Long Term Trim -100%

[Robert King]

I have been running a 14CUX on a 3.5L; when I run the NA catalyst map (with O2 feedback) on the R3526 chip, the long term adaptive on both banks show -100% with the short term switching from +8% to -30% at idle. Pulse width is 1.9 to 2.1 ms; the LR manual indicates normal is 2.4ms. This indicates to me that the engine is running rich and the ECU is trying to correct for the rich condition. Probably the-100% long term is because the ECU cannot correct enough above idle. 
Fuel pressure at idle is 30psi, and 38psi with vacuum disconnected. Manifold vacuum at idle is 18”. TPS is showing 6% at idle, and AFM is at 21% at idle ( actual). These measurements are via Rovergauge. I have tried disconnecting the purge and PCV with no significant change. Engine runs well, but fuel consumption seems a little high. 
I am running Ford FOTED5A injectors, that supposedly have a similar flow rate (19#), but with better atomization. I have tried two AFMs with similar results. Any ideas? Could the long term reading be a glitch?