LLC converters (SMPS) do not have zero voltage switch OFF?

[treez]

Hello,
I thought LLC converters were supposed to have zero voltage switch_ON and Switch_OFF of the FETs?

However, as this waveform view of VDS (red) and I(FET) (green) shows, it definitely doesn't have zero voltage switch off.
https://i44.tinypic.com/2w4zqqa.jpg
Why is this?
LTSPICE simulation and schematic attached

By the way, the secondary diodes are zero voltage switched ON and OFF.
-All this refers to operation with F(sw) at the upper resonant frequency, that is, the resonant frequency of the resonant capacitor and the resonant inductor.

 

[mtwieg]

Zero voltage switch off seems fundamentally guaranteed, since at the moment before switch off occurs the drain voltage will, by definition, be near zero (because the switch was on). What would ZVS at switch-off look like to you?

And zero voltage switch off is not really important, since it doesn't save any energy (unlike ZVS on switch on).

 

[treez]

for zero dissipation in switch-off, the fet drain current would decay to zero, and then just after that, the drain voltage would rise to its off state value.....this does not happen with the simulation attached of an LLC resonant converter.
No adjustment I tried gave zero switching loss at switch off.....there is considerable overlap of fet current and fet drain voltage at switch off......it is "hard-switched" off......the beautiful "resonant" converter is not completely resonant......the purveyors of LLC control chips have kept this from us...do you agree?

 

[mtwieg]

for zero dissipation in switch-off, the fet drain current would decay to zero, and then just after that, the drain voltage would rise to its off state value.....

What you're describing is ZCS, not ZVS. To simplify things, we usually about the condition just prior to switching, not after switching. So ZVS is by definition guaranteed at turn off, just like ZCS is guaranteed at turn on.

No adjustment I tried gave zero switching loss at switch off.....there is considerable overlap of fet current and fet drain voltage at switch off......it is "hard-switched" off......the beautiful "resonant" converter is not completely resonant......the purveyors of LLC control chips have kept this from us...do you agree?

In practice it's not possible to get both ZCS and ZVS consistently. It's been found that one can be maintained, and ZVS is usually the better choice, and this is no secret. Any description of LLC operation will start by describing the divide between the ZVS and ZCS operating regions, and demonstrating that they are not simultaneously achieved except at one operating frequency (which varies depending on the load):