Hello,
I'm trying to evaluate components for a ADP5050, but I can not get coherent values for the compensation networks.
The design tool provides values that can not be retrieved using formulas from the datasheet, by far. It seems to be related to the output capacitors and the way they are used.
It is not clear if there is a derating of capacitance values in the computation. On the other hand, if one use small ripple, the values of the output capacitors become "ideal" and much smaller than "standard" values provided for lower ripple.
For example, using buck1 and 2 in parallel for 3.3V/8A at 300kHz :
Using 0.1% ripple propose 6 100uF capacitors (3 per buck) and the compensation network is 1.2nF/34k. I can not get these values using the formulas from the datasheet using frequency values from 1/ to 1/12 of the switching frequency.
For 0.025% ripple, the capacitors become two "ideal" values of 102.4uF (!) and the compensation network is 1.2nF/26.1k. Which makes no sense (less ripple with less output capacitance !).
This does not match at all the proposed values from table 17 of the datasheet.
It is more confusing when reading examples from the datasheet where the same compensation network is used for buck 1 and 2 with different output voltages (but same capacitance and current) - see for example figure 64.
At the end, I'm not able to evaluate these network as all sources are incoherent or contradictory. In addition to that, there is nothing about this when using buck 1 and 2 in parallel. Is the network is for both (8A and all capacitance) ? It is not said.
So, what is the correct way to compute this ?
Thanks in advance,
J.Prévost
