Overunity.com Archives

Hi People,

"newman spark effect" not in pspice simulation...

Thanks Igor, for resparking intrest in this metheod, Wherever you are.
The old 7 reed relay circut worked exactly as described. EXACTLY.
only problem was relays welding shut and short circuiting, Igors 4pole relay should not do that :)

Enjoy

see also:
http://www.overunity.com/index.php/topic,133.0.html
http://drspark.com/idea003.php

For more power add more to the value of the zener
In the simulation; If D3 is removed, power flows back to the battery, probably ruining it...

Igors addition of the inductor has made this act more strange imho...
and is why it pumps up

Im useing student version of circuitmaker

any ideas on the wattage and cap voltage ratings so nothing breaks?

Have built and test this circuit?

Pakis_ch

Hi,
what is primer/seconder properties T1 transformer (2x12 V-220 V ?).

Hi People,

I have constructed recycle circuit.
Auto bulb replace R2 and another is in series with T1 and C3/C4
Seperate 555 circuit drives the relay...

here is photos of operatioin


:)

The circuit

6volt bulbs replace the 12volt 12watt bulbs

.
.
The Relay must be converted into a rotating commutator,
Or, a solid state 4pdt relay.......

Anyone know how to do the solid state (FET) 4pdt Switch with no biasing or back currents that will change dynamics of switching action?? true open(s) and close(s) of a few Mohms or 0ohms

Thanks
Sparky  got this nick because i tuned a lodge coil, with amazing results (they had way to much cap)....

Hi People,

I replaced the light bulbs with 10-ohm resisters.
The resister in series with the supply gets hot, but you can still touch it for maybe 3seconds, and the output resister is untouchable!

have fun everyone....

:)

Dr.Spark,
very nice !
Do you get already more Watts at the second
lamp than you put into the circuit yet ?

Regards, Stefan-

Hi there Stefan,

Sad to say, I dont have access to equipment to perform measurements worth arguing over, anymore, BUT, I have replaced the Bulbs and their varying resistance with two identicle 10ohm resisters.
The resister in series with the powersupply is noticably cooler than the resister between the capacitor banks, which becomes to hot to touch, and is now discolored.

That is an easy to feel, and somewhat visable caloric differance!

The Spice simulations seem in error, and I have sent a copy to them and will post any reply.... :)

If their not in error, all we need is a *perfict* switch like what is simulated by the computer running spice or CircuitMaker I used.

At this point I dont know how to replace the relay(s) with solid state but that has to be next. (or a spinning comutator, tubes, anything faster) The higher the frequency the better the effect...  noticable when I change the 555 circuit trim pot controlling the relay.
I can see the relay loose it when the freq is above 25-30hz and the circuit malfunctions.

555 circuit has R1=1k, R2=1k+10ktrimpot, C1=10uf.

This was not expensive to repeat, I did the sims while waiting for the relay and caps in the mail from mouser, after seeing wizkycho's rep here!

spark

Hi DrSpark, without mechanical contacts it will probably not work. Was the same with the Newman setups. It needs the sparking in the air to do electron clustering at the contact points to convert additional energy into the circuit. With just electronic transistor switching it will not work. So a faster turning mechanical commutator and higher voltages is the way to go for more energy output.
Regards, Stefan.

Hi Drspark,

Ok I'm convinced.
I'm going to build this with mechanical switching.
If it works ok for me too, I have a friend of mine with huge experience in electronics 
and he is going to convert it to electronic switching and see what happens.
Do you thing that a bigger transformer will make any difference to the output?
I have access to many parts from PC PSUs and I can change the transformer with something bigger than what you use.

Pakis_ch

Hi Pakis_ch,

My transformer is only 24 turns of two strands #24.
I measured out 3feet found center on ONE strand and sandpapered a bare spot and soldered the greeen wire on. Then I put both strands in a drill and lightly twisted them to make treading them thru the toroid easyer.

I connected a 12v lightbulb to the uncentertapped strand and wired the primary as showen in the various drawings...

It barely made red the filiment. So, now their both connected like litz wire but only one is centertapped to the diode.

Now my reason for a low coupling was fear, Im afraid that the spice sims maybe correct, and the cap-trapped cemf from the transformer will cause a voltage rise and pop my caps. Ending the fun for atleast a month. The sims all ran away without the zener clamping the voltage to a limit.

If you look at the circuit, any backemf or cemf from the transformer cant get to ground without integrateing onto one side or the other cap bank.....  and there lies the uncharted side of this concept, which is wizkycho's addition, (the Transformer) also sugjested by Gyula a couple of years ago.
I wish I had simulated it back then  :-\

Looking forward to your replication and insights  :)

spark

Hi People,

Making the circuit as simple as possable;
I went back to the simulator, I still get a greater output than input, but at the lower frequencys, thats what I experience now with the circuit in real world.

By removing the inductor from the sim/circuit, it no longer runs away in voltage, very interesting.

Im begining to think spice is correct.

Anyone wanting to Sim this can dload a free, student copy of circuitmaker, just google circuitmaker download.

At 8-10hz the input resister is cold can be touched to your lip, however the output resister is far to hot to touch.

This is inverted from light bulbs that appeared to show greater output at higher frequencys...

spark

CircuitMaker2000 Setup
In analog options only GMIN is changed.
that fixes the time step error from low freq calcs (10hz is a long time)

To make sim run longer edit stop time in transient and fourier analysis setup dialog box...

spark

Hi People

With Gyulas help; I have converted the circuit to solidstate in the sim.
Im searcing for the best cap values and freq, then ill be building this one...

Enjoy

Hi drspark,
thanks for circuitmaker help but ithink power&cop calculation..

total power given circuit is : Power(V1) 
output power =Power(Rin)+Power(Rout)+P.other components.. , so? how can compare input/output power or energy.

like Igor Knitel circuit, with 0.1 ohms/5Watt serial resistor voltage and current
we calculate given power  to circuit.

Hi there Geszin,

If *collect data for; node voltage, supply current, device current and power* is enabled, the simulator will display the power data when you place a probe ontop of the device.

The 10ohm resisters are devices posistioned in series with the battery and at the load point. I use the wave forms calculated as disapated power for those two devices as a measure of the performace.

The power required for switching is not included yet...

spark

Hello,
I just joined this extremely interesting forum. Yersterday I have built and tested your circuit (the one with relays) and it works as you described. Playing with freq and duty cycle you can (quite easily) get battery lamp off and load lamp lit up. I added two 1 Ohm resistors in series with the lamps and felt the different temperature as you did; this further proves that battery current< load current.
However I think we should evaluate the input power as battery current x battery volts  x power factor (because it is not DC); the battery resistor dissipate some power but this is not the input power.
Am I missing something?
Can you calculate the input power as Volts x current in your simulations?

Thanks a lot

Hi Guido,
exactly, the input power must be measured by battery voltage x input current x power factor, as the input current is pulsed.
The input current can be of course smaller than the load current, cause it acts as a capacitive current transformer.

Hi Guidoc, Stefan,

CircuitMaker calculates the PIV data every timestep which in most cases is 1-250uS.

The waveforms of r1[p] and r2[p] are spice calculated power.

Stefan is correct, this is a current transformer made from caps and it divides the voltage by 2 if DC averaged,
BUT
In the sim a most interesting thing happens when you add a transformer to the middle of this. (thanks Gyula and Igor)

When a simulated capacitor (current) transformer, is merged with a simulated inductive (voltage) transformer, simulated power output rises such that a simulated zener was needed to clamp the simulated output down. :)

Im working to get the frequency up (in a real circuit) so I can place a transformer in the circuit and go back to dc like Igors claim.
Then ill be able to measure the voltage of the cap on the other side of the transformer and bridge with a my multimeter.

If thats real, this is OU and Power is selectable. (true Tesla switch?)

Now im not saying the capformer is not interesting alone...
I also agree with Guidoc analysis of real circuits need be done.

I am not an expert with the simulator ;)  or electronics, just a retired software engineer... being retired from industry I dont have access to the equipment needed to do real world analysis anymore and the people who would already have made the fet version.

I cant and wont argue with only a multimeter(s) in hand, unless its dc.

I only have 11days left before the trial version expires.

Spark

Drspark, Stephan
thanks.

I'll try to make some more tests and maybe take some scope pics.

Stephan,
I would like to test the so called spark-Newman device" you mentioned already. Where can I find some drawing? I remember a plan with a relay connected to a transformer and then to an 8W neon lamp but I was wondering if there is something different or updated.

Thanks.

Berkleys SPICE Simulates overunity solid state!

If this is real, cop is thousands :)

sims with relays and and now mosfets!

The zeners are 11v each and limit output...

spark

Hi People,

A much simpler circuit! this, I plan on building next month.

Enjoy...

Spark

Quote from: guidoc66 on February 09, 2007, 12:55:41 PM
Drspark, Stephan
thanks.

I'll try to make some more tests and maybe take some scope pics.

Stephan,
I would like to test the so called spark-Newman device" you mentioned already. Where can I find some drawing? I remember a plan with a relay connected to a transformer and then to an 8W neon lamp but I was wondering if there is something different or updated.

Thanks.

Hi look into hear:
www.overunity.com/newman
www.overunity.com/newman2

Thanks !

QuoteBerkleys SPICE Simulates overunity solid state!

