The LTM8060 Sync inputs are multimode function selectors as well as external clock synchronization inputs. When are the Sync pins sampled? There is no description or spec of timing relative to Vin-on or to RUNn pins. Can we assume that the mode can be changed on-the-fly? When transitioning from any of the three modes (Burst mode, FCM, Spread-Spectrum) to external sync clock, is there a clean ramp from the RT set frequency to the sync input frequency (required to be higher than RT setting)?
I've configured the LTM8060 jig in the LTSpice examples library for 3.3V @ 6A plus 2.5V and 1.8V @ 3A each, circuit below;
The following waveform is a zoomed-in section of channels 1,2 at the end of a 5msec run where a 1.4MHz clock had been applied at 1.5msec following startup. The 3.3V stabilized at 600usec after startup. The green trace is the applied 1.4MHz signal on SYNC12, the pink trace is CLKOUT12, and the grey trace is the ripple current in Rload1. By this point of the waveform, the device had 3.5msec of SYNC clock present.
Three things are noticeable in this capture; first, the load ripple current shows that the Fsw has not locked onto the 1.4MHz and is still only a bit higher than the 1.23MHz preset prior to application of the SYNC clock. Second, the ripple is 2x the Fsw setting, meaning that the two channels are running at 180 degrees to each other. That's not a surprise and is a good thing although there is no mention of this in the datasheet. Third, the two phases appear to be a bit imbalanced from each other. They were nicely even prior to the application of the SYNC clock.
I know that the SYNC input is not being ignored given the following trace capture showing a Rload1 ripple step disturbance at 0.5msec where SYNC12 clocking begins and CLKOUT12 starts sending (I set data storage to 1msec from start, so the clock application occurs 1.5msec from sim startup and nearly 1msec after the 3.3V output stabilized into a 6.6A load)
At this point, either the Spice model of the LTM8060 does not properly capture the chip behaviour, or there are a number of unspecified parameters regarding the SYNC function. I'm glad that it appears that the SYNC mode can be changed on-the-fly, however, I need to know the stabilization time after a mode change, or if there are any restrictions on which mode changes are permitted while the channels are running.
Edit Notes
Some further investigation regarding mode configuration using LTSpice[edited by: RP555 at 10:30 PM (GMT -4) on 29 Apr 2026]
