In a switch mode power supply with a flyback transformer you can not view the transformer in the same way you view an AC transformer. Notice the polarity dots of the transformer, they are reversed. In this case, with a standard AC transformer, a positive wave would yield a negative output. This is counter intuitive until you understand what the transformer really does in an SMPS.
In a switch mode power supply (SMPS) the transformer is an energy storage device that is charged up when the switch device is on. When the switch device turns off the field collapses and the transformer acts as a constant current device to charge the output capacitor through the diode. This is a wonderful thing. Using the transformer in this way means you are not restricted to the turns ratio and still get high efficiency.
So to cut it short. the diode turns the transformer from a voltage device into a constant current device, which is far more efficient.

 

,I didn't know that the transformer has grounding at all.
I know almost everything about the regular transformer but this smps is totally different.
Anyway,those examples in the photo help me to understand why the wave form could be unidirectional(positive or negative)or bidirectional.
In the photo,at the second possibility,where the ground is in the middle and the wave form is bidirectional,we need rectifier with 4 diodes,don't we?
If i connect the smps to 230vac,the voltage after the 4 diodes rectifier will be 325VDC,this pulsed dc flow to the large capacitor.Does the voltage at the capacitor will be around the 325vdc?
From the capacitor the current will flow to the primary winding of the transformer in aform of high frequency pulses that are controled by the controller that also controls the duty cycle of the pulses,depends on the load consumption .Is it mean that the PEAK voltage on the primary winding is always at a round 325vdc.while at 100% duty cycle the average voltage is 325vdc but at 50% duty cycle(-for example)the peak voltage is 325vdc,but the average will be 162.5vdc(325/2)?

Does the secondary winding has less number of winding from the primary winding?does the voltage decrease due to the difference of the winding numer?(This question is because i would like to know if it does,at least the ordinary thing that the normal transformer does.)
If the output of the smps is 12vdc is it mean that the secondary winding has pulsed average voltage of 12v that stay constant opposed to the primary winding that its average voltage change by the controller (duty cycle)in order to keep the secondary winding constant average 12v?
And the final one:does the pulsed current that flow from the secondary winding to the rectifier get out from the smps to the load as pure direct current(with out pulses)thanks to the capacitor as the same pure direct current on the output of the ordinary power supply(with no pulses)?

This is very confusing subject so i have been trying to give all the details that i am not sure about them.

 

In the photo,at the second possibility,where the ground is in the middle and the wave form is bidirectional,we need rectifier with 4 diodes,don't we?

Two would be sufficient.

If i connect the smps to 230vac,the voltage after the 4 diodes rectifier will be 325VDC,this pulsed dc flow to the large capacitor.Does the voltage at the capacitor will be around the 325vdc?
From the capacitor the current will flow to the primary winding of the transformer in aform of high frequency pulses that are controled by the controller that also controls the duty cycle of the pulses,depends on the load consumption .Is it mean that the PEAK voltage on the primary winding is always at a round 325vdc.while at 100% duty cycle the average voltage is 325vdc but at 50% duty cycle(-for example)the peak voltage is 325vdc,but the average will be 162.5vdc(325/2)?

If there is a bridge (four diodes) and filter capacitor at this point as you say (most smps will not need the capacitor), then yes it will be charged up to 325volts.

For the rest it would be best to draw at least a block diagram, since there are several types of smps.

 

" why is there a rectifier"? to rectify the high frequency ac coming out of the transformer. has nothing to do with the phase of the windings, charge time of the filter caps, or the phases of the moon, it is just there to rectify the ac coming out of the transformer to DC.

 

" why is there a rectifier"? to rectify the high frequency ac coming out of the transformer. has nothing to do with the phase of the windings, charge time of the filter caps, or the phases of the moon, it is just there to rectify the ac coming out of the transformer to DC.

No,Studiot is right (& I was wrong!).

In the special case of a flyback transformer,which stores energy,the current in the primary sets up a unipolar magnetic field ,which collapses when the current source is removed.
The effect of this is to induce a unipolar pulse in the secondary.
(This is very similar to what happens in an old style Kettering car ignition coil.)

In this case,the rectifier only conducts during that pulse--there is no other half cycle.
(In a practical circuit there is a small overshoot in the other polarity,but it is negligible,)

Sorry,Studiot,--I have worked on many SMPS,& I am familiar with charge storage in a transformer,but somehow,never thought it through properly in this case.

Now I understand that you are thinking from the perspective of that kind of transformer,your reference to various pulse shapes makes perfect sense!