If this is real, cop is thousands Smiley

sims with relays and and now mosfets!

The zeners are 11v each and limit output...

spark

Hi People,

A much simpler circuit! this, I plan on building next month.

Enjoy...

Spark

Great Dr.Spark.

Looking forward to more infos from you.

I guess this is a simular circuit with capacitors
as what Mike has now shown with his modified Bedini-Cole-
Window motor. He switches coils back and forth and uses dL/dt
and you switch capacitors back and force and use dC/dt.

Normally with coils you have Voltage_induction = L x di/dt
Now if you also change L over time it must be something like:

Voltage_induction =  d ( L(t) x i(t) ) / dt
which has a much different solution, where more energy important terms
come up.
I have to dig this math calculations up in my old Newman files.

The same goes for the case with switching capacitors
with the additional dC/dt terms.

Regards, Stefan.

Stephan,

Yes! please dig for the math...

Thanks in advance...

Spark

Quote from: drspark on February 21, 2007, 02:14:16 AM
Stephan,

Yes! please dig for the math...

Thanks in advance...

Spark

Okay, will probably take a few days, cause I don?t know where I
have put this paper and have to search it all over my
place and now I firstly need to play with my old Newman machine
for this Mike Motor thread...

Regards, Stefan.

Hi People,

Berkleys SPICE Simulates overunity solid state and SELF-RUN
by replacing the battery with large cap, removing the blocking diode, and telling the simulator some Initial Conditions, one to set the start cap to 12v and the others help the sim start the multivibrator.

Run the sim and the start cap voltage drops as the circuit charges up and then recovers as the output stablizes. another zener was needed to protect the startup/run cap....

another most interesting result.
anyone out there build this yet? please share your findings.

Best regards,

Spark

Quote from: drspark on February 03, 2007, 10:14:43 AM
Hi Pakis_ch,

My transformer is only 24 turns of two strands #24.
I measured out 3feet found center on ONE strand and sandpapered a bare spot and soldered the greeen wire on. Then I put both strands in a drill and lightly twisted them to make treading them thru the toroid easyer.

I connected a 12v lightbulb to the uncentertapped strand and wired the primary as showen in the various drawings...

It barely made red the filiment. So, now their both connected like litz wire but only one is centertapped to the diode.

Hi Spark!

Question about the toroid transformer.

Is this the correct schematic for the replication you made?
http://drspark.com/2cycle.jpg

http://drspark.com/6vbulbs.jpg

I?m unsure about the toroid winding.
You have two 3 foot lengths of  #24 gauge wire.
On one length  you found the middle and attached a green tap wire.
You then twisted the two 3 foot lengths together.
same twist direction for the length?
Or, twist one way to the tap then twist other direction the rest of the way?
Then the two  3 foot wires are soldered together at their ends?
Or is one primary and one secondary?

I?m unsure about your bulb connection . you said you connected it to the
Untapped 3 foot wire which suggests that at least on one end the
Two 3 foot wires are not connected together ?

Thanks,

Hello Electrostatic

In the final version of the relay circuit, the transformer was disconnected. Just connect the input diode to the top of c1/c2 bank,
and the output resister between the two banks.
like 2cycle.jpg with the inductors shorted.
The 1000uf C5 is also disconnected.

Hope this helps

Spark

Quote from: drspark on February 23, 2007, 02:15:26 AM
Hello Electrostatic

In the final version of the relay circuit, the transformer was disconnected. Just connect the input diode to the top of c1/c2 bank,
and the output resister between the two banks.
like 2cycle.jpg with the inductors shorted.
The 1000uf C5 is also disconnected.

Hope this helps

Spark

Hi Spark,

Thanks.  I thought that the ?interesting? part was the transformer.
I still am not that convinced that the input bulb is letting RF through
and that , that?s the reason that the bulb is glowing dimmer than
the output bulb.

BUT,

As Tesla said himself I believe, about his fuel-less generator,


?The radiant energy system is a self-oscillating capacitive system. Once it is set into oscillation, very little power is expended in keeping it going. Because it is an electrostatic oscillating system, only a small amount of charge moves through the system per cycle?

http://home.comcast.net/~onichelson/Fuelless.pdf

From,
http://tesla.nichelson.googlepages.com/home

So I think you might be on the right track

Hi,
Yes i tested 4 contacs relay switch device with 4*10.000uF/63V Cond. ,toroid coil,
power supply 12-18 V
condansator voltage was increase to 1000V and multimeter reach limit.Im worrieng conds.explosion .
in circuitmaker simulation, playing C&R values getting strange results & errors.
sometime show big voltages.

Hi Gezgin,
watch out, that your caps dont explode. Put a zener diode or a few white LEDs in series across them, so when the cap voltage reaches zener or LED threshold voltage, they will conduct and dont increa e the voltage anymore. Please post pics or videos. thanks.

Quote from: gezgin on February 23, 2007, 08:04:29 AM
Hi,
Yes i tested 4 contacs relay switch device with 4*10.000uF/63V Cond. ,toroid coil,
power supply 12-18 V
condansator voltage was increase to 1000V and multimeter reach limit.Im worrieng conds.explosion .
in circuitmaker simulation, playing C&R values getting strange results & errors.
sometime show big voltages.

Hi Gezgin,

a 1000 volts through 63v caps? RF spikes?
Was this simulation only?

Hi Electrostatic,

The relay i have is only realy capable of about 20hz, the transformer i made for the circuit needs many more turns to work welll and higher frequency than 20hz, so i removed it,

In the sim the transformer is the reason why the voltage climbs and is the most interesting part of the circuit.

I have to wait till next month before I can gather the needed componants to build the !!self running!! fet version. see bottom of pg 3

Interesting about Tesla; he would have tryed this, probably used the earth as one of the capbanks  ;)  humm.....

Best regards
Spark

Hi Spark!

Well I may have some not so encouraging  news on this device.
this in not the definitive answer , and your circuit might be doing things
with that inductor, but I would like to point out the (simple) reason
the bulbs can have a difference in brightness.

I also, like you, looked at the resistors or lamps as being  equal so could
not see how a difference in brightness could occur.
But its really simple for this device to light (or heat) the R1 brighter than R2(off the battery) (step A,B,C, diagram at top of page http://drspark.com/idea003.php)
without RF and without sparking being involved.

Think of what could happen if the two resistors where NOT equal.
A difference in temperature is possible.

It works like this, the I?s  (current) is not equal. Effectively if this was all in one cycle
You could look at R1 as a higher value resistor ,the ?bottleneck? in the circuit.(its not though, the real reason Is the other side of the equation  the amount of amps running through R1
In the end)

From Gary  in one of my yahoo groups,
???..
?The circuit in step A has both resistors in series. If they're the
same resistance they will absorb the same amount of power because:

power (w) = R i^2

Since the R's are equal, and since they're in series the i's are
equal, so their powers are equal.

But in steps B and C resistor R2 isn't doing anything but cooling off.
Note that this is for five sixths of the entire cycle. During that
time (the five sixths of the cycle) resistor R1 continues to absorb
power (in diminishing amounts) so it gets and stays hotter than R2.?
???.

So, simply put the I?s are not equal, the R1 resistor simply had more cycles
(more current absorbed)
Which in a way is the same as a *different Value resistor*
**and different voltage for R1. Hence
different brightness, different heating.
very tricky,,,,

But I must say, the series parallel switch is still very interesting to me.
And might have interesting possibilities. Especially with an inductor added.
And also thinking of ?mikes? Bedini generator with a trifilar inductor
which I think does do  series parallel switching somewhere in there.

My test circuit, digital multimeter show interesting volts.
i think ,its maybe coil back-emf ..
I checked old analog voltmeter, its show only 10-12 Volts..
I will try with zener&led.

another circuit(simulation only)

Hi Gezgin,

Great!
Can you describe your transformer?
I have a couple of torroid cores from pc power supplys,
Is that where you got yours?
My relay started malfunctioning, after about a total of an hour.
So try to limit your run time(s)....

sugjestion; get the highest voltage value cap you have and use a bridge diodes after your transformer to convert back to dc and then measure voltage at that cap after 1sec then discharge every cap and run 2sec then 3, 4...

The resister in series with initial charge source must be there. In the sim, if its not the sim_zener breaks from the startup charging current and the voltages go straight up, causeing the transient analysis to fail and in real world if this whole concept works a cap!

In a MOSFET version of this, there must be delay (sawwave drive) in the switching state or they will short out and circuit will malfunction, simulatedly speaking.

How can I tell if my driver really turns the MOSFETS COMPLEATLY ON?

EVERYONE HAVE EYE PROTECTION! maybe face and throat too.


Spark

Hi People,

For those out there experimenting with the circuitmaker or other spice program; To help with the *Timestep to small error* I copied this from the manual;

Spark

Quote from: drspark on February 23, 2007, 01:28:18 AM
Hi People,

Berkleys SPICE Simulates overunity solid state and SELF-RUN
by replacing the battery with large cap, removing the blocking diode, and telling the simulator some Initial Conditions, one to set the start cap to 12v and the others help the sim start the multivibrator.

Run the sim and the start cap voltage drops as the circuit charges up and then recovers as the output stablizes. another zener was needed to protect the startup/run cap....

another most interesting result.
anyone out there build this yet? please share your findings.

