MP8017 Switching Waveform

DC-DC Power Converters
1

Hi,

I’m using the MP8017 together with CoilCraft AC1584 to create a 8.5V, 1A output (I realize the transformer is designed is for 5V output). Please see attached for the schematic.

I’m seeing that the switching waveform has a small notch before turn on, which coincides w/ the SR turning off. Is this expected, or is there an interaction between the MP8017 and the SR controller (UCC24612). I see this notch even if I disable the SR controller and let the SR FET act as a diode. Is this notch when the converter would normally start to go discontinuous and enter resonance? Is there any way to ensure that the converter is achieving ZVS (the efficiency @ 1A is around 88%, not counting Iq from the PSE controller, PD controller, or output LED).

Additionally, I do not see the snubber switching. The voltage at VSNUB is ~94V and VBUS is 48V. According to the datasheet, the snubber only kicks in when VSNUB-VBUS<1.5V, but how can I achieve this? Do I need a discharge resistor across Csnub? Wouldn’t that lead to lowered efficiency?

Any feedback would be greatly appreciated,
Thanks!

Schematic:

Schematic
Schematic3687×1408 201 KB

Switching waveforms:

48V_8.5V_1A_whole
48V_8.5V_1A_whole800×600 128 KB

48V_8.5V_1A_SRoff
48V_8.5V_1A_SRoff800×600 124 KB

2

Hi Bloom,

I will take a look at the issue and will get back to you soon on this.

Regards,
Adhish

3

Thanks Adhish!
I was able to capture the primary winding current during switching, if that helps. This is with an additional RC across the SR, but the notch occurs with or without the RC network.

48V_8.5V_1A_Ipri_withRC
48V_8.5V_1A_Ipri_withRC800×600 54.6 KB

4

My guess is the secondary side rectifier is messing up. Try just using a diode and see if it goes away

5

Tried that, with no luck.

6

Hi Bloom,

Can you try adding a gate resistance of around 5-15 ohms for SR MOSFET in addition with your RC snubber design? If this does not works, then try replacing the MOSFET with a slight higher Rds_on which can be useful to some extent, however it will be a tradeoff for efficiency.

I would also recommend that you should add a resistor across the SNBR pin and C13 for the active clamp to work efficiently as it shows in our typical application circuit for SSR
image

Let me know if this works.

Regards,
Adhish