Free Energy | searching for free energy and discussing free energy

Solid States Devices => solid state devices => Topic started by: nwman on September 21, 2008, 11:25:56 PM

Title: Behaviours of a transformer questions.
Post by: nwman on September 21, 2008, 11:25:56 PM
Hello all,

 I asked this question in a different thread and category but I think it would be more suited for this category. I'm working with a few ideas of other people and I'm a novice when it comes to electrical physics and magnetism. How ever I do know a little and have seen a few good ideas that I am working on improving. So on to my questions:

First of all I have a question about how the core size of a transformer effects the transformer. For example if you had a transformer that looked like the one below. Where the core inside the primary coil A is smaller (say 1/2) then the core inside the secondary coil B. Lets also say A and B coils have the same number of turns. How would this effect the power transfer through the transformer?


Also, at what saturation level does a transformer work at? Is it below, at, or above the core's saturation point.

Thanks

Tim
Title: Re: Behaviours of a transformer questions.
Post by: valveman on September 22, 2008, 03:54:30 AM
Core size is irrelevent.  The size of the core is determined by what size of wire your using in the transformer and number of windings.  i.e. use the smallest core size that will do the job.

However shape of the core can be important.  It depends on how you want the magnetic field to interract.  Look up toroid transformers.

Saturation of the core is dependant on a few factors.  Material used for the core and frequency of operation.  For instance, most transformers used in powersupplies use  metal that is layered in slices to help reduce the effects of eddy currents.  High frequency cores are made of many matereials.  These are usually powered down, mixed and formed using an epoxy in the form desired.

Saturation of a transformer is not a good thing.  This means that the transformer cannot keep-up with the rate of changing magnetic field.  This will cause the transformer to overheat and eventually fail.

Robert
Title: Re: Behaviours of a transformer questions.
Post by: nwman on September 22, 2008, 08:16:28 AM
Thanks for the input and more is always welcomed. I think this answers most of my questions. I'm in the process of having a modified transformer made and I'm just trying to think through what I might see when its finished and how to test it. More questions probably will fallow.

Tim
Title: Re: Behaviours of a transformer questions.
Post by: gyulasun on September 22, 2008, 12:46:33 PM
Hi Tim,

If you have not seen this link it is worth reading,  an interesting pondering on some of your questions.

http://amasci.com/elect/mcoils.html       Its Fig.11 and downwards is especially interesting. For the iron core that can be curved / bended,  insulated iron wires (garden wires) that are led close parallel to each other i.e. bunched could be used  ;) 

A useful (text book) info on magnetic core characteristics that includes saturation is here:

http://focus.ti.com/lit/ml/slup124/slup124.pdf

Also the first page here shows nicely the saturation on a core's B-H curve:

http://www.powerit.vt.edu/AA/chapters/pdffiles/c4pdf.pdf


rgds,  Gyula
Title: Re: Behaviours of a transformer questions.
Post by: nwman on September 22, 2008, 11:43:32 PM
Thanks Gyula,

Those are great resources. A little over my head in some cases but that's what I need.

Tim
Title: Re: Behaviours of a transformer questions.
Post by: khabe on September 23, 2008, 12:36:04 AM
  High frequency cores are made of many matereials.  These are usually powered down, mixed and formed using an epoxy in the form desired.
Robert

I dont bieleve they are making powder cores using epoxy ... perhaps somewhere in garage or in kitchen ::)
I use often enough Micrometals mix 26 cores, those are machineable, can cut, mill ,drill ... what ever, I have heated, etched - no vestiges about epoxy.
Sand what milling lefts I do collect - then yes - I can use it for some quick-idea experiments mixed with epoxy and press-molded - its good enough, but never so good when real thing.
When more serious projects, then I use Somaloy700. Very hard to find in kilos, more easy when to buy at least 1t. But OK, I have some kilos - then I have to go to factory, to make contract ... they making mould, then pressing, then heat ...
Somaloy its in principle pure Iron - and very good. Added info speaks about Somaloy, there is small precent of lube inside what makes insulation between particles and keeps all together after pressed-heated. For very small parts thats enough if you have 10t press and small good precise(!!!) el.oven with sure timer (!) - I have tried with success.
As you understand Im not much busy with transformers, very seldom, more like motors and generators, but cores anyway ;)
cheers,
khabe
Title: Re: Behaviours of a transformer questions.
Post by: nwman on September 28, 2008, 07:10:41 PM
khabe, they do make laminated c cores and are epoxied together. That's what I am using and they seem to work excellent.

