SMPS problem

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi to all.. i hope to get an answer here! :)
I've build the following schematic
An externally hosted image should be here but it was not working when we last tested it.

Now i've the following problem..
At the output rail i've measured +80 and -50 instead of +/-40 as expected and the IRF540 become very very hot... (one time they've unsoldered themeself).
Now i'm ansking myself... what's wrong? A ppl tell me that from schematic frequency is around 28Kh and he suggest me to set it to 500-700Khz...
I don't own an oscilloscope so i can't check directly sorry...
Maybe the problem IRF540? How i've wind the transformer?
"how i've wind the tranformer"
Please some one help me...
Thanx in advance to all.
mail:maxiz@inwind.it
ICQ:47573544
 
Where is the output inductor? I don't see an inductor in the output, after the transformer. Without an inductor, how is the energy per cycle being stored? Also, what is the purpose of the full wave bridge rectification on the output? This is obviously a push-pull topology, but the output inductor is missing, and the expected full wave center tapped output rectifier has been replaced with a full wave bridge rectifier. I'm puzzled. If you answer the questions above, I might be able to help you troubleshoot the SMPS. Best regards.
 
Claude i'm a beginner so i've just build the schematic without know exactly what i should do o that's missing in the schematic.. sorry! :) Anyway
where should stay Inductor output? How to calculate it?
this SMPS will be used for an audio Amplifier...
Could u explain me the difference between:
full wave center tapped output rectifier and
full wave bridge rectifier.
I know that these question for u are stupid and a lost of time.. but if u could give an answer.. privately too..
mail:maxiz@inwind.it
ICQ:47573544
 
First of all put a heatsink on your mosfets. I believe the reason your mosfets are getting hot is because you don't have enough turns on your primary. Put about 10 turns and go from there. Also make sure your very careful about how you go about finding your center tap. this could be why your having such large offsets.
Once you have your primary figured out your secondary shouldn't be so hard to figure out.
 
You should use output inductors (chokes). Don't worry too much about polarity on the chokes now. Also configure the rectification differently than you show. You may also have to add windings to the secondary after doing so to the primary. All the suggestions offered have been good. I would not even have seen the problem with your rectification if it had not been pointed out and I am surprised that I didn't think of the output chokes.
 

Attachments

  • transfout.gif
    transfout.gif
    3 KB · Views: 2,626
Something else may help you. I wind my chokes air core. I take magnet wire and wrap it around a smooth, round stick form in two sections separated by a short length of magnet wire. Then I bend the connecting wire to make the two halves next to each other so that they will pass current in opposite directions. This method should cancel EMI at a distance.

Try different sized chokes after the rectifiers until you get best performance. I usually make them just large enough to increase the efficiency to over 95%. But you will not reach that kind of efficiency with your circuit.
 
Correction

It occurred to me that the bridge rectifier in the schematic MaXiZ posted is the same as the discrete component set-up I posted. :bigeyes: Now I am thinking the problem is likely to have something to do with the transformer windings. I suggest you rewind your transformer, removing all the present turns and starting over. Make sure that you add more turns at least to the primary side than you had before. Try adding about 25% more turns and try it out. If the mosfets still overheat, try adding another 25% and do that up to 4 times, if needed, and let us know the results
 
Where?

Sorry for my stupidity about what u say.. but.. ehm... ;)
Where should be connected the pull-down resistor? :scratch:
at pin 3 of SG3524? Where? And then did they must connected to before or after 4.7ohm resistor (gates is 4.7ohm connection)? :)
"Also make sure your very careful about how you go about finding your center tap" So now my question is.. What's exactly center tap? :)
 
Re: Where?

MaXiZ said:
What's exactly center tap? :)
It located where you have the ground connection at the secondary side. In your case you don't have _real_ center tap. You have two wires which you have tied together, which creates the center tap.

A real centertap has a transformer with only one sedondary winding and at some point is a wire attached, usually at half of the windings. By this you won't get totally isolated secondary windings like toroids usually have.
 
Heating Problems

If the MOSFET turn on time or rise time is slow, the MOSFET will stay in the ohmic region to long. That causes the FET to burn more power in the device lowering the supply efficiency.

You might need a small signal transistor to switch the FET faster.

JW
 
HOW

There are number of methods you can use, a push pull transistor configuration using 2N4401 and 2N4403 at the input of each two MOSFET MaXiZ . You turn them on as fast as possible, this is a command practice to increase the rise time of the MOSFETs in switch mode supply. As you know, you must swap out the gate to source capacitance to increase the speed MOSFET and that takes current.

If you are looking for an application note see the IR web site.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.