Best regards,

Spark

Hi Dr.Spark,
what is the switching frequency ?
This looks very simple to build !

I will do this after I am am finished with the Mike motor.

Regards, Stefan.

Hi Stephan,

Looks to be about 150hz,
I renumbered the componates.
Also starting to gather parts for construction.
Odd caps are ones I have...

Time to settle this, on the bench!

Spark.

Hi Spark,Stephan
toroid from local store , it maybe from www.chinadmegc.com(iron powder toroid core) .
diameter: 36mm.coil 0.5 mm wire .inductance is 10-15 mH.
power supply 12-16 V adjustable.
I calculated 555 osc.frequency 32-33 Hz (RA=10k,RB=4.7k,C=2.2uF+100nf).
digital multimeter shows stupid results, idont think  problems about capasitors any more.because connected to caps. a neon bulb and no light on.(only small light at one caps.) I'm curious about caps. wave forms & need scope! ???

my opinion is that creating solid-state overunity is hard .. still we need rotating magnetic fields, magnets ..

Quote from: drspark on February 24, 2007, 07:53:04 PM


Time to settle this, on the bench!

Spark.

Hi Spark,

Since there where two different setups on this page 

I just wanted to check something out with you.
Was the battery input on this http://drspark.com/12vbulbs.jpg
pulsed at 1/6 cycle compared to the output bulb?
Like this diagram  http://drspark.com/capamp.jpg

Or like this one where the battery is always in the circuit
http://drspark.com/2cycle.jpg which I think would need
A different explanation to show the extra cycles through
The output bulb..

ADDED:

Ok, the explanation for this http://drspark.com/2cycle.jpg
would be the same as for this one http://drspark.com/capamp.jpg

Only slightly different.

In the shown switching position you have a LC tank circuit
Comprising of  C1 +C2 in parallel through L1+L2
To C3+C4 in series. This is a ringing circuit (AC) which would have the same effect
as the manual (extra cycles) switching of  http://drspark.com/capamp.jpg

So the first explanation still holds I think.

Cliff,

Some members ask me for this scematics.
possibly (sure) it cant work
because TR1 primare the bottom winding are
shortet with centertap via C4 against ground - and also the
food-end of this windings are SHORTET with C7 +C8 against
Ground.
So never an Frequency can bee transformed , and never start up
any Voltage to selfrunning  or only running.

never have this worked , possible als an error in desing.
Inventor will find out this...

Gustav Pese

Hi Electrostatic,

There are 3 versions of this circuit now up on my site.
The first one i built back in 1993 had 7 reed relays and was controled by computer useing the printer port output byte, it ran as described and was dismissed by a freind as a simple transformer.

But, it sure seemed strange to be able to burn a resister thru another identical resister, and since in theroy and practice this cap bouncing effect divides the voltage in halve and also *should* double the current.

But still the resister at the load point burns instead of one being twice as hot as the other...   

I left it on my site as a definite possability imho.

years later Igor added a transformer at the load point and the power connection to the middle of that, as soon as I saw that I started playing with the simulator and repeated my result from many years ago. So next i order a 4pole relay and a caps to try it again, same result as before, slightly different because you cant have any number of BC cycles (as per v1 capamp.jpg) but the same thing happens, the caps bounce the *charge* from one side to the other thru the t1 (like a push pull amp?)  and whenever there is an inballance in the voltage of the supply current flows untill the state changes or it equalizes, when the caps go series the voltage is greater than the supply so the inballance causes a flow back, now add in fact that the caps dont compleatly discharge every step or *state* the requirements from the battery decrease.

After making this simulate and work again on bench with 555 / relay.
I quest to convert to solid state. When I get solid state version to sim (thanks gyula pointed out fet circuit over at JLN site)
http://jnaudin.free.fr/html/scalchrg.htm

I change the load resister for the transformer trying to get back to DC when the whole simulation got really strange with voltages rising and in less than a second megawatts are on Rout and the sim chokes out of bits and times out. So I add zener to clamp off voltage and the sim voltage rises to about that voltage and becomes steady state. then I ask all, whats going on here? and save to buy parts for ver3.

I have all but the zeners, fets and a good idea how many turns to use on the torroid...

And the big question is IF the circuit as drawn will turn the fets ON compleatly, or burn them up?

Spark

Hi pese,

I will make an animated .gif that will show how there IS a varying voltage on the transformer.

I start on it now.

Best Regards
Spark

Hi People,

Here is a little .GIF animation I made for pese.
I hope this helps to visualize whats happening

Spark

This will also not work  Contact 1 and 3 are shorted over Cap to ground.
Possible to use as LC resonance device , But not useable with 100 uF Elektrolydt Condensors !!  (Must bipolar) Value fr L if C = 100-200uF ? Frequency ?
Something is possibly wrong in anyway .

Also in your circuit Pin 2 and 3 from Transormers are connected wit condensors to ground .  On Pin 2 (Condensor with Zener , you "hope" that an DC.voltage must be upload himself , never possibel that AC resonate in this point .
I think "some" will be changed in this circuit from you or members here...

Hi Pese,

I have never claimed this was a resonate ac circuit/consept.
The basic circuit is DC.
My understanding;
The abrupt switch(ing) causes the current in the winding to stop suddenly and that causes flyback or increase in voltage no matter what the polarity that pulse can integrate onto a cap, there is nowhere else to go...

here is another DC circuit (the multivibrater again) that converts DC to AC. Notice this circuit is square (basicly) waves accross the transformer, yet this circuit works great for running small appliances that can handle the noise from harmonics and switching.

Finding the smaller caps and higher frequency that would resonate the T1 could be a great improvement, but the sinewaves I see in that would kill the flyback integration and voltage increase, thats added to the current increase caused by the capasitors being swapped.

Spark

Hi Spark,
what kind of transformer did you use in your simulations ?

Is it a normal 50 or 60  Hz type with iron core
or a ferrite transformer, which can be used with higher frequencies ?

What is about the turn ratio ?

Many thanks.
Regards, Stefan.

Hi Stephan,

The transformer has a one to one turns ratio.

With the simulator I set up a test;
1khz from side to side 1v in, 1v out,
from one side to centertap 1v in 2v out.

Seems to be two wires equal lenght, one center tapped the other not.

The core type is unknown to me...

I plan on setting the multivibrator caps to 0.01uf for about 1.5khz and using a torroid. Question I have is how many feet of wire should I use?

Spark

@drspark

offcause , if your schematic in Reply #43 uis not rsonating wt the
elektrolytics than it is an "shortet (AC) Circuit !!
Tve C4  L /from Transfo and C7 are an closed Ring (over Ground Connection) So the Transfprmer ist shortet  for AC Signals that will be transferred to bridge rects ....
This is all . Pls try to understand
(otherwise i must not give any comments more)
Pese

@ drspark

also in message #51
the inverter circuit is wrong designed and can not work.

if sombody need to know, why ... ask on message or email
pese

...we are getting closer, what a nice progress - even if it is "only" a math simmulation...
Now look at the picture bellow... at first it doesn't seem much... Oo but think more than twice.
E Output rises by one unit (+1)... we've been ignoring other points of E extraction in our circuit, till now. This change will just make higher output, leaving input as it is.

disadvantage: little more complex switching.

wiz

@DrSpark,
if you use normal standard Iron cores for the transformer,
you can only use frequencies up to about 70 Hz,
otherwise your hysteresis losses will be too high.
If you want to go to 1.5 Khz you should then use a ferrite transformer.

@Pese.
where are the exact errors in the sims ?
I guess, if the sims show some excess, then
it is indeed something there, if he did setup the simulations
right with all the start parameters !

@Igor,
thanks for your new graphics, but it is hard to understand.
What do you want exactly to change now ?
Can you please explain some more ?

Many thanks.

Regards, Stefan.

@stefan,
ich reklamiere nur , das trafo in #43 mit elkos AC m?ssig jurzgeschlossen ist und somit BALLAST (load) darstellen.
C4,C/+C8 sind an Pin 2+3 des prim?rkreises angeschlossen.
Beide C gegen Ground.  Das ist derselbe "Kurzschluss" als w?rde man
einen Kondensator an 4+5 des sekund?rkreises h?ngen. Der Ballast
durch den Kondensator (f?r AC) ist so gross dass diese Stufe SO
nicht arbeiten kann . Sorry , diese theoretischen Dinge sind f?r mich schwierig in englisch auszudr?cken . G.Pese

Die Inverterschaltung (#51) hat auch "dicke" Fehler . SO ,nicht nachbauen!

Quote from: hartiberlin on February 26, 2007, 09:27:11 AM
@Igor,
thanks for your new graphics, but it is hard to understand.
What do you want exactly to change now ?
Can you please explain some more ?

Many thanks.

Regards, Stefan.

... to cut the story, I made a mistake. I expected for a moment that R2 would behave like a aditional voltage rise...well it is but with reverse polarity for R2 so If R2 is in circuit (R2=R1) voltage on R1 drops to half of what it was without using R2 so Ecase1=Ecase2...
I desperatly wanted to save some more charges...maybe there is another way...

igor

drSpark. When using simmulator circuitmaker2000 i'm getting same very much OU results.

The main difference when switching with relays (especially 4pole ones) is that there can not
be two outputs closed with one input contact. so there is allways strictly one paralel and one serial cond bank.

