Microwave oven transformer as a 12V power source

[boylesg]

Any words of wisdom about this sort of thing before I consider giving it a go?

According to this, you can remove the secondary windings, add a new secondary winding of 10-20 turns of suitably amp rated cable and create a 12V power source that can deliver high amps.

 

[tpetar]

Any words of wisdom about this sort of thing before I consider giving it a go?

According to this, you can remove the secondary windings, add a new secondary winding of 10-20 turns of suitably amp rated cable and create a 12V power source that can deliver high amps.

Yes, that is true, You can find a lots of videos on YouTube portal about this, and usually people make spot welder with this transformer. This transformer cannot run a longer time with extra high power, core will be very hot. Check needed core sqaure area for power which You plan to use on longer time.

Example videos:
http://www.youtube.com/results?sear...9.5.0.0.0.0.0.0..0.0...0.0...1ac.1.11.youtube.


Best regards,
Peter

 

[boylesg]

Yes, that is true, You can find a lots of videos on YouTube portal about this, and usually people make spot welder with this transformer. This transformer cannot run a longer time with extra high power, core will be very hot. Check needed core sqaure area for power which You plan to use on longer time.

Example videos:
https://www.youtube.com/results?sea...9.5.0.0.0.0.0.0..0.0...0.0...1ac.1.11.youtube.


Best regards,
Peter

Well I have no intention of using it as a spot welder or anything like that.

I understand that such an application would draw hundreds of amps from it and I could understand why the core and the primary windings would get rather hot if the secondary current remained flowing for an extended length of time.

I just wanted 30A or so at 12V. 30A would not be too much to ask of it, re heating, would it?

As I understand it MOTs are designed to run with the primary in saturation and that, if I use it as is, then it will probably wasting a lot of primary current to produce my 12V/30A.

I have also read that, for the purpose of driving tesla coils etc, you can use another MOT, with its primary shorted or removed, as a ballast inductor to limit the primary current and bring the main MOT out of saturation.

So what if I was to wind a some turns of the mains line around the ferrite core of a flyback transformer? I have save a dozen or so of these for just such occasions. Is that likely to provide enough ballast for my purpose?

I have a few ferrite toroids, through which they often wind a few turns of the power cord of a TV, as well. Perhaps they would be better?

 

[tpetar]

You can regulate current with triac dimmer on primar winding.


Best regards,
Peter

 

[boylesg]

You can regulate primary winding with triac dimmer.


Best regards,
Peter

As in a standard dimmer switch from the hardware store I presume?

So my understanding of the saturation issue with these transformers is accurate then?

I am just starting to try and teach myself about transformer theory so my knowledge is rather patchy at present to say the least.

I take it that if I have the dimmer set to low, such that that the transformer can't supply the secondary 30A or what ever, then the secondary voltage will start 'sagging'?

 

[tpetar]

Yes you will control current on secundar winding.


Check this dimmer:

Triac regulation of mains voltage
http://danyk.cz/triako_en.html


Best regards,
Peter

 

[dick_freebird]

30A is big, but on trash day there is probably at least one
PC with an ATX power supply sitting by the curb along
your street (if you aren't willing to pay the $10-20 for
a used one at some PC repair shop). You could expect
somewhere between 10 and 20A capability, with no more
work than figuring out which low voltage supply needs
the dummy load, and which control wire needs grounded
to enable startup.

 

[boylesg]

30A is big, but on trash day there is probably at least one
PC with an ATX power supply sitting by the curb along
your street (if you aren't willing to pay the $10-20 for
a used one at some PC repair shop). You could expect
somewhere between 10 and 20A capability, with no more
work than figuring out which low voltage supply needs
the dummy load, and which control wire needs grounded
to enable startup.

Already have an ATX power supply in mind and the best I have done thus far is one that can provide 22A at 12V with the two 12V supplies paralleled.

It is for a RGB cube with a peak current draw of around 26A.

The reason for using the laminated core is to limit current induced in the core itself by the primary windings.

So why doesn't the welds, across the laminations that bind them together, create circuit paths through which current can flow within the transformer core?

Welding the core laminations together seems to defeat the purpose of electrically insulating the individual core laminations with varnish.

 

[tpetar]

Laminates in iron core of transformer are used to limit Eddy currents. Without laminations, iron core would heat up very fast, with high waste of energy.

Microwave oven transformer is cheap transformer, and fast for production. This transformer is welded on few points with TIG method.

https://www.youtube.com/watch?v=KRoPHKpCYmg


Best regards,
Peter

 

[boylesg]

Laminates in iron core of transformer are used to limit Eddy currents. Without laminations, iron core would heat up very fast, with high waste of energy.


Best regards,
Peter

Which direction do these eddy current flow. From diagrams on the net and the fact that each individual lamination has to be varnished it appears that those eddy currents would flow from one lamination to the next if they weren't insulated.

So if you do a weld across the edges of the laminations, e.g. one at the top of the E and two on either side of join between the E and the I as with the MOT core I have at present, then why are you not undermining the insulation between the laminations?

