• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 41 subscribers
  • Views 234 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

PMP22764: Mosfet Losses

d_zero
Genius 4470 points
Part Number: PMP22764
Other Parts Discussed in Thread: LM5156H, CSD19531Q5A, CSD19534Q5A

Dear All,

we are designing similar flyback based on PMP22764 with LM5156H IC.

e2e.ti.com/.../3929193

I this e2e thread is stated that the PMP22764 was designed for best efficiency, but we would like to have better so i did a calculation with CSD19531Q5A ( currently used on PMP22764) and CSD19534Q5A and i get approx. 1W difference, please can you look at may calculation if i'm missing something? Also i used Schottky diode on the other side STPS10200SF .

Transformer: ZB1368-AL Coilcraft

Vout= 55.75V

Iout_max = 1.2A

Vin from 18V to 30V -> Typ 24V

My calculation

Vcc 10 V External from Transformer Aux Winding CSD19531Q5A
Isource 1.65 A Figure 8-14 Peak Driver Current vs VCC Qg 37 nC
Isink 1.75 A Figure 8-14 Peak Driver Current vs VCC Rdson 5.3 mOhm
Coss 560 pF
Idriver 1.7 A
tr 22.42424242 ns Rise time EQ5
f_sw 252.5 kHz tf 21.14285714 ns Fall time EQ5
I_rms 4.85 A EQ 1
V_ds 58.2 V EQ 2 Rdson_factor 1.75
I_peak 7.56 A From the SNVC240 calculator
P_cond 0.218171188 W EQ3
P_sw 2.659586246 W EQ4
P_tot 2.877757434 W
Vcc 10 V External from Transformer Aux Winding CSD19534Q5A
Isource 1.65 A Figure 8-14 Peak Driver Current vs VCC Qg 17 nC
Isink 1.75 A Figure 8-14 Peak Driver Current vs VCC Rdson 12.6 mOhm
Coss 257 pF
Idriver 1.7 A
tr 10.3030303 ns Rise time EQ5
f_sw 252.5 kHz tf 9.714285714 ns Fall time EQ5
I_rms 4.85 A EQ 1
V_ds 58.2 V EQ 2 Rdson_factor 1.75
I_peak 7.56 A From the SNVC240 calculator
P_cond 0.518671125 W EQ3
P_sw 1.221844923 W EQ4
P_tot 1.740516048 W

Would you suggest CSD19534Q5A  over CSD19531Q5A , or i'm missing something in the above calculation?

Br,

David.

  • 0
    TI__Mastermind 40100 points

    Hi David, the 1W savings will be ~1% savings in efficiency at heavy loads which is possible and what I've seen in the past. The tradeoff with the faster FET is you might have a larger overshoot at the rising edge so might need to increase the RCD clamp. It might end up less than 1% improvement. 

    For the rectifying diode on the secondary, I didn't see a reverse recovery spec in the datasheet. You typically want a fast recovery time if trying to optimize efficiency. You may want to test the new FET first, then see if it improves performance. Then you can test with the new schottky and test performance again. Thanks!

  • 0 in reply to Darwin Fernandez
    Genius 4470 points

    Hi Darwin,

    thank you for the replay.

    1)

    please let me know if i understand correctly. For given transformer i have leakage inductance Llk and primary side parasitic capacitance Cp and if i choose a faster FET with lover Qg i will save on switching losses, but the down side is that the FET with lower Qg also has a lower Coss and i will get larger Vds spike per down equation:

    and thus i have to increase the RDC clamp snubber, to have same overshoot , and then i have more losses on RDC clamp is that correct?

    2) Isn't this an advantage of Schottky diodes that they don't have reverse recovery time? No reverse recovery time means less losses in CCM Flyback?

    In the DS of STPS10200SF page3 figure 4 

    Qrr=Vr*C ~ 170V*50pF = 8.5nC

    How can we compare Qrr STPS10200SF  of to trr of PDU620-13

    Best Regards,

    David.

  • +1 in reply to d_zero
    TI__Mastermind 40100 points

    1) That's correct, I've done something similar in the past and saw a slightly higher lead edge spike but because I didn't want to increase the snubber, I chose a higher VDS rating of the FET.

    2) That's right the schottky will not have reverse recovery (sorry was thinking about the STTH family while this is the STPS you're talking about). Just note the schottky will typically have higher capacitance where you might also need to increase the snubber.

    In the end, I believe your selections are okay and this will be a matter of testing on bench.