TOP250YN primary circuit problem

Hi,

Your design is a high power flyback converter.

1. If the power dissipated in the clamp circuit is excessive, that can heat up the diode. the power dissipated in the clamp circuit is dependent on the actual leakage reactance of the transformer primary winding. the leakage reactance of the primary is dependent on the transformer construction. Generally a Sandwich construction in which the secondaries are wound between two halves of the primary should help.

2. Given that this is a high power design, it is possible that the actual clamp dissipation is high and you may need to heatsink the diode.

Can you send us the primary side current waveforms for the operating condition where you see this issue?

Regards

PI-Sarek

Hi Sarek,

Thank you for your reply. I changed the FR106 to MUR4100 and it isn't very hot now but there is a new problem that the 180V/5W zener heats so much that even melt the tin of its pads! According to your explanations I guess that problem may arise from my PCB. I send you the PCB and would you please check it? Also I don't know how to view the current waveform of primary circuit. I guess If I add a resistor to drain to measure the current it affects the overall circuit.

Best regards

Hi,

Please refer to Figure.48 for the recommended layout configuration which is given in the data sheet. Here is the link for data sheet

https://ac-dc.power.com/sites/default/files/product-docs/topswitch-gx_family_datasheet.pdf

Could you please send us your transformer construction document and leakage inductance information.

Could you please measure the leakage indcutance of the transformer which is there on the existing test unit.

If you have current probe, you can measure the primary side current waveforms by inserting the current probe in series.

Could you please capture the primary drain voltage ans send it to us.

 Please try inserting series with primary clamp zener diode D2, so that it limit the current into the zener. You can start with a 1Kohm series resistor.

Resistor R5 you can replace with 0ohm, if you do not have any EMI problems.

Let me know what happens.

Regards,

PI-NANO

hi!
I have a problem with my flyback converter. I use TOP 250 for my design. With one output 5V, everything was ok (with max current 10A), but with one more addition output 12V, if output 5V has load (even with 1A load at 5V), the voltage at 12V will increase to 15V and more. Do you guy explain this.?

Hi Tony,

Fly-backs are known for poor cross regulation.

You can improve the cross regulation effect

1. By using sandwiched transformer construction with AC coupled winding's on secondary,

2. By adding small amount of pre-load on each output

3. By using waited feedback method etc.

Regards,
PI-NANO

sorry it's Weighted feed-back method.

Hi! piano. Thanks for useful ìnfomation , i will try them all. With weighted feedback method, i allready use it in my design. but there is a problem, when outputs have load , both output value are changed :12V íncrease, 5V is dicrease ( to 4,7..4,8 V) . This is not good because i need output 5 V have good regulation. do u have some ideas for this ?

One more question, please.! Can you give me any model or figure about sandwiched transformer construction with AC coupled winding's on secondary,thanks again!

Hi,

If you need tight regulation on 5V output then you can not use weighted feedback method, 100% of the feedback should come from 5V output. However this will result in poor regulation of 12V output.

you can have better regulation on 12V if you have some pre-load, however this would increase your light load consumption.

the other option is to use a post regulator on 12V if you need both outputs tightly regulated.

I would suggest you to look at our InnoSwitch product as well, since this uses Sync Rec, regulation will be much better compared to diode based secondary rectifier.

https://ac-dc.power.com/design-support/reference-designs/design-examples/der-544-40-w-dual-output-power-supply/

https://ac-dc.power.com/sites/default/files/PDFFiles/rdr469.pdf

Here is the design example for sandwiched and AC coupled

https://ac-dc.power.com/sites/default/files/PDFFiles/der299.pdf

Regards,
PI-NANO

Hi, pinano! Regulation on 12V is better if the value of resistor pre-load is decrease, does it right? One more question, how to decrease value of ripple on output ? I try some methods as output filter and snubber, but ripple is still high.

Hi,

Nice to hear that the regulation issue is resolved.

Regarding output ripple, how are you measuring measuring output ripple?

Did you use 20MHz Bandwidth filter?

Did you use short probe?

Do you have any capacitor connected on the ripple probe?

Could you please send me the output ripple waveforms?

Regards,
Sudhakar

hi, pinano. I use short probe, but don't use 20MHz bandwidth and any capacitor on the ripple probe. But now, it is still not problem. I have a serious problem now. I design another power supply with output 5V. But when I connected ouput to load 1Ohm ( 5A ), diode TVS VR2 is died. when I took this diode away, and put load to ouput again, then Top 249YN is exploded. @@ I sent you my design, can you guy give me some help?
For transformer, Primary : 60 turn , secondary 5V: 2 turn, and Vcc is 4 turn.
I guess may be problem is in resistor R7. Does it right?

Can you give me your email?

Hi Tonycnbel,
Can you try adding light load first then increase it step by step (maybe in increment of 0.1A) and check voltage accross Drain to Source of TOP249YN? Make sure that the voltage across Drain to Source does not reach 700V maximum limit. In case the maximum voltage is too high, we need to improve the primary snubber circuit.
Regards,
Jedidiah

Hi Jedidiah!
Thanks for your recomendation. I dont have adj resister so i used Resistor 20W-5 Ohm, and put to output step by step with increasing number of paranel resistor ( decrease value of resistor from 5 to 1 Ohm). About voltage on pin Drain, i think it reached 700V because clamp Diode was burn at 5A. So may be improving snnuber is needed, but how to improve it? can you show me more clearly? And i think about another reason, may be transformer was not good, and leakage induction is hight ( I measure that it reach 30-50 uH), can you give me discussing for it?

Hi...i am a new user here. As per my knowledge if the power dissipated in the clamp circuit is excessive, that can heat up the diode. the power dissipated in the clamp circuit is dependent on the actual leakage reactance of the transformer primary winding. the leakage reactance of the primary is dependent on the transformer construction.

http://percentagescalculator.com/

Please use the link below for additional information about clamp design.

https://www.power.com/sites/default/files/PIU-107_OverheatingComponents.pdf