The AEM documentation also recommends that you tilt the tip of the sensor down a few degrees so that deposits don't form on it. So when you weld on a bung, it should be at least someplace where the downpipe is not vertically oriented. Food for thought.

Just philosophizing...-> It seems strange that moving the sensor that distance would make a noticable effect on MS monitoring. It would think that the high velocity of the exhaust gas would minimize the impact of such a small distance. Also, I wonder just how much EGT goes down after 36" of travel. Every recommendation just seems so arbitray - how can we know what is optimum with so many variables that come into play?