Allan's Transformer as a Generator

[AllanV]

Thank you so much again, very specific   

Hi,

My mind went a bit blank this morning. The previous post would give half the principle, it would pay to go bigger with the core and double wind with 15% tappings. Get the full effect.

It may be best if you can find the transformers cheap enough to just copy what I am doing for this next power supply.

A lamination 115 x76, window 19x52, 38x45 bobbin. It will have 4 windings 2@ 55turns 0.8 enameled wire, on first. 2@45 turns x 2.0 enameled wire on last, insulate between rows. Your smaller transformer could work but if it is welded that makes it difficult to get a part without some damage.

The transformer with the gap is wide window 175mm and a 50mmx50mm core area.

If you want to use what you posted just strip it and then find some wire. The bigger the core area the less windings usually. The small transformer will need thinner wire and will have less output.
Do not put too much wire onto the bobbin it becomes a nuisance. The wire as a rule of thumb fills the space to 40%.

Once set up it is not difficult to wind but care must be taken. Keep it tidy and without any abrasion on the wire other wise there will be shorts and failure.

The three phase transformer is not the best for this project. It is a bit small.

2mm wire on two windings, each with a 15% tapping. My transformer has 300turns + 30 on each winding 1.5mm enameled wire.
I have a lot of choices with what has been accumulated. It will be tried to see what happens. It may be rewound with 2mm wire and less
turns.

all the best,

Allan

[Jeg]

Hi all

Allan may i ask, what did make you to go with Amber/Turns variation instead of the phase shift method?

I think a hall sensor inserted in to the core gap would help a lot to visualize any flux changes. Probably we should start by building the right equipment first before testing the idea.

For example
http://www.electronoobs.com/eng_circuitos_tut12_1.php

[AllanV]

Hi all

Allan may i ask, what did make you to go with Amber/Turns variation instead of the phase shift method?

I think a hall sensor inserted in to the core gap would help a lot to visualize any flux changes. Probably we should start by building the right equipment first before testing the idea.

For example
http://www.electronoobs.com/eng_circuitos_tut12_1.php

Hi Jeg,

Experimenting is a long and slightly tortuous path and leads to many dead ends. Each parameter needs to be explored to reduce error. It is a lot of work. But in retrospect it becomes obvious and simple.
Thanks for showing the circuit, I have all the parts necessary to build it. 

The flux gain is significant it is noticeable just by placing a piece of metal close to the gap. And it also shows in the output wave form and voltage increase.

The transformer as a generator, rate of change and strength of the mag field produces the output.

When a current does the circle with a floating supply through uneven turns there is opposition as flux is produced. But it seems that this is necessary to put some tension between the windings and the turns ratio does not allow them to produce too much flux. Saturation occurs when the current increase is greater than the core flux can accept. Once this ratio is established the hi volt pulse current will mostly become AT and then become flux in the core.

To change the flux quickly the currents must cross over. Each current must take a turn at being the greater in one semi cycle, or half wave.
I takes no great DC power to do this when it is compared to the output.
DC only could be used but to cross to the next semi cycle, voltage must be built and then released to expel the flux for maximum current where the currents again cross over. 

The transformer usually operates just above the residual magnetism that stays in the core with no current at all. It is driving down this residual flux to the opposite polarity that gives the best output.

If two phase shifted circuits are used the input will draw maximum current but still has not addressed the problem of potential saturation.
What is being sought here is a substantial flux increase which means a smaller device with fewer turns. Heat is usually the problem IR conductor losses and the core losses. The core losses are two fold and there appears to be a sticking point where the flux will not usually exceed a certain amount.
If the length of the flux path can be shortened it is a bonus as well.

If you have doubts, fair enough. But a starting point is presented that could save a lot of work.

Accepted that to commit to making anything, materials and parts have to be purchased or found from junked parts.

Bigger is easier in some ways, if every thing is too small and wire too thin it is difficult, 0.5mm the smallest for ease. A hand winder with a counter helps.
Anything above 2mm has its own difficulties on a small core. Using multiple wire needs a good setup.

It is noted that you are thinking about how to make it easier.

A couple of L298N stepper motor drivers and a battery to isolate them from Hi volt circuit.
Two home made, wound coils, are all that is needed, one with a gap.
Experimenting is different to building. There is a different mind set.

All the best,

Allan   

   

[AllanV]

Hi all

Allan may i ask, what did make you to go with Amber/Turns variation instead of the phase shift method?

I think a hall sensor inserted in to the core gap would help a lot to visualize any flux changes. Probably we should start by building the right equipment first before testing the idea.

For example
http://www.electronoobs.com/eng_circuitos_tut12_1.php

Hi Jeg,

Just a thought, with a hall sensor the 80 watt transformer test could be verified.

With 240volts on the transformer check the flux outside the core and note the current as well. The RMS current will be greater than a DC current applied next. Why is there a difference?

Disconnect the 240volt power and take a 1.5volt battery and connect this to the 240 volt I/P winding. Place a diode across the winding to stop any spikes.

A reduction to 0.9volts is best therefore a resistor is needed plus a variable one helps. Check the flux outside the core with a meter.

This indicates the DC equivalent and power in the link. The output current dampens the flux and transformers need a 10% over wind on the output to make up the voltage.

In an 80watt transformer with 0.032watt in the link it calculated out to 1/2500th of the load.

When a low power DC current is applied to the two winding set up the flux makes a full change and it represents half the full wave.

But the magnetism is to drive the link and would not produce much power into a load. A setup must include an I/P and an O/P, one leading and one lagging.

Allan     

[AllanV]

Hi all

Allan may i ask, what did make you to go with Amber/Turns variation instead of the phase shift method?

Hi, another thought as an explanation.

When the link flux is modulated the current flow increases significantly. An example is a three phase motor 5kw and 5x+ current at start up, there is no way that power is producing the flux. It is impossible. The generator and the meter notice the power consumption of course.
When a magnetic field is created it does work and when that field collapses it gives back most of what has been used. If the circuit is not interrupted the current comes from the source generator. Each phase in the motor is tightly linked and current draw is balanced between them.
The rewind variable speed drive with a 5Kw input only needs 50Watts maximum to control it. This 50watts DC field is the only link between the 5Kw input and output shaft load.

With a phase shift method;
If the phases are 180degrees opposed plus the input and output driver driven situation the modulation of the small flux allows the maximum current to flow as the voltage increases. This current needs to be redirected.

It would still require another winding on each of the input and output, 15% more or less and a floating supply to pass a current around in a circle through each pair. This creates the increase and decrease in the field. Another inductor would be required to direct the current in a leading capacitor circuit and the lagging load circuit.
What is being done is to create a variation of current in each separate I/P O/P circuit in relation to each other without any reference point except the flux creation. 

A phase shift could increase with flux increase and there would be nothing added to the output.

Allan