On the other hand in schematic with 6 MOSFETS there is a time when all 6 of them are opened/closed (no matter how fast rise and fall are) because when 10Hz source is at V/2
both Transistors are still conducting (probably satturated) so the conds are not neither in serial or paralel mode (on both sides)

In a schematic two with 3 mosfets one capacitor bank is allways paralel and the other changes paralel,serial (but again not strictly).

I run simmulation (sligtly modified schematic with 6 FETS) but with small pause between each transistor closes then other opens and gave me NO OU result.

When simulating Your schematics (no strictly defined opened or closed cap bank in a time period)
I wonder what are all the possible ways where the charge is flowing.
One of the most important,I think, is that when current flowing from source to both banks at the same time creating bucking flux flow in a core of Trafo which returns high back EMF which must be collected by the either of 1+2 capacitor banks and shown as rise of voltage.
I give this circuit very high probability it will work.

The question is: Are the simmulation programs that advanced that they are actally calculating
bucking fluxes in a trafo itself and calculating possible back EMFs ?

If so that would be an explanation why this device really works

Igor

A snipet from the CM manual.

in simulator , some coil tests circuit&waves :

next ill try to buck the ct winding

Bucking ?

Hi People,

Here is Simpler circuit! yet again,
This only works if the caps, c2 c3 have 2/1 ratio
This looks TESLA, Q1 input gap, D1 limit gap.

Far to easy to be true???  i think so...

Spark

9 Watt?s  output  ???

9 volts ?

If volts  this is "thing" of winding/ratio !


Circuit have never OU !


G.Pese

Igor,
Tried to validate your Tesla switch, see enclosed picture. Only got LAMP #1 to light and not lamps #2 (at all, not even dim). Used 6800uF caps 50 volts, 4 pole relay, doubel throw, Mos-Fet buffered 555 50:50 oscillator circuit. Variable oscillator from 6Hz to 60Hz. Lamp #1 glows very small at 6Hz and become brighter, linearly to 60Hz (brightest).  My feeling is that capactive reactance and inductive reactance cancel eah other, offering very little resistance to ground. Thus Lamp #1 glows as freq increases.  Inductance is 190uH coil, center tapped.  Resonance, for lowest impedance to ground formula is HZ=1/((2*pi*sqrt(L*C)). 

??  Any pointers on how to get Lamp #2 to glow at all?
v/r
Ken

Hi Ken,

Great looking rep,

check your wiring on back (standard reply)
remove connections to inductor
remove 1000uf cap

connect bulb where inductor was, between cap banks.
connect input power to top of any cap bank (diode to caps)

try again!

Spark

Spark,
Removed the 1000uF cap (ground lead), shorted the inductor at 3 points (essentially no inductor now) and still ONLY 1st Lamp glows, the faster the oscillator is switching. Measured the line on Oscope between the switching caps, and I do see a Very small and short duration of spike voltage, but maybe just 1ms or less duration. Just not enough difference from 12 volts to spiked 20+ volts to see the #2 Lamp light.

humm, Lamps used are 12 volt 20W.
Ken

Hi ken,

Sounds like the relay is malfunctioning or is not connected correctly???

Best regards
Spark

Quote from: kcarrigan on March 04, 2007, 10:25:07 PM

humm, Lamps used are 12 volt 20W.

Hi Ken,

Could you use lamps with lower power like 2-3W? Seems they are too low resistance for the current involved in your circuit, hence probably this is the reason why L2 remains dim. I mean your changing the lamps in your first shown circuit in which L1 slightly glows.

Maybe using 6V lamps also helps (also with types of 2-3W of power first).

You could try increasing NE555 frequency up to 70-80Hz but this depends on whether the relay is able to keep up with the speed...

rgds
Gyula

Hi,
I've checked the connections, did resistance checks on the relays, all hooked up correctly. Tried 100 ohm loads instead of bulbs and the 1st resistor gets hot, the other one cold. This was with inductance shorted.

I can not see how this switching of capacitors from series to parallel and from parellel to series - how it could produce more energy in the second bulb/resistor. once the first set of series caps are charged to 12 volts, and the parallel one are charged to 12 volts; it switches - which would make the new parallel caps now at 6 volts, and the new series caps at 24 volts. Only the 24 volts has 1/4 the energy to what the paralled caps had, (1/2CV^2) and needs more power drawn from the 12 volts DC input to charge up the 6 volt paralleled caps.

I've tried the squarewaves up to 70Hz, and think with a diode across the relay (dumping the back emf - relay chatter) I could make it go higher, except that just make the first bulb glow brighter.

Any thoughts?
Ken

I used 27.9=~28 mH coil. and some times increased voltage 12V to 13-15 V.
bulb light doesnt matter so much, did you get battery current, capasitor voltage?
For self running, self powering operation adding 3. coil to toroid core and with rectif. diods feedback power to capasitors(except relays). I tried but unfortunately wasnt work.

and my friend tested circuit(includes irf510 mosfet) with sim.multisim-7, result :
Volt R2 - 1 Ohms : 12 uVolts..

Hi People,

I got my old (real old) scope working again!
And a new relay for the circuit!
The new relay (radio shack 275-0218) works much better and there is no arcing
(untill I added inductance but thats for another post...)

The fet version is built BUT the monovibrator does not give anything like square waves, maybe I toasted a transister during construction, need help with 50/50 duty driver for the fets, a 555 is difficult to maintain even duty cycle, any ideas welcome!

Notice how close the sim is to reality!!!

Spark
p:s:
the edit on the scope photo top voltage is 25 not 30
also notice that it is multiplying the voltage by about 1.5 average!

Maybe something like: http://www.geocities.com/tjacodesign/motor/motor.html

Can use inverters instead of nand gates.

hi, great postings all.

can anyone tell me how to get a copy of circuitmaker student demo, as they no longer support it acording to the webb site?

regards

dorro

Hi People,

Multivibrator circuit had transister backwards, you can see in photo.

FlyBack effect from audio transformer at R2 point in recycler circuit caused disterbance here!...

I gathered 8 AA nicads all dead and charged them 9.99v, then with r2 HOT10ohm1/4w, running circuit 12h I discharged batterys to 1.1v, then swapped r2 for audio t1 36ohm winding and was trying to get lowvoltage picture of flyback pulse w/o disterbance to cammera when i notice pulse/spike waveform peek voltage increasing! allmost faster than i can get photo.

I decide to let it continue, no input from 555relay circuit only drive coil, seperate grounds and sources, still battery voltage climbs to 4.7+v after about 8hours
and it still now veryslowly climing and the caps are swapping thru the transformer and the spikes are still being made... i give it till i wake up then report hours and final volts, then onto the fet circuit and higher frequencys...

drSpark

Hi,Dorro1971, ;)

you can find circuitmaker student version onhttp://www.archive.org (http://www.archive.org):a very usefull site:  add the address http://www.circuitmaker.com in the waybackmachine

Terry M.

Hi People,

The next morning its 4.7v  so i think is memory effect of battery(s).

Last night I remove transformer, put 22uf cap where battery was, start 3:00am at 7.82v
2:15pm same day 3.99v

Now all thats happening is the c1-c2 caps are changing state from series to parallel at 10hz accross the 22uf cap and the d1 diode that are in series....

The sim shows 22uf cap voltage drop <100mv in 30seconds.

reallity much much different hours have passed and it still has half voltage!?

Im going to get a wall wart to run the 555circuit and set this (relay far) away and let it run....

on to the quiet FET VERSION...

Spark

cheers Terry,

waybackmachine!!!! should have thought of that..DOH!


dorro

Hi dorro1971 Terry,

There is a egroup that has the servicepack1 (major upgrade) for CM2000 and other infos
http://tech.groups.yahoo.com/group/muug/?yguid=204074595
muug : CircuitMaker / TraxMaker User group
I hope this helps
for googling try;
search token / size of file, if found.
cm2000servicepack1.exe 2mb+-
cm2000.zip 10mb+-
cmstudnt.exe 3mb+-

drSpark

Hi People,

I have the solid state version of the recycler running
heres a quick look (needs bigger caps) and i forgot the diode.

R1=100ohms R2=6vbulb run circuit bulb glows dim yellow, put bulb in series with 100ohm and power supply and bulb does not glow at all...

drSpark

Hi DrSpark !

Can you make some scope shots on Q4s Gate-Source and Q4 Drain-Source ?
I don't think it swtches properly... but none the less, again very close and very efficient.

Igor

Hi DRSpark,

I included a slightly modified schematics to improve the switches performance.
It involves the addition of 3 diodes and 2 resistors. The diodes preferably are Germanium or Schottky types to reduce forward power loss but for a test the Si 1N4001-1N4007 types would serve. The resistors help to make conducting path for the Gate-Source control voltage, making switch on / off to happen more easier. The resistors also dissipate a little power. I did not build this circuit yet.

What is the cold resistance of your 6V bulb?  Or how many Watts is its power?

Also looking forward to see the scope shots Igor asked for.

Regards
Gyula

Hi Spark,
well done.
Could you pleae post scopeshots directly across R1 and also across
the Lightbulb?

I want to see the voltage ( and thus the current) in these devices.
Many thanks.

Regards, Stefan.

