Solution Finder Get Tech Support

Blocking diode too hot

Posted by: pablomkz on

I used TOP258YN to design a SMPS, but I have one problem: blocking diode is too hot. Do you have any idea about this question? I just attached designed circuit. Thank you.

Comments

Submitted by PI-Cochrane on 12/11/2012

Could you upload your schematic and point out which part is the blocking diode.

 

thanks and best regards

I just attached schematic; I was measure leakage inductance and it´s whitin design´s margins. I hope that you can help me. Thank you.

I just attached Vdrain waveforms measured at Vin=100Vac (FLV6012_1) and Vin=220Vac (FLV6012_2).

Temperature of blocking diode is 110ºC, but when I putting within box it´s over 190ºC.

 

I´m sorry, that´s waveforms was measured at full load (I = 5A).

 

Submitted by PI-Cochrane on 12/12/2012

I guess you mean MBR10100 is too hot for your design. For a 5A 60W design, it is not surprised to me that the output diode suffers a lot from thermal, since the conduction loss alone should be at least 3W. To solve this issue, you may have to use better diode with lower forward drop volateg, several diode in parallel and bigger heatsink.

 

regards

 

Submitted by pablomkz on 12/12/2012

Blocking diode is D1 (FR106)

Submitted by PI-Cochrane on 12/12/2012

The thermal issue of DI should be caused by the reverse recovery time. PLS try some new diode with faster reverse recovery time.

 

regards

 

Submitted by pablomkz on 12/13/2012

It´s an ultrafast rectifier: UF5408; another sugestion?

Submitted by PI-Cochrane on 12/13/2012

If the reverse recovery and leakage inductance are well controlled, PLS try to use higher damping resistor for the RCD snubber. PLS try a 47ohm  resistor in series with the snubber capacitor first.

 

if the thermal issue can not be solved, you have to reduce the leakage inductance, and improve the thermal condition for the block diode.

Submitted by pablomkz on 12/19/2012

In reply to by PI_Crusher

The 47 ohm resistor in series with the snnuber capacitor appears to solve the thermal issue; I just doing some measurements and I´ll comment later. Thank you for your help.