Hi NRamaswami,

First of all please bare with me, I try to answer in due order as my time permits.

At the moment it is the core you may wish to decide on, what are the mechanical sizes of the straight "I" laminations you have got?  (If they are I shaped, that is, for I assume you wish to use three electromagnets in a straight line, right?)  So you may wish to think about fitting I laminates with rectangular cross section into say a 4 inch dia cylindrical PVC tube, if this is what you are planning.

Regarding your trifilar and bifilar experiments and the different current draws, it is against common findings: it is okay how you connected the trifilar (or bifilar) windings but then the L inductance for such trifilar coil (three windings connected in series as you also described) should increase about 8-9 times with respect to that of a single wire winding having the same # of turns. And this should be a 3-4 times increase in case of a bifilar coil. 

So the AC impedances should also increase in the same amount: from a single wire coil having say Z=40 Ohm, the impedance would be 9 x 40= 360 Ohm for a trifilar coil having the same number of turns than the single wire coil. 
And then the current draw should decrease: from 220V/40 Ohm=5.5A to 220V/360 Ohm=0.61A.  I understand that you found the 32A office tripper tripped for your trifilar coil and your bifilar drew 18A, so there is a puzzle here: perhaps the winding styles for the trifilar (and for the bifilar) coils were very different with respect to the single wire coil, not allowing for an inductance (hence impedance) increase but a decrease!

Regarding your Ohm meter readings: when you directly short the two measuring tips to each other, ideally you should see 0.000 Ohm in any range, including the 2k range. If this is not the case,  then you have to substract the non-zero Ohm part from the measured part: say the display shows 0.01 Ohm when you directly short the measuring pins and then you measure 0.08 Ohm on a coil, the true Ohm value would be 0.08-0.01=0.07 Ohm for that coil. By the way, using the smallest range (which is perhaps 200 Ohm for your meter) would give more precision (resolution) in your low Ohm measurement efforts.

Regarding your post on the magnet wire: it sounds the magnet wire available for you is not insulated but bare, otherwise you would not need to add insulation to cover it, right?  OR if you mean the magnet wire which is available for you is already has the usual enamel coating, then it does not need any further insulating layer and could be re-used several times. Please address this too.

More to come later. No need to hurry.


Gyula

Gyula:

Thank you so much.. There seems to be a diffeence in the way factory made 3 core coils work. They have huge insulation which is over and above the 1100 volt insultation of the single core wires and are rated to be capble of withstanding about 11000 volts I think. It is this insulated wire that took so much of amps.

When we duct taped the wires ourselves this did not happen. At that time the wires quadfilar wires took only 7 amps which is consistent with what you are saying. Does insulation huge insulation and gaps between wires have any thing to do with increased amperage consumption and increased magnetism.

If your statement on quadfilar wire is correct then I should be able to immediately replicate the 630 volts output experiment and check using a step down transformer what is the output. I guarantee that in this device when you keep the poles NS-NS-NS the output from the secondary increases with voltage increase. Not only does voltage increase but amperage also increases wth voltage.

I do not have that many I cores. They are very sharp and cut the hands and are only 6 inches long. We have electromagnets that are usually 5 to 6 feet long when we use the 4 inch tubes and so we use the cheap soft iron rods.

Doug:

We have done a small Square device Like this. CW-CW-CCW-CCW. All of them are placed in a rectangular way. It essentially means that the parallel electromagnets are wound in the opposite direction. and so we have the NS-NS-SN situation here. Not just that it is NS-NS-SN-SN ..

Results for you..

Each electromagnet 2.5 inch diameter. Soft iron core. 8 inches long.

Primary 4 sq mm wire on two oppposite electromagnets. Secondary1.5 sq mm wire on all the four.

Primary turns 35 x 3 layers. Secondary turns 65x 3 layers.

Total number of primary turns 105+105=210

Number of secondary turns 65x3=195 per electromagnet.
For 4 electromagnets Total secondary turns 195x4= 780 turns

Primary 210 turns and secondary 780 turns.

