Voltage.  I'm not sure if that was due to the moving permanent magnet, the BEMF, or both.  But it was interesting all the same.  I'd like to see a similar setup that allowed the solenoid to to be energized for longer.  I don't see this as being easily possible with a mechanical set up and instead would require a variable rate switching circuit.  Again, I'm not sure what this would accomplish, but now I'm just playing while trying to think where the current drop effect could possibly be useful.

Thanks again to everyone for all the info and ideas.

M.