In retrospect the previous blog, posted about a year ago, shows how naively I approached the problem. The correct calculation, done around September gives a 1/4 correlation in certain configuration. The paper was written with QuTools, Munich based entanglement experts. The experiment was run by QuTools in March in Germany and gave us some interesting anomalous results which required several rounds of debugging and interpretation.
By inserting only one QWP in one arm we detect phase effects with great sensitivity. A modulation in the signal correlation, otherwise computed to be 1/4 still remains. The updated interpretation puts the blame on the source itself and that the pumps would not be perfect resulting in the amplitude and phase we see and detect. This was double checked with a full state tomography on the source and seems to give results that roughly agree with what we measure in the simpler QWP experiment. We therefore have a way of characterizing the entangled sources in a simple way.
I have received the system after much hassle with the customs. And train myself on it.