Hi Spark,
what happens, if you put a bigger choke inductance in series with the L1 bulb ?

Or what happened with your transformer version ?
Why do you leave it out now ?
Many thanks.

Regards, Stefan.

Hi People,

Were to start?
Turned on scope, it will be an hour or more stablizing (old tektronix Type 502)....

Stephan, work on the transformer version is as soon as the zeners and transformer I ordered last night, arrive. After the fets did not burn up and the circuit function like relay circuit, I decide to get the rest of the parts. Also ordered 4more 4700uf to replace the 220 and 270uf caps. These new ones dont want to leak like the old ones :)

Gyula, great! I will scout around for the diodes and resisters will be easy to make that change. But first I want to get to know this circuit so I can (get the scope shots) and a good feel for how this behaves, then I will be able to tell the differance with the changes you sugjest.

People, it ran all night and is still running start bat 11.3v now almost 12hours later 10.03v

6volt bulb  same as the ones from last month. 1.6 - 1.7 ohms cold.

Still no traces on scope....   it needs time ill be back later with the scope photos...
Im going to switch to computer type powersupply and leave the batterys for allnight runs untill they discharge........

Spark

Hi Spark,
hopefully your scope will work again.
Without a scope such work can?t be analyzed exactly.

Looking forward to the upcoming scopeshots and good
luck on your battery tests.-
Many thanks.

Regards, Stefan.

Stephan

scope is working, im cropping and anotating photos now  i left the battery on it
few more mins  one done 3 to go...

Spark

Hi Stephan.

Here are the shots of the R1 resister and the R2 light bulb
working on other fet point shots next....

Spark

Hi Gyula,

The cold resistance of the bulb is 1.5 - 1.6 ohms.

Spark

Hi Igor, Gyula,

Sorry some of these shots are upside down.
BUT it was easyer to clip the gnd aligator to the drain and poke around with probe.

Still on battery so no ground loops same for all shots today.

battery now 9.88v on dvm filiment still hot, but not as bright as last night of course.

The new larger caps will not discharge like the current bunch, that was visable on the scope/relay version of this circuit

Q4 FET scope shots:

Spark

Hi people,

I added the blocking diode and now i can tune the monovibrator.

New test;

Mains driven 10dcv power supply,
R1=10ohms 1/4w,
R2=10ohms 1/4w,
multimeter

whole circuit input  (no r1, draws 12ma@10v)
9.93V
72.4mA
0.072W

across R2  (0vdc must use acmode.  I know...)
1.95V across R2
0.2A   in series with R2
0.39w

multimeter COP 53 to 1

Finger test you can touch R1 to your lip,it is warm, I canT feel it with my fingers.
R2 will burn your finger! again the burning resister effect like with relay.
(disclaimer: Touching R2 to your lip will result in serious burn and personal injury)
Is R2 53 times as hot as R1??? probably not, BUT its more than 3 times as hot!
way more IMHO!

Spark

Hi drspark,
if input current is 72.4 mA = 0,0724A then
power =9.93 x 0,0724 = 0,718 Watt ...

or 10V x 0,072 ~=0,72 watt ..

is there math. fault or current value different?

Hi Gezgin,

Milli = 1000;  I calc  (9.93V * 0.00724A)=0.072W
IS WRONG you are correct
53 sure did seem odd.

drSpark

Hi Gezgin, Members,

I suppose I asked for that one just by using the dvm :D

100ohms in series with bulb and battery = no light AND 100ohms in series with battery to cap swapping circuit lights bulb

Same with 2 ten ohm resisters one is burning hot the other is not.

Do those observations show sign of overunity?
or am I wasting our time with this?

drSpark

whatis Lamp  power? 5W?

in circuit,serial 220 uF C5 C6 capasitors charging&increase voltage 10 to 20 v,
so 6 V lamp lighting.

fe theory, relay spark effect  and with special coils (bifiliar ..)
resonances (LC) maybe create extra energy..

Hi Gezgin,

Lamp 6v stamped T55 6F. cold resistance 1.55ohms.

drSpark

Hi Gezgin, Gyula, Members,

Lastnight I made some measurements of the bulb hot.

Battery 6v lantern type.
Open voltage 6.32v
Bulb on voltage 5.86v
Series current .39A
DVM straight DC  :)
I guess its rated 2.5w @ 6v

drSpark

Hi People

Here is a photo of the circuit, bulbs with 10ohm resisters in parallel.
The resister does not burn connected this way.

9.9v 324hz

drSpark

Hi drspark,
what about your self running circuit (emf recycler 3.0) with transformer&mosfets.
I tried (with relays) but didnt work.
scop show coil emf picks but when connecting load its disappearing.
Here is gnd to capasitor wave photo:

Hi Dr. Spark,
sorry for the late reply, but I was busy with other work.

It seems from your  scope shot IMG_5942a.jpg picture,
that if you fold the spikes down to the waveform,
that you have about 4 Volts steady of voltage at R1= 100 Ohm,so
that means an input current of about 40 mA.
That means at a voltage supply of about 10 Volts you
have an input power of around 0.4 Watts.

Now the question is , is the bulb glowing brighter than 0.24 Watts ?
Already at the R1= 100 Ohm resistor you "destroy" about U^2/100 Ohm= 0.16 Watts,
so you have left around 0.4 Watts - 0.16 Watts= 0.24 Watts for the bulb).
(Neglecting the other circuit components,
which have much lower currents..)

From the Scopeshot at the bulb it is hard to say.
I would say, if you fold again the waveform, so it
shows a steady line, this line would probably only have around 1 Volts.
Now the question is, what exact resistance your glowing bulb has...

You should better test this with a normal temperature stable
resistor, as the bulb probably has a resistance of about 10 Ohms or more,
if it is glowing.

Maybe you can do a comparison and test with a DC power supply the same
bulb at the same brightness and measure voltage and current and
tell us, how many Watts does it take to glow this bright as in the scope shots you posted.
Also please measure the dc resistance in this operating point in the DC case of your
bulb filament.

So if the bulb would have 10 Ohms resistance, if it is glwoing like this,
then you would have about (1Volt) ^2 / 10 Ohms= 0.1 Watts of power
generated at the bulb,
but if it only would be 2 Ohms, then you would get (1Volt) ^2 / 2 Ohms= 0.5 Watts,
then you would have Overunity.

Regards, Stefan.

Hi Dr. Spark,
please repeat the scope measurements with
2 normal resistors instead of resistor and the bulb:
R1= 100 Ohm 1 Watt type or more
R2= 1 Ohm   5 Watt type or more
and post the scopeshots again at these resistors.

Then we can exactly say, if your circuit is overunity.

Many thanks.

Quote from: drspark on March 21, 2007, 05:16:45 AM
Hi Gezgin, Gyula, Members,

Lastnight I made some measurements of the bulb hot.

Battery 6v lantern type.
Open voltage 6.32v
Bulb on voltage 5.86v
Series current .39A
DVM straight DC  :)
I guess its rated 2.5w @ 6v

drSpark

Okay, now at 5.86 Volts / 0.39 Amps= about 15 Ohms
resistance of hot glowing filament of the bulb.

Now please try to put a series resistor ( potentiometer)
in series with your 6 Volts latern battery
and bring it to the same brightness as your  bulb in the emf recycler circuit.
Now then calculate again the bulb filament resistance.
R= DC Voltage at bulb / current in bulb

Or better just use as I pointed out before 2 resistors ( 100 and 1 Ohm) and drop the bulbs
for the measurements.

Hi People,

Gezgin,
Transformer version next week....

Stephan,
scope is warming up, it takes about 20mins for beams to find the zero or you chase zero around for 20 mins
ill dig for LARGE resisters that i can make 10 and 1 ohm resisters from.

drSpark

Hi Stephan,

scope shots

drSpark

Dr. Spark,
where is the 0 Volts line of the lower trace ?

There very down there where you wrote 0V  2V/ div ?

Hi people,

I have 4-1ohm 10w and 2-8ohm 20w large resisters.
Can arrange into 2 10ohm resisters for voltage with 1ohm or 2ohm for current measurement.

Would they help make better scope shots?

Yes Stephan at bottom just below 0v.

drSpark

Okay, now the result.

The input is about 10 Volts x 80 mA= 0.8 Watts.

The 80 mA can be seen, by folding the square wave
ontop the DC level to a steady value, so the voltage
gives about 8 Volts, if you integrate over all areas
of the cycles..
8 Volts / 100 Ohm is 80 mA.
You already heat via the input at the 100 Ohm Resistor
a Wattage of (8 Volts )^2 / 100 Ohm = 0.64 Watts.


Now output power at R2= 1 Ohm is
about (0.4 Volts)^2 / 1 Ohm= 0.16 Watts,
when you fold again all the areas to a steady DC level.

So the output is much lower than the input.

You can try different resistor values,
but without a transformer it would probably
always be output < input.

Regards, Stefan.

Dr. Spark,
just try it again the other way around:
R1= 1 Ohm
and
R2= 100 Ohm.

This way it will make more sense,
as the 1 Ohm at R1 will only be a shunt to measure the
input current and not all input energy  is already heated
there.

Regards, Stefan.

Thanks Stephan,

The bulb thru the 100ohms and the 2 10 ohm resisters one burning hot, sure had me fooled!