 

(This is very similar to what happens in an old style Kettering car ignition coil.)

Exactly......The purpose of the output diode in an SMPS is to select whether it is a "forward" or a "Flyback" topology. These two modes of operation are so different in nature that to say that the diode rectifies the output of the transformer, in a SMPS is, at best, a miss-leading statement.

 

No worries, vk6zgo, we have all learned something here.

Welcome Les, hopefully your travels will add more pearls to the forum in future.

xchcui, don't forget that there are many types of switching regulator, some have more sophistication than others. Keep the good questions coming.

 

Two would be sufficient.



If there is a bridge (four diodes) and filter capacitor at this point as you say (most smps will not need the capacitor), then yes it will be charged up to 325volts.

For the rest it would be best to draw at least a block diagram, since there are several types of smps.

Unfortunately,i don't have a scanner,and even if i had one,i don't know exactly how to draw it.i read the operation of the smps in wikipedia(the other sites that i checked,explained the subject in the same way)and i copied the steps of the circuit in steps as it shown in wikipedia(the isolated one).Between the steps i asked my questions since their explanation are not clear enough.
The main things that i would like to know are(the same idea as i mention on the last post):
1)In order to maintain constant voltage to the load the controller change the duty cycle on the primary winding.If the capacitor is charged with about 325vdc,it is mean that the peak voltage on the primary winding is around 325vdc,while the average voltage on it is depended and determined on the duty cycle from the controller,isn't it?
If the load need 12vdc,the secondary voltage is about fixed 12vdc and only the primary winding voltage is changed by the controller in order to keep the secondary winding on 12vdc when the load consumption is changed,Is it right?
Does the output current of the smps is the same output current of the ordinary power supply-pure direct current with no pulses(of course with some ripples)?
I understand that there are many kind of smps but,as i mention,i copied the steps from wikipedia and i try to understand in general through these step by step questions the general idea of its operation.

 

You need to take my comment about different SMPS more seriously.

I have no way of knowing how the actual unit you are looking at is configured.

How do you know for instance that it has a transformer, and that what you see is not an inductor?

Probably the most popular type of SMPS these days has an inductor only.

http://cds.linear.com/docs/en/application-note/AN140fa.pdf

You can draw a block diagram in MSWord or MSPaint.

There was a question about this from another member recently

http://forum.allaboutcircuits.com/showthread.php?t=97820

Alternatively have you a digital camera or phone, you could then photograph your sketch.

 

a switchmode power supply putting out dc without a rectifieer in the output? dont sound very reasonable to me if it is supposed to be putting out dc. sparkplugs dont need dc, and dont get it from a kettering ignition. one pulse might have more output in one polarity than the other, but it is still not dc without a rectifier.

 

in most smps I have worked on, the output winding is one big thick humk of copper, normally one turn for a 5 volt dc output. how does this magic unipolar output pulse charge up the 1000 or so mfd filter cap without letting it discharge back through the very ow resistance winding? a hint, there is a rectifier in the way.

 

You need to take my comment about different SMPS more seriously.

I have no way of knowing how the actual unit you are looking at is configured.

How do you know for instance that it has a transformer, and that what you see is not an inductor?

Probably the most popular type of SMPS these days has an inductor only.

http://cds.linear.com/docs/en/application-note/AN140fa.pdf

You can draw a block diagram in MSWord or MSPaint.

There was a question about this from another member recently

http://forum.allaboutcircuits.com/showthread.php?t=97820

Alternatively have you a digital camera or phone, you could then photograph your sketch.

I am sorry,studiot,but i don't have any special smps that i am looking at.
I just copied what wikipedia said a bout the smps and i tried to ask question that i didn't understand in there explanation.
They said that there is a transformer in the smps,so i asked questions about smps that has transformer.
Now you said that the most popular smps has an inductor only.
This issue become more complicated from what i thought it will be,and since it is not possible to answer to my last question,i ask only one question the refer to the all types of the smps:
Does the current output of the smps to the load is,after all,pure direct current(besides the ripple)with no pulses?I mean,the same as the regular dc current that in the output of the ordinary power supply?

 

Yes.

The goal of any DC power supply is to supply DC. The switch mode supply only does it with a higher frequency than power line supplies. That makes the output capacitor not need to be so large as it is replenished thousands of times a second instead of dozens of times a second.

 

Thanks everyone for your replies.

It is true that there are couple of things that i don't understand about the smps operation process,but the answer for the main question of this thread had been answered.

Thanks