We gave a load of 220 volts and 7 amps and connected the circuit like this.. Mains Phase - Primary input - Primary output - Resistive load of 10x200 watts lamps - Mains Neutral.

Secondary was connected to secondary load.

Primary input 220x7 amps. = 1540 watts.

Secondary output -- only 12 volt lamp is able to burn in a light way.. Nothing more than that. Magnetism is present in all cores but is very weak in two of them and strong in two of them. Though the magnet is small the current given is significant and output is unacceptably low.

Primary and secondary are both only single core wires. I did not put them as routine electromagnets for the wires will not stand a chance to act as electromagnets. We will need a lot of turns of the primary wire being 4 sq mm wire and out experience is that it requires a minimum of 240 turns of wire for the magnet to remain stable.

We will repeat the experiment of Dieter as well and report.

My boys will test it now in NS-NS-NS configuration and I will report the output. I will also provide the primary and secondary turns. We will see the results. We will use the same above method and then find out what is the result in the secondary. I do not know theory and so I really go by the experimental results.

We will accept the experimental results.   

Doug:

I concede the magnetic flux loss issue is there. I had been told by a client that at 90' magnetic flux loss would be very heavy. Seems to be the case also here.  I will not be able to do much this week due to heavy pending work and will get back to testing later.

Gyula: Your insight is so accurate that I'm really stunned. But I'm still unable to understand why the multicore wires perform well only when they are wound two times and do not perform well when the layers are more than 2.

If my understanding of your writing is correct, greater inductance will increase the magnetism but would also consume more amperage and greater impedance would reduce the amperage consumption but would lead to lesser magnetism in the coil. Am I right? Please reply at your conveience.

Hi NRamaswami,

I think it would be a very good step for you to obtain an L meter and check the inductances of your single wire, bifilar and trifilar coil constructions. I say this because all the coil calculating formulas or online coil calculators are based on normal enamel-insulated copper wires (i.e. 'magnet' wire) and not based on wires covered with thick insulation layers which introduce space or distance (and capacitance) between adjacent wires, either between turns one above the other or between turns next to each other.

I do think that obtaining an L meter is the quickest way to advance your electromagnet constructions because by knowing the actual primary coil inductance in an assembled setup you can calculate the AC input impedance in advance, this makes the input current estimation possible in advance, you could also see the effect of the secondary load on the primary input coils inductance, how the presence of the secondary load would modify the input impedance of the primaries,  this way you also could quickly browse through some electromagnet coils you have left from earlier experiments etc.

When you have a measured primary coil(s) inductance, and you have the DC resistance of the same coil(s) also measured with your Ohm meter, the input Z impedance can be calculated as I showed earlier, I can help you in that.

With the L meter suggestion I do not mean to toss the ball into your field of course but to indicate for you that there is no calculation readily available for arriving at coil inductance (hence impedance) values when using wires with thick insulation, at least I am not aware of such.
Especially with open core coils this becomes more problematic because manufacturers specify permeability data for their soft iron cores when the core has a closed magnetic path (usualy the permeability data is specified for ring cores made from a given soft iron material).
This means for instance that a normal transformer lamination may have say a permeability of 800 in a closed core, no air gap, and using a certain section of this same lamination as an open core its permeability becomes much less and quasi unpredictable. This is where an L meter can help tremendously.

I understand that you do not have many I laminations but if you are going to use iron rods again, then your electromagnets would become 'ill-behaved' again, heat losses and not readily repeatable results would prevail.

So either you obtain more 6 inch long I laminations or you may wish to consider the steel pellet core I referred to already.

A notice to the I laminations: you can place some 6 inch long piece in a single line and then put a second row onto them which would overlap the gaps and so on, of course this would need a sufficient amount of I laminations. Yes they have sharp edges indeed, they need careful handling.

