Hello Lan,1k or 1.015k isn't going to make much difference, it's 1.5% and you'll not get an inductor that accurate.
Going back to the table in post #22, if you want to use a 1k source impedance and a 3.3k load impedance, then you should use 8th row (load/source = 3.33) which gives 0.2447 for Cl and 5.31 for Cc.
That gives an inductance of 1.28mH
I found this one
https://4donline.ihs.com/images/Vip...1-1.pdf?hkey=6D3A4C79FDBF58556ACFDE234799DDF0
and it has a self-resonant frequency of 1.3MHz, which means that above that frequency it behaves like a capacitor.
That means that if you want to remove frequencies above 1.3MHz it won't work, and a single-order RC filter would work better (and an 2nd order RCRC filter would work better still).
I have also calculated and got the same values.
| RL | 3300 | |
| RS | 1000 | |
| RL/RS | 3,3 | |
| L_coefficient | 0,2447 | These values are chosen from the table in the post #22, since I have calculated the ration of RL/RS is 3.3 and n=2 selected. |
| C_coefficient | 5,3126 | |
| L1 | 0,001285435 | 1.28mH |
| C1 | 2,56269E-09 | 2.56nF |