Ill still try the transformer circuit. Soon as parts arrive...

drspark

Stephan

How does this explain the burning resister?

drSpark

Quote from: drspark on March 21, 2007, 05:39:48 PM
Stephan

How does this explain the burning resister?

drSpark

You seem to have used 0.25 Watts types resistors.
If you have used more power in one resistor,
it surely gets hot...

Stephan,

both resisters are 10ohm 1/4W r1 cool r2 HOT


I am puzzled still  by only that

drSpark

Hello,
Back to your first circuit board (without inductors and replaced bulbs with 10 ohm 2 watt resistors). Measured c1 and c2, c3 and c4 in steady state. C1 & C2 are paralleled, C3 & C4 are in series. OK, the Paralled caps are 11.59 volts. C3 is connected to ground measures 3.51 volts and C4 measures 8.08 volts. Humm.  Turned on circuit switching.. measures the same. Placed a OSCOPE on caps. C1 lead always connects to ground, measures NO Chopped waves at all. Of course C2 did by about 5 volts peak to peak. What bothers me is that C4 is always lower then C3. Actually if left for a while, C3 will turn negative while C3 is positive - adding them up gives paralleled c1 & c2 voltage.

SO... maybe the inductance 'peaked' or distroyed my 50 volt caps...  going to order new ones, maybe about 8 this time, and keep the resistors in the circuit with no inductance.

Very strange... never seen a cap actually reverse it's self while slowly discharging.
v/r
Ken

Hi People,

Stephan,
I made two of the 8ohm+2-1ohm 10ohm resisters, and somemore scope shots.

drSpark

Hi People,

Here is a photo...

spark

Okay, Dr. Spark,
now with the diode and the 2 x 10 Ohm resistors,
things have changed.

You now have around 1.2 Volts at 10 Ohms= 0.12 Amps
input current.
That means you have around 10 Volts x 0.12 Amps= 1.2 Watts of input
from your battery.
Now you burning up at R1: (1.2 Volts) ^2 / 10 Ohms= 0.144 Watts of heat power.

Now at the output you have , when we fold and integrate the waveform
around 2.5 Volts at 10 Ohms, that means we get a heat power there of
(2.5 Volts) ^2 / 10 Ohm= 0.625 Watts.

So now you see, why one resistor burned almost up,
when you used just only 0.25 Watts types.
The R1 only comsumes 0.144 Watts , thus it is just lightly warm
and is lower than the rated 0.25 Watts.

But R2 burns almost up, cause it gets 0.625 Watts, which is more than twice
the rated 0.25 Watts...

But all in all the input power is much higher than the output power.

Please try it again with
R1= 1 Ohm and
R2= 100 Ohm.

Many thanks.

Regards, Stefan.

Dr. Spark,
as long as you don?t use any transformer and don?t use mechanical switching
with the right contact materials, I guess we will never see any additional energy
coming into this circuit.

It is just a capacitive current transformer type circuit and we need
external energy input via electron clustering at the mechanical switching contact points
or something simular to generate additional energy from the outside of the circuit
to get additional energy input.

I am looking forward to your new tests with the transformer.

Regards, Stefan.

Hi Stephan,

Just to complete the record I performed the 1 to 100 ohm measurements you sugjested.
freq 307hz

The parts arrived today, transformer is not an audio freq transformer. is standard mains 50/60hz unit.

That yellow torroid any idea how many feet of #24 i should wrap on it?

Okay, this is also not overunity.

Now if we fold and integrate again the waveform,
we get about 25milliVolts atthe R1= 1Ohm shunt.
This means 25 milliAmps flow into the circuit, which means
an input of 10 Volts x 0.025A= 0.25 Watts.

The R2=100 Ohm resistor gets a power of about
(3.6Volts) ^2 / 100 Ohm = 0.13 Watts, so the output at this resistor is
still lower than the input of 0.25 Watts.

This way, it just will always work only in underunity mode.

Regards, Stefan.
P.S. If you wind a transformer,
try to go first with so many windings, so that you have around
10 Ohms DC resistance of the coils.

Hi guys,
New caps on order. Thought you'd want a diagram of the 50:50 duty cycle timer circuit I use. Instead of using transistors and components, use this circuit.  The output of course will need something like a MOSFET or biploar transistor to keep the 'Ra' resistor isolated from any output loading affects.
v/r
Ken

Yes, this site about 555 osc. circuits:
http://www.doctronics.co.uk/555.htm

Hi all !
maybe we can switch things this way.

at first only C1 is charged, C2 is empty.

At this point (if K1 is closed at O1,O2) current through s1(trafo or coil)
starts to flow and induces voltage which in return rises C1+ potential and lowers C2+ potential making it possible for C1 to additionaly discharge to C2.

can you please answer:
1. I expect that voltages will end up (obseving right and left branch) at level simmilar or same as when using four caps ? will they ? or only discharge current is much stronger but making it imposible to reroute C2 charging to C1 ?

2. Is there already any place to place load in this schematic ?

None the less I asume that if trafo is replaced with two coils (on one core) will have I think very strong magnetic pulse which can be utilised to push some magnets on rotor... not putting to much to input.

just a specualtion, but should be tried.

Igor

Hi People,

New Caps and board.
finer frequency and duty control.
Plenty of room for Gyula's diodes&resisters,
and adding the other 3fets to ballance the circuit.

I am ready for a transformer now...
anyone have a hifrequency transformer pt# Mouser.com I should try

Motor application, interesting, gray(ish)

Stephan, check out new scope shot, reintegrate...

drSpark

Hi all !

Please reconsider using schematic in my previous post:
advantages:
- much simpler (only two switches)
- no loss in charge potential(U) (unlike previous paralell conection where charges split, and capacity(charge number) splits when in serial - in return demanding current refill)

This is exactly what I have tried to achieve. Two Cups (one full, other empty). The thing is how
to lift one cup over another without losses (or small ones)...Here full Cup lifts itself (with trafo) and lowers Cup that is refiling. Full charge and even its potential is flowing from one cup to another over and over again...is it cheap ? is it maybe free?

Please brainstorm this schematic...haven't tried it (many things can happen here that are not obvious) but must soon.

Hi Igor,

Nice!
Sim OU first try!
Need to think about this one....
I think I have all parts but the board.

drSpark

Hi People,

Here is my best simulation of Igors latest.
Lets call it *bouncer*?

HELP! I have that small torroid core from pc powersuply and #24 magnet wire
HOW CAN I MAKE A SUITABLE TRANSFORMER FROM THOSE THINGS?

NOT GOING TO PUT 10-OHMS OF IT ON HERE >300 FEET OF WIRE  ;D

DRsPARK

Quote from: drspark on March 27, 2007, 02:15:37 AM
Hi People,

New Caps and board.
finer frequency and duty control.
Plenty of room for Gyula's diodes&resisters,
and adding the other 3fets to ballance the circuit.

I am ready for a transformer now...
anyone have a hifrequency transformer pt# Mouser.com I should try

Motor application, interesting, gray(ish)

Stephan, check out new scope shot, reintegrate...

drSpark

Hi Dr. Spark,
after folding the waveforms and integrating,
it does not show overunity.
Input average voltage at 10 Ohm resistor about 1.2 Volts,
means 0.12 amps in and 10 Volts means 1.2 Watts input.

Average voltage at Loadresistor R2= 10 Ohm = about 3 Volts.

Power at this resistor thus:
Voltage ^2 / R = 0.9 Watts.
So you have now to try with the transformer.

Just try to wind first with low turn numbers and thick diameter wire,
maybe just 10 turns, so the ohmic resistance is low and you can switch with
faster frequencies !

Good luck !

P.S: Nice new circuit from Igor too.

Regards, Stefan.

Hi all !

Made sim.
For pits sake look at this wonderfull 120Wpeaks at Rload.

input is allmost zeroed.
battery is really not needed, and if needed it is recharging itself !!!
No need for sparks.

situation stays same after 100 seconds of sim (capacitors doesn't deplete a bit)
simmulation is stable(tried with different values) and gives correct wave forms unlike in parametric parallel/serial switching.
so I think that this really works

This is so simple it can be made motionless. Have we finally found a holly Grail ?

Harti please make aditional link to this Igors switcher topic at www.overunity.com

hope this works. It can be made light, small, and powerfull...works in space

Igor

Quote from: drspark on March 27, 2007, 02:15:37 AM

...I am ready for a transformer now...
anyone have a hifrequency transformer pt# Mouser.com I should try
Motor application, interesting, gray(ish)

Hi DrSpark,

Unfortunately, there are not much choice  to choose from.

Perhaps these two types will be worth trying, but they would need a switcher frequency of higher than 15-20kHz.

mouser part # 673-PE-63385 1:1 pulse transformer $6.14 one qty
                 # 673-PE-63387 1:1:1 pulse transformer $10.62 one qty

These are the prices for someone not willing to trouble/wind many feets of enameled wire onto a toroid core...  :D

To get as high as 15-20kHz from your multivibrator, you have to change the feedback C values for a lower value, or you may like to use a CMOS version of 555 timer IC like TS555CN (part #511-TS555CN, $0.39 one qty) and in this case you can easily tune the switching frequency by a single potenciometer (and duty cycle with another one, independently). If you wish schematics on this circuit later I can upload.