Ferrite cores would function correctly too but they have become a bit expensive during the years. On ebay you can find 8 to 10mm dia ferrite rods with 180 to 200mm length like this offer http://www.ebay.com/itm/16x-Large-Balun-Ferrite-Rods-10x200mm-/201042983709  but you would need more than 16 such rods to fill up the inner volume of a 45-50cm long coil bobbin with 2.5 inch dia and then you would need still twice as many for the other 2 electromagnets.

More on your questions later on.

Gyula

Hi all,

This thread has grown a lot. 20 pages in 15 days!!  I post here a little detail from a Buforn patent to get relaxed among so much technical info:

Wings,
Very interesting old news snippet. Where did you find that?


I have an intuitive believe that there is indeed a possibility to break the law of energy conservation. Call me mad if you want. Free energy is bad for business and businessmen made the rules. J.P. Morgan, David Rockefeller etc., these guys rule trough control of energy, this is true.


Anyway, thinking about permanent magnet motors, there has always been the lack of being able to turn a magnet off in the right moment. But you know what? You can turn an electromagnet off! You can store its magnetism in an electrical charge and turn it off that way. The electrical charge is then useable by the next electromagnet, with few losses, but you already got the work of the last one, eg. done by attraction. So there is a logical gain.


Going out of phase with one primary could achieve a similar effect. It's all about timing. If this were a motor, let the attractor always be ahead of the rotor. This doesn't require more magnetism than  any sticky point, it's just a timing issue, that can be solved with an inductive delay. And an electromagnet becomes permanent when you short circuit it in magnetized condition. So basicly this is a permanent magnet that can be turned off.


The same applies to solid state generators.


Today I tested my generator with a steppermotor controlcard, "smc800", this is for centronics printerport, so I had to reactivate an old laptop.

Unfortunately my prototype has bad dimensions. Only 2.1 ohm in each primary, so I again had to use resistors to make sure not to draw too much current from the SMC800.

Finally, the voltage on the primary was only -.44 to .44 V, so (0.88*0.88)/2.1= 0.368 Watt consumption on each primary, 0.73w in total. The output was rectified and smoothed by a big cap, the cap went up to 25Vdc, when measuring the amps with the digital meter (the one with defective AC measurement), it read 80 DCmA, so that may be 2 Watt output.

The SMC800 was run with up to 1000hz and NRamaswami was right, higher frequency brings better output (well, not radio frequencies of course).

I am not sure if these measurements are all correct and I have to say, the smc800 is using amplitude synthesis that had a sawtheeth by its own, maybe 10khz, that I can clearly hear when I run it with silly speeds like 0.2 hz , amusingly even these frequencies, that are not my sinus amplitude, generate pretty high currents, although I have no idea if these are out of phase in one primary, obviously they are, because as I know, otherwise the output would be cancelled out completely,

I will try to find a better "driver", a steppermotor control card, powered by a pc supply, plus a laptop, just to generate 2 Watts, that's bulky. I'd rather have some smart transistor pcb that runs from a 9volt dc source. I will however build my next prototype with about 17 ohms in every primary, that's much more practical.

Regards

Wings,
Very interesting old news snippet. Where did you find that?

Regards

in the book see this Farmhand post:
http://www.overunity.com/12794/re-inventing-the-wheel-part1-clemente_figuera-the-infinite-energy-machine/msg390602/#msg390602
lot of ideas   
https://ia700302.us.archive.org/16/items/inventionsresear00martiala/inventionsresear00martiala.pdf[/font]

Hi Gyula:

By the way I have no bobbins. I use plastic tubes which are 2.5 inch or 4 inch dia and 30 cm to 50 cm in length for winding the wires. Each one of them is packed with softiron rods. 2.5 inch takes about 60 of 6mm dia soft iron rods while 4 inches one take a lot more. Soft iron rods are about 43 cm long. or 30 cm long. All cores are either soft iron or made up of iron powder which is packed in to the tubes with plastic caps on both sides. 99% of the time we do not use iron powder. All winding is by hand so far. No machine winding and no enamelled wire so far.