Heatsink for DPAK

[Eeyore]

Then I think it's gonna be something like that :

fs = PWM freq.
tc_on = switching time off-on;
tc_off = switching time on-off;
T = PWM period;
t_on = time when the PWM in high;

P = 1/2*(Rds_on*I^2)*fs*(tc_on+tc_off)+Rds_on*I^2*(t_on/T);

Correct ?

No. During what you called tc-on and tc_off the device is operating in linear
mode. You'll need to calculate those losses more accurately as the integral
w.r.t time of Id * Vds

Graham

 

[Pasquale]

No. During what you called tc-on and tc_off the device is operating in
linear mode. You'll need to calculate those losses more accurately as
the integral w.r.t time of Id * Vds

Graham

I've done it. And the result during the switching time is

Ps = 1/2*(Rds_on*I^2)*fs*(tc_on+tc_off).

That result takes into account the integral that you mean, at least that
is what I've understood from a book. Wasn't it waht you meant ?


Thank you,
Paquale.

 

[Eeyore]

I've done it. And the result during the switching time is

Ps = 1/2*(Rds_on*I^2)*fs*(tc_on+tc_off).

That result takes into account the integral that you mean, at least that
is what I've understood from a book. Wasn't it waht you meant ?

I still don't follow your equation. It looks from first glance to be
dimensionally incorrect.

Ignore the stead state dissipation, that's simple enough. How exactly did you
calculate the dissipation during the switching periods ?

Graham

 

[Pasquale]

I've done it. And the result during the switching time is

I still don't follow your equation. It looks from first glance to be
dimensionally incorrect.

Ignore the stead state dissipation, that's simple enough. How exactly
did you calculate the dissipation during the switching periods ?

Graham

It's not dimensionally incorrect :

- Rds_on * I^2 = Rds_on * I * I = Vds * I = Watt
- fs = 1/sec
- tc_on and tc_off = sec.

so it remains just Watt because the secs delete themselves. I've found it
in my book about Power Eletronics

http://www.amazon.com/Power-Electronics-Converters-Applications-Design/
dp/0471226939/ref=pd_sim_b_title_1

Thank you,
Pasquale.