LTM4686 datasheet shows two parameters, TOFF_DELAY and TOFF_FALL. TOFF_DELAY allows a user to specify the delay between the RUN input signal going low and the actual beginning of the (channel) output turning off. TOFF_FALL allows the user to specify the amount of time for the output to actually fall from the controlled output level to 0V (in increments of 100us).
I'm trying to simulate this behavior in LTSpice (latest and greatest) but I'm not seeing what I expect. What I see is that the model responds to the TOFF_DELAY parameter as expected where a delay happens between the removal of the RUN signal and the turning off of the output. However, I expect the TOFF_FALL to operate like a "soft stop" (as opposed to a "soft start") where the output gradually tapers to a 0 voltage / 0 current but what I see is that it seems to behave as a simple additive delay. For example, if I set TOFF_DELAY to 1ms and TOFF_FALL to 1ms, the output voltage is maintained for 2ms after RUN is brought low. If TOFF_FALL is changed to 5ms, the output voltage is maintained for 6ms. In either case, the output current simply goes from full on to zero at the end of the delay and the output sags according to the simulated load (e.g., in this case a 110milliohm resistor on each rail).
The simulation is set up with a 5V input source with no source impedance, 900mV parallel output with the two rails joined by a 1 microohm resistor, 200uF decoupling capacitance on each of the two rails and the various parameters adjusted to simulate the desired operation. Here are the two SPICE lines:
Vout_0=0.9 Vout_1=0.9 Ilim0_range=1 Ilim1_range=1 OC_limit0=1 OC_limit1=1 Mode_ll=2 Fault_response=0 Retry_delay=.1m
Freq=750K PHs_0=0 PHs_1=180 Ton0_delay=0.3m Ton0_rise=.5m Ton1_delay=.3m Ton1_rise=.5m Toff0_delay=1.0m Toff0_fall=5m Toff1_delay=1.0m Toff1_fall=5m Vout0_range=0 Vout1_range=0
This is a (more) complicated beast, IMHO, so I'm hoping that it's simply a matter of my changing the parameters to get it to work; I allow for the possibility that the model is unable to simulate this "soft off" operation...?
Thanks in advance.