Well, I am more confused now. I just built the transformer shown above and it sucks. I'm not sure why but for simplicity reasons if I put one watt in the primary I only get like .2 watts out the secondary. Both sides have 400 turns of 20 AWG wire all nicely packed  and turned. The only difference is that the core inside the primary has been notched out do that the core material is have the thickness as the core inside the secondary.

Am I missing something? Is my construction just crappy? Or does the thickness actually change the functionality of the transformer?

I took the transformer to a  friend of mine and he said that changing the thickness is like making it a step down transformer. The idea is that its not the number of turns but also the size of the circumference of the turns. The larger the size the more wire the field has to interact with. I would estimate the larger secondary has 50% more wire then the primary.


Could there be any other reason why the efficiency suck?

Tim
Title: Re: Behaviours of a transformer questions.
Post by: gyulasun on September 29, 2008, 01:01:24 AM

Hi Tim,

I assume the air gap between the C cores was at the minimum possible the core manufacturer matched the touching facing areas with their machining? (I mean you use cores with matching surface areas the manufacturer made the cores to be a matched pair.)
Would you mind saying a few words how you measured the input and output power, what was your load, you used mains frequency (sine wave? ).

Here is a transformer calculation link with formulas that includes cross sectional area for the core, though the cut-out core cannot be the main explanation for the low output power problem: http://en.wikipedia.org/wiki/Transformer_design

rgds,  Gyula
Title: Re: Behaviours of a transformer questions.
Post by: nwman on September 29, 2008, 03:26:15 AM
Gyula,

Yes the cores are out of the box condition besides the cut-out area which I made with a large mill. Now I'm not sure if I have the same pair. I bought 3 pairs of them and have only mixed two pairs possibly together. I didn't think it would make a difference? I'll get into more details about the testing later when I have more time to write. I did have a peace of paper between them. Would that small of an air gap cause that much loss? Again, I do not claim to know anything! Thanks for the input.

Tim
Title: Re: Behaviours of a transformer questions.
Post by: nwman on October 09, 2008, 06:50:55 AM
I have a newish question. I've ordered a variable power supply [0-30v,0-5A] to power the DC transformer shown in the picture above. To be able to take Amp reading from the secondary coil I need some kind of load. What would be the best setup to achieve a load appropriate to the varying voltage and amperage? Would it be an array of light builds, DC motor [what size], or a capacitor? Preferably something I could find in town would be great.

Tim
Title: Re: Behaviours of a transformer questions.
Post by: Nali2001 on October 09, 2008, 05:31:40 PM
Hi Tim,
How exactly did you determine the efficiency?
And you say:
"I've ordered a variable power supply [0-30v,0-5A] to power the DC transformer shown in the picture above. To be able to take Amp reading from the secondary coil I need some kind of load"

I also then presume you have a pulsing circuit to pulse the coils? I mean how do you intend to measure the output of the secondary with steady dc?
And you absolutely need a scope with a current shunt if you want to know what is going on with dc pulses. Since pulses work with duty cycles and are not 'on' all the time you need devices that take these into account. And a digital or analogue meter will not show good results. See it this way:
watt is volt times amp right? But that is only so in a dc situation. Now if you were to drive a transformer/motor/whatever at 100v dc, and pulse it with 100amps BUT with only a duty cycle of say 10% your meter will probably say something like 100v(or less) and about 75amp, So 7500 watt. But that is bogus. it is more like (100voltx100amps) / 10duty= 1000watt
You can only truly determine these ratios with a scope. Also look up 'Rms' or 'true Rms'

Title: Re: Behaviours of a transformer questions.
Post by: nwman on October 11, 2008, 03:29:52 AM
How exactly did you determine the efficiency?

