N119ST.  Here are a couple of flow visualization pics. Hard to see, but, (in the pic on the left) the flow in the valley between the spinner and the inlet is going down a lot higher than I expected.  Likely do to the prop/cowling interaction.  In the 3rd prototype, I moved the high spot down an inch.

With the exception of the recirculating flow in the corners of both inlets, there is no flow separation on the entire cowling.  The only way to cure the flow separation in the inlets (and, it isn't much) is to install flow splitters and move the flow in stages around the corner of the inlet.  I could probably get away with smaller inlets if splitters were added.  Lotsa work for a few percent increase in efficiency.  See the installed pic below.

---