Gyula

Quote from: hartiberlin on March 28, 2007, 05:23:00 AM

Hi Dr. Spark,
after folding the waveforms and integrating,
it does not show overunity.
Input average voltage at 10 Ohm resistor about 1.2 Volts,
means 0.12 amps in and 10 Volts means 1.2 Watts input.

Average voltage at Loadresistor R2= 10 Ohm = about 3 Volts.

Power at this resistor thus:
Voltage ^2 / R = 0.9 Watts.
So you have now to try with the transformer.

Just try to wind first with low turn numbers and thick diameter wire,
maybe just 10 turns, so the ohmic resistance is low and you can switch with
faster frequencies !

Good luck !

P.S: Nice new circuit from Igor too.

Regards, Stefan.

@ DrSpark, Gyulasun, Harti

After conducting many tests, changing parameters of components, etc. I can confirm
that first measurements of EMF Recycler (http://www.overunity.com/index.php?action=dlattach;topic=133.0;attach=6634;image) is not showing FE.
Gyulasun pointed out in his previous posting that resistor (84.3mA) for measuring current at the output is missplaced before C5 and that is exact cause of this false mesurement.
Nevertheless when I put resistor in series with only output lamp (means after C5) I get much better results (O/I Watt ratio) than DrSpark and it is most times near unity 90 - 98 % . The
real difference in my setup is that two currents are flowing through trafo (which is actually Rload). One is from preserved charge in caps and the other additional from input.
I suggest to DrSpark to make small adjustment and use two Rloads, and put input inbetween.
this way O/I ratio will increase.

What do you think of this last Only two caps schematic and theory behind it. I think it's amazing
and would like some speculative confirmation or denial, scince I have no time to make it today, but weekend is near...
It is easyer (allmost free) to "lift" (potential) allmost massless bunch of electrons, than lift full cup of watter for purpose of filling the other empty cup.
Think that massless electrons (a lot of them) can do work of say 10KWs and compare how much mass of water we have to lift and at what height (spend work) to observe same ammount of work. "Cup" filled with electrons is easily lifted, it is massless...
Hi Hopes this works.

best regards

Igor Knitel
Perihelion Labs

Hi Igor, Members,

The battle of the sims...

All in fun, you never know, till you build em...

drspark

Hi Gyulasun,

I like the 555 circuit but, we need the inverted signal and that meens another pnp transister anyway so I go for dual combined 2 transister oscillator and be done. No problem changing resisters and or the caps.

Thanks for shopping for the transformer! I cant get my little finger thru the core I have let alone 2 300+foot strands of #24 and im lazy about reinventing the wheel. AND its a new month (real soon)  I will order the centertappeed one.

vbr
drspark

Hi, guys!

Sims, probably, have a bug
Take "Transformers -> Coupled Inductor" in place of "Transformers -> Subcircuit" and you will be surprised - OU will not be!  :'(

Vlad

Hi Vlad,

You are correct, the T1 on free1kw.jpg is magic DC transformer! (does not exist yet)
I think thats what pese was trying to say to me a month ago.

vbr
drspark

Hi Igor,
what do you know about (Alexander Kochurov's) akoil generator& its russian patent
its special transformer or another energy genator_?

http://peswiki.com/index.php/Directory:AKOIL_Power_Generators says:
Russian company claims to have a fuel-less generator. The technology consists of a self-running special ?transformator?.

in photo 3 caps and 2 transformer
is there any similarity your design with akoil gen.?

Hi People,

Here is the 3fet circuit with R2 replaced by audio frequency transformer.
scope shots; open secondary and shorted secondary.

Next scope photo of the output side of the transformer,
is 20vpp with 10ohm load... and is square???

drSpark

Hi

Scope shot of loaded audio transformer.

spark

Hi Dr.Spark,
many thanks for
the new postings.
But we need to know the ohmic resistance of the transformer
primary and secondary windings, so to say anything at all.

Also in the test with the 10 Ohm loaded secondary transformer you forgot
to test the input current.

Please let us know.
Many thanks.

Regards, Stefan.
P.S: The first posted scopeshot had an input power of about 0.2 Watts.
( The one with the open scondary transformer winding)

Quote from: drspark on April 05, 2007, 03:39:29 PM
Hi

Scope shot of loaded audio transformer.

spark

This looks like it has around 40 Watts output at the 10 Ohms resistor !
around 20 Volts folded, when integrated so Voltage^2 / 10 Ohm= 40 Watts.

What wattage type resistor do you use there ?
Does it get warm ?
Can you please show a scopeshot of the input current at the other input 10 Ohm resistor ?
Many thanks.

Dr.Spark,
maybe you can verify with 2 x 12 Volts / 20 Watts  or higher type halogen lamp in series
instead of the 10 Ohm resistor,
if your last circuit really puts out around 40 Watts ?
So do these bulbs shine bright ?

What is then the input current scope shot  at the other 10 ohm resistor ?

Many thanks.

Hi People

The transformer;
33.6ohms
2.4ohms
hair fine wire could never pass 40W
resister is brown black (real dark blue) LOL sorry ill remeasure later...

Working on a better transformer.... with 1 to 1 turn ratio and centertap.
Soon as i got the driver to work >10khz, I tryed that tiny transformer.
The 36ohm winding has no DC effect on the ping pong action (like its not there to hifreq)

Drspark

To all !

This is the frist time I hear bout AKOIL.
Any schematics  ? is it the same as mine.

I know that schematic is so simple, and one can't believe I have no time to try it...
Well I'm preparing for gigantic jump to future with more certanty or rough downfall to the present chaotic oil mud.
so i must allso have time to concentrate...one of these days certanly...
Havent tought of any "obstacles" that this wouldn't work and give FE, a lot of it.

anyway a very nice progress...

Igor

Quote from: drspark on April 05, 2007, 08:50:56 PM
Hi People

The transformer;
33.6ohms
2.4ohms
hair fine wire could never pass 40W

Power transfer in a transformer does not depend so much on wire diameter size..

Quote

resister is brown black (real dark blue) LOL sorry ill remeasure later...

So the output resistor is not 10 Ohm or melted up and you can?t see the
rings anymore ??

I guess the value of this resistor is much higher,
otherwise your secondary with 33.6 Ohms would have had even more Watts
on there ! And then the transformer would have burned up with
its small diameter wire size..

Regards, Stefan.

Igor, Members,

I am sorry if it seemed like I was negitive regarding your latest circuit, I most certanly am not.
I have collected all the parts required, Except one, the transformer.

For your circuit to work we have to use relay and low frequency transformer 1 to 1 turns ratio, thats the brick wall for me. I/We need a resonate transformer for 20-30hz max operating range of relay. I know its not impossable, but thats the limit of my knowledge.

I have a spool of #24 and that torroid,  the torroid currently has 2wires 3ft long, one centertapped, anyone know what that frequency is?

For the recycler I need a transformer for 10Khz that can take about 5amps.
(or how many double #24 turns on a AM radio ferrite stick. I have one of those also)

Pese, can you help us design 30hz and 10kz centertapped 1 to 1 transformers capable of amps, I really want to get to the bottom of this as do others im sure.

I see a transformer as two gears with many teeth and a clutch or *hydralic torque converter*, go to slow and the clutch slips, go to fast and the clutch slips, when just right the gear ratios transfer power. My analogy is broken when we introduce resonance.

With the caps raising the current, a little bit of resonate voltage rise will make up for the voltage losses at componates and loads, keep fingers crossed I am.

------------------------------------------

I do not understand the need to bring more electrons into the circuit, ALL the electrons in the circuit are *in the circuit* allready. We need a way to make them move, a charge seperating dipole (battery) or a disterbance. If we kept *adding* electrons to the circuit it would ionize.

Whats needed is ElectroMotiveForce to give motion to the current Druid gas electrons in the metals of the circuit NOT more electrons (mass).

Why speed of EMF (c) not speed of magnet passing coil in a operating generator?
There are many things about electric not understood - that hasent stopped us from harnessing it.


drspark

Stephan,

With the audio transformer bridged with a real 10ohm not 10meg ohm resister the output looks like the shorted t1 scope shot. Im sorry as I climbed the stairs I was wondering why the resister wasent hot. I kind of expected the same heat. Neither way would light a 6vbulb.

I think because the output flatlines the core is getting saturated?...

The waves look basicly the same no matter which side of the transformer I plug in, when the other is shorted.

The audio trasformer is labeled TSP-1036

spark

Hi People,

I have greater voltage and current now using 2-1 transformer.
The core makes hi-frequency sound when all is just right...
I will be making up scope shots after i play with other size wire(s)...
The bulb is the brightest ever and a real 10ohm smokes quickly!

:)

drspark

Dr.Spark,
looks good.
Yes, don?t use too many windings on the core.
Then you can go to higher frequencies and still have
a good and fast current rise in the coils
and a good power transfer.
Looking forward to the scopeshots.

Regards, Stefan.

Hi People,

Scope shots of the circuit running in photo
FETs get hot now, from the top down Q3 real hot Q4 hot Q5 warm,
adding 50k like Gyula sugjestion helped but did not fix

drspark

Igor

On paper/drawing/sim, Convert your circuit to 4 or more FETs and I will build it.
Must be solid state to make 20khz and beyond.

drspark

Hi People,

I have moved the R1 measurement point back to a series resister with the source. A DC value, 0.4v across 1ohm is observed at 9.95v.

