Post Go back to editing

LT8616 Simulation at low current load not plausible

Hello
I want to use the LT8616 dc/dc-converter to power my application circuit.
The application needs a 3.3V and a 5V power supply, with an input voltage between 20V and 38V.
The normal load current is between 75mA und 800mA.
I went through the data sheet and tried to avoid the discontinuous current mode as well as I tried to reuse components, which are already implemented in any other circuit.
For the inductance I wanted to use 47µF, and for RT I wanted to implement 27k (-> f = 1,3 MHz) or maybe 22k (->f = 1,5 MHz)
in my Calculations, my typical current ripple (delta I) should be 42 mA at maximum input voltage (which would be far away from discontinuous current mode)
But in my LTspice Simulation, even at 24V input voltage @ 100mA load current, the converter is in discontinuous current mode, and the frequency of the current ripple is not plausible too.
I can not explain this behavior. With a load current of around 1A the simulation is pretty close to my calculations.
Can you help me, do i have a mistake in my Simulation?
  • Hi ,

    SYNC/MODE: Ground the SYNC/MODE pin for low ripple Burst Mode operation at low output loads. Tie to a clock source for synchronization to an external frequency. Apply a DC voltage of 2.4V or higher or tie to INTVCC for pulse skipping mode. When in pulse-skipping mode, the IQ will increase to several hundred μA. Channel 1 will align its positive switching edge to the positive edge of the external clock and channel 2 will align its positive switching edge to the negative external clock edge. Do not float this pin. 

    If the SYNC is tied to ground, the circuit runs in burst mode.  Fig 8616 G19 shows the burst frequency versus load current in a specific circuit setup.   

    1) Your simulation need SYNC to be set properly. 

    2) In light load conditions, the real frequency of the converter is not the preset frequency unless it runs in forced continuous mode which is not the case.  Your ripple calculation is therefore not correct.