I used a digital multimeter with a max setting to read both the input and output Volts and Amps. I used both a small DC motor and small light bulbs as the load and had both two AA batteries in searies and a 9v. So its not a full proof testing situation and my numbers may be flawed but on most occasions it averaged out to 75%+/- .

I also then presume you have a pulsing circuit to pulse the coils? I mean how do you intend to measure the output of the secondary with steady dc?  

Again, not too perfect but I just used a push button switch to pulse the coil by hand. It seems to give fairly consistent readings besides the drain of the batteries. Is there any simple product I could buy that would just keep a constant pulse at the varying V and A?


And you absolutely need a scope with a current shunt if you want to know what is going on with dc pulses. Since pulses work with duty cycles and are not 'on' all the time you need devices that take these into account. And a digital or analogue meter will not show good results. See it this way:
watt is volt times amp right? But that is only so in a dc situation. Now if you were to drive a transformer/motor/whatever at 100v dc, and pulse it with 100amps BUT with only a duty cycle of say 10% your meter will probably say something like 100v(or less) and about 75amp, So 7500 watt. But that is bogus. it is more like (100voltx100amps) / 10duty= 1000watt
You can only truly determine these ratios with a scope. Also look up 'Rms' or 'true Rms'

I'll look into this. I may need to buy a scope but for now I'm taking you advice and taking small steps. Would you have any recommendations on a scope? I have seen people show reading from that look like a software based scope. Is this expensive to  buy?

For now I just want to see a crude test of pulse in and pulse out and see what happens. Input is welcomed.

Tim
Title: Re: Behaviours of a transformer questions.
Post by: Nali2001 on October 11, 2008, 05:15:30 PM
Hi Tim,
First of all, I know I often sound somewhat direct and such in my answers but I mean no 'harm'
Nice to finally see someone diving into the vast field.

Some answers:
Your meters in this test are more or less useless especially the digital once, analogue would be better but still not all that useful. Any experimenter needs a scope. This makes it all much more clear and makes ya pulses visible as well. But it is a learning curve on it's own so just to remind you that this field is vast, even the basics. But coming back to the scope, there are a lot of second hand scopes on ebay and such. These are often oldish analogue scopes which do not much more than just show the curve and thats it. Now for a few more dollars you can get yourself a hand usb scope that comes with software http://blog.cc-robotics.de/wp-content/uploads/2008/02/dso2090_1.jpg (http://blog.cc-robotics.de/wp-content/uploads/2008/02/dso2090_1.jpg) that lets analyze and measure on the fly what your curve is doing. It can also calculate rms valves for easy calculation of wattages. I guess I would recommend the dso-2090,  http://cgi.ebay.com/New-DSO-2090-100Msa-s-USB-PC-Digital-Oscilloscope_W0QQitemZ300263917184QQihZ020QQcategoryZ104247QQssPageNameZWDVWQQrdZ1QQcmdZViewItem (http://cgi.ebay.com/New-DSO-2090-100Msa-s-USB-PC-Digital-Oscilloscope_W0QQitemZ300263917184QQihZ020QQcategoryZ104247QQssPageNameZWDVWQQrdZ1QQcmdZViewItem)
It comes with two probes as well. http://www.partsforhdtv.com/images/tools/dso-2090_02.jpg (http://www.partsforhdtv.com/images/tools/dso-2090_02.jpg)

Well about the pulsing. It more or less depends on a few things I could send you a circuit digram which will show ya how to build a fet based pulsing system. Or maybe a relay, although don't expect them to last long at such speeds... http://en.wikipedia.org/wiki/Relay (http://en.wikipedia.org/wiki/Relay)

You are right and take the small steps so that they are understandable. That way you have something to show for in the long run.