We make a little 4-channel ARB that works fine, but a customer wants to use it in some RF kind of application, and he's seeing spectral spurs in the -75 dBc sort of range, not too awful for an ARB but terrible by RF generator standards.
Turns out there are some mathematical DDS artifacts but the bad boys are switching power supply noise. The main culprits are a couple of LTC3411 synchronous switchers. I'm seeing big spectral lines at the
5th and 7th harmonics of the switching frequencies, practically nothing at the fundamantals.This is not the first time we've had sync switchers make a lot of noise. LM3102 was incredible. There's shoot-through current and sometimes step-recovery effects on the substrate catch diode, which is still needed in a synchronous switcher. The noise can be so mean that it spreads everywhere.
I decided to measure the supply currents as a preliminary to redesigning the power supply. The idea is to measure the DC drop across the tiny inductors, in the presence of all the switcher flailing. I made this probe-filter thing out of q-tips and junk.
and I made this to help solder stuff.
I think I'll use some LTM8023 modules next spin. They are very quiet.