An observation that I dont understand:
I have the wire from the D1 that feeds power to the cap bank top C1, in hand. When I touch it to the top C1+ the circuit charge and run, BUT when I disconnect it, the circuit surge in output before it discharge and die, when I touch it to the top of C3+ cap the circuit charges and runs... and when I disconnect the circuit discharge no surge and dies.

This is behaviour inline with original reed relay circuit, where power was only connected every so many cycles, for one cycle.....

No idea what this is doing to the FETS.... Tapping the wire as fast as I can results in a erratic mostly continious surge!

A counter chip can controle a relay to do that to this once every 5000 to 10000 cycles for x cycles :)
(surge lasted 1/4-1/2 second)

drspark

I have done your "EMF recycler V3", but what is ?.ic? ?

Hi mhackgyver,

The .ic tags look like sources but really are Inital condition setings.
Used in the monovibrator to help the sim simulate an oscilator.
also used to precharge the cap in the selfrunning sim circuit.

drspark

Hi Dr. Spark,
at the scopeshot:
3fett1_10ohma.jpg

you get about (4 Volts)^2 / 10 Ohms= 1.6 Watts of output power.

But what was the input power at R1 in this case ?

Many thanks.
Regards, Stefan.

I've build that:

(https://overunityarchives.com/proxy.php?request=http%3A%2F%2Fmhackgyver.free.fr%2F10.jpg&hash=3b71d8c61922bdf517d9ad565113135bf775dd92)

and the result:

R1:
(https://overunityarchives.com/proxy.php?request=http%3A%2F%2Fmhackgyver.free.fr%2FR1.jpg&hash=be5f5d9d6950405c47bd2bdb2b4eaf0cbca90733)

how that can be possible after the diode?


R2:
(https://overunityarchives.com/proxy.php?request=http%3A%2F%2Fmhackgyver.free.fr%2FR2.jpg&hash=3e016ea02e0520685b5373a2a9693d2597c4ed2b)

@mhackgyver

probably your diode does not work anymore.
YOu have about 0.12 Ampere average input current, thus at 10 Volts
1.2 Watts input.
Your output at R2 is about (3.5 Volts)^2 / 10 Ohms= 1.225 Watts,
so the graphical interpolation could be a bit less or more, so
it seems what you put in is also
coming out at R2, so this is right.

Hmm,
what I just wrote is not right...
Where exactly is the zero line of the input voltage at R1 ?

If it is in the center of the square waveform,
then when the current is negative,
we could have a charging of the battery,
if the other multivibrator circuit is not broken...
Strange,
please check the diode and the rest of the circuit.
Thanks.

mhackgyver,

Great Work!
you replicate also!

Your R2 should be >4x temp compared to your R1.
Both are circuit outputs...

drspark

Hi People,

Here the sim calculates COP, power across R1 and R2.

The outputs A and C on 3fetcopraw.jpg are what you see on a scope.

Notice circuit change, right side of R1 connects to right side of R2.

drspark

Quote from: drspark on April 21, 2007, 04:41:40 AM
Hi People,

Here the sim calculates COP, power across R1 and R2.

The outputs A and C on 3fetcopraw.jpg are what you see on a scope.

Notice circuit change, right side of R1 connects to right side of R2.

drspark

Dr. Spark,

do you have also a graphical output file
from your CKT text simulation file ?

I don?t understand your new setup...

Why don?t you use a transformer ?
Why the change with the connection of R1 and R2 ?

Please explain more.

Many thanks.

Regards,Stefan.

ok

Hi Dr.Spark,
I don?t understand your green waveform...
Does it go with a spike up and then reverses and go lower ?
If I try to follow the green line it is hard to see, when it comes up,
where it exactly goes !?

Also what does your red line mean ?
Is this the output ?

What is the output power versus input power ?

Did you try to take new scopeshots so one could
see this ?
With your last scopeshots you did not show the
input current scopeshots, so one could
not calculate the input power.

Regards, Stefan.

Hi Spark,
can you please explain a bit the new circuit and the graphs
posted in your last RAR file ?
Many thanks.

not eng.

Quote from: vala on May 13, 2007, 12:41:42 PM
not eng.

Not in English or what should that mean ???

DobÃ...â,,¢e, skus to otestovat.

Hi Vala, People,

Here is Vala's circuit converted to picture.
Interesting fet driver

VBR
drspark

Thank you.

Quote from: vala on May 13, 2007, 12:41:42 PM
not eng.

I think, that vala is trying to control FETs with impulse trafos which I think is good approach
scince we don't have PNP optocouplers in our shops to make discrete solid state relays and be sure when actually our fet is closed or opened. So this is the next best thing before we manage to buy some real solid states (no triacs or SCRs at output).

His schematic approach is novel in that manner that he uses parallel-serial,serial-parallel recombination of caps but allso uses part of my last schematic that helps to discharge serial combination much more (and in same time preserving dipole from otput) than just recombination would allow.

I wellcom this approach but there is one additional loss when charge is splited to paralel capacitor combination it losses it's potential (V) and much of energy is lost that way so I wold stick to complete detail in my last schematic.

goosh I don't have time to make this simple setup.

Igor

Quote from: vala on May 13, 2007, 01:21:14 PM
DobÃ...â,,¢e, skus to otestovat.

OK, now we have to test it !

If building uppon schematic with only two caps (the one I recently proposed) it is advised to use
unipolar capacitors (sadly known unipolar caps have small capacities).electrolites will blow.

Are supercaps unipolar or bipolar caps ?

Igor

l am sorry at present have not time.

Capacitor RAD BIP 10/100 RM3.5  (www.ges.cz)

It is whim synchronic resonance.
Heed not regulation (R3= 10K - 1K) max 60V
R1 = (In) or out energy.
Start charge T1 (primary - sekundar)= coil capacitor
Deepr analysis next week.

Sorry not englich.

vala

This looks quite promising!Ã,  How are things going for all replicating experimenters on this thread?

What brought me here was investigating what appears to be an extremely basic version of what the OP is working with.Ã,  The experiment was very crude, starting with manual switching sequences.Ã,  I added a microcontroller(PIC16F628A) to drive some Photo-MOS relays(653-G3VM-352C) and so converted it to solid-state.

I'm working with very low capacitance(~100uF aluminum electrolytic) and so far nothing worth mentioning has come of it..Ã,  My desire is to see a compact device capable of running a couple of LEDs or various microcontroller applications while trickle-charging a Supercapacitor as the main source dipole.

Here is the first iteration of the "FlexFlo (http://www.textfiles.com/bbs/KEELYNET/ENERGY/flexflo.asc)" circuit inspired by George Wiseman's Energy Conservation theory:
FLEXFLO1.GIF (http://anomalies.net/archive/Keely-Net/Diagrams/FLEXFLO1.GIF)

I might get around to trying the second iteration:
FLEXFLO2.GIF (http://anomalies.net/archive/Keely-Net/Diagrams/FLEXFLO2.GIF)

Though this thread has gotten me to think biggerÃ,  :)

It's been fun following what you've all been up to, hoping to see what comes next!

Is this project still alive?

Fausto.

Dr. Spark's circuit, Igor's Switcher and Wiseman's Conserver bear some resemblance.  Charging two caps in a parallel unit, which is also in parallel with two caps in a series unit, then by a prescribed sequence of closing switches/relays placing the first two in series, connected to the second two which are now in parallel, produced a compelling situation:  Current was sloshing back and forth between the two banks, being re-used at the load, with a battery replenishing the miniscule losses during one of the switching states.  One issue I encountered was that both capacitor banks eventually balanced out at full charge and the current sloshing between them would diminish, at least until one of the banks was discharged manually.  So you end up using and re-using the charging current for a while, then you have to do something with the accumulated power to stave off stagnation.

Hi There All,

a youtube sim_replication

http://www.youtube.com/watch?v=nHiGDo47ibw

Thank You Fausto. 

IF you do build use photo flash caps for lower charge and discharge resistance...

Dave

hi all

sorry, i didn't see this thread before - i joined in on the 'Tesla Switch, need help thread' recently with some supporting results for NerzhDishual about Conservation of Charge violation i recorded when switching charge from one capacitor to another and then i started a new thread ('OU/COP>1 switched cap PS cct like half Tesla Switch' in Super Capacitor forum) about the Cap switching results i'm seeing

i'm seeing measured COPs slightly greater than 1 when accounting for output plus all other cct dissipation compared with supplied power

i'm using Super Caps as input supply so i can  get a better measure of energy supplied

interested to see your cct here DrS, my cct also uses the energy travelling between caps but only on a single pass

in the circuit posted, i have transistors in Darlington pairs as switches but in last couple weeks i've been able to replace with MOSFETS, reducing losses, and now getting very interesting results!

last night i think i finally understood what's going on with the apparent '50%'approx input energy losses and i think i know what mods to make to test my suspicions about what's happening

keep up the good work, all - there is definitely some non-Classical ElectroDynamics occuring with these cap-to-cap energy switching arrangements we're using!!

sandy

Sandy, Group,

Thank you for your comments,
the only info i can add at this time is to use the lowest resistance caps possible.

and from a utube comment maybe put a piezo element at the load point.

Dave