I ran a PD9160 charging test with the real battery:

I wish the converter-to-battery wiring was beefier to make for a tougher test. But for this test bed, the converter performed as one would expect it to. Once the current got down to 58 amps the converter voltage was over 14.3, and stayed there for 3-1/2 hours as the current tapered. There's no "falling off the cliff" as seen in Salvo's data on page 11. Fisherguy's data on page 11 is exactly what I'd expect if the AC voltage was more than a few volts under 120. KendallP's data on page 13 is pretty similar to mine, just a bit weaker, once again what I'd expect if the AC voltage was below 120. I see nothing in any of the examples, other that Salvo's, that the converter behaves differently with a battery vs. a resistive load. It would be interesting to see somebody else re-run the data with a verified 120 volt power source.
From the Vconv vs. Vbatt spread, the calculated wire resistance is about 16 mohm. Based on the wire size and length, it should be at least a few mohm less, including the circuit breaker and battery relay loss. It looks like further investigation is in order here.