- - - Updated - - -

Which direction do these eddy current flow. From diagrams on the net and the fact that each individual lamination has to be varnished it appears that those eddy currents would flow from one lamination to the next if they weren't insulated.

So if you do a weld across the edges of the laminations, e.g. one at the top of the E and two on either side of join between the E and the I as with the MOT core I have at present, then why are you not undermining the insulation between the laminations?

Is there any practical way to mitigate the sound they make? That would be rather irritating if you were using it as a power source as opposed to a spot welder.

 

[nilses]

30A is big, but on trash day there is probably at least one
PC with an ATX power supply sitting by the curb along
your street (if you aren't willing to pay the $10-20 for
a used one at some PC repair shop). You could expect
somewhere between 10 and 20A capability, with no more
work than figuring out which low voltage supply needs
the dummy load, and which control wire needs grounded
to enable startup.

Welding ?

Connect the green wire to ground, any of the 3 black wires next to it. Then connect a faulty harddisk or a lightbulb to +5v and ground and you have suddenly +5v @34 AMPS and +12Volts @manyamps for many moons. If the powersupply stops working, use a vaccum cleaner in the fan area. Lubricate the fan ballbearing and Thou will have a powersupply for many moons.

 

[boylesg]

Welding ?

Connect the green wire to ground, any of the 3 black wires next to it. Then connect a faulty harddisk or a lightbulb to +5v and ground and you have suddenly +5v @34 AMPS and +12Volts @manyamps for many moons. If the powersupply stops working, use a vaccum cleaner in the fan area. Lubricate the fan ballbearing and Thou will have a powersupply for many moons.

The problem is that standard ATX power supply in PCs likely to be dumped on trash days never seem to be rated above 18A on 12V. However my RGB cube could draw about 26A if all 512 RGBs are white.

A better option would be to just purchase a dedicated LED supply switch mode transformer on ebay. I still could do that but I thought I would look into this MOT business.

Even with the purpose made switch mode supplies for LED displays, you still have to wire them up to the mains, place them in an appropriate insulated or grounded hosing and make sure they have sufficient heat sinking or ventilation. So I suspect using them is not all that much less dangerous than using a modified MOT.

If I could use 5V to power my RGBs then a standard ATX power supply would do the trick because they are often rated for 30A at 5V as you pointed out. However I am using 1 transistor on each cathode column and two layers of multiplexed transistors on each anode row.

That is likely to be cutting it pretty fine with respect leaving enough voltage on the RGB anodes to drive them, particular the blue bit of the RGB that requires a higher voltage than the other colours.

 

[tpetar]

Which direction do these eddy current flow. From diagrams on the net and the fact that each individual lamination has to be varnished it appears that those eddy currents would flow from one lamination to the next if they weren't insulated.

So if you do a weld across the edges of the laminations, e.g. one at the top of the E and two on either side of join between the E and the I as with the MOT core I have at present, then why are you not undermining the insulation between the laminations?

- - - Updated - - -



Is there any practical way to mitigate the sound they make? That would be rather irritating if you were using it as a power source as opposed to a spot welder.

This is a cheap transformer. You can see that laminated are not even arranged on standard way for EI core. Welding of tranformer laminates gives high level of automation and high speed in production of transformers. It does indeed increase losses but helps manufacturers to sell the trafos cheaper. The core is made out of transformer sheet steel laminations assembled to provide a continuous magnetic path with minimum of air-gap included. The steel should have high permeability and low hysteresis loss. For this to happen, the steel should be made of high silicon content and must also be heat treated. By effectively laminating the core, the eddy-current losses can be reduced. The lamination can be done with the help of a light coat of core plate varnish or lay an oxide layer on the surface. For a frequency of 50HZ, the thickness of the lamination varies from 0.35mm to 0.5mm for a frequency of 25 Hertz. Microwave ovens transformers operate pretty far into saturation, as a result of this, they need to be fan cooled.

I will say this is acceptable compromise between the cost of manufacturing and product quality.

Eddy currents also called Foucault currents:

Best regards,
Peter

 

[FvM]

As far as I'm aware of, microwave oven transformars are stray-field transformers. They aren't well suited for general transformer purposes.

 

[boylesg]

As far as I'm aware of, microwave oven transformars are stray-field transformers. They aren't well suited for general transformer purposes.

By the looks of it a MOT is a stray field transformer as long as it has the magnetic shunts in it between the primary and secondary.

Based on what I just read, the shunts decrease the coupling between the primary and secondary windings and thus limit the current in the HV secondary.

But if you remove the HV secondary and the shunts and add a new LV secondary then it becomes a standard transformer as far as I can see.

 

[FvM]

But if you remove the HV secondary and the shunts and add a new LV secondary then it becomes a standard transformer as far as I can see.

Yes, if it's feasible.

 

[tpetar]

Here is one example:

Burnt MOT - Powerful source of low voltage
**broken link removed**


Best regards,
Peter

 

[FvM]

Here is one example

But not in the way discussed above, because the magnetical shunt is still in place. In fact, I'm not sure if it can be easily removed.

For some applications, the shunt may be even wanted, e.g. battery chargers.