VCCINT-0P95V Below Spec

Question

On multiple PicoZed SDR's we are observing VCCINT-0P95V at 0.919V, which is below min spec (0.93V). I traced the problem to excessive voltage drop across inductor E6. (VOUT3 is at 0.964V).

Just an observation so far, we seem to be operating fine despite the low rail.

I would use the ADM1166 margining feature to boost the rail, but I am worried about over-voltaging VCC-BRAM, which is also fed from VOUT3. VCC-BRAM is only exposed on the bottom of the SOM (E5 and C194) in the middle of the connectors, so is a difficult rail to probe.

-Jer

mbrown · Accepted Answer

The DC resistance of E6 is, as you've both identified (thank you!), turned out to be a problem for this narrow tolerance Zynq rail.

In fact our designers recognized a need for higher current capacity for the 0.95V rail and we are in the middle of building prototypes of a Rev D module that removes that ferrite component and increases the capacity to around 6A. The reduced filtering is not expected to have any negative effects on the overall system.

Shorting E6 is the easiest remedy for now. It's a big component and easily found (see below).

/Matt