I don't know how many elements are in the transducer. It has 21 pins so there are at least several. The coverage area displayed is not a cone. Each element, if pinged separately and analyzed separately would have a traditional cone. It's not done that way. The timing of the firings produces multiple cones. The result is a coverage area. Now I am about to get away from facts and into more subjective conclusions based on observations. I do not know the shape of the coverage area but it varies with both depth and range. Garmin has disclosed that there is a weak area at a range equal to the depth and a little farther. We can use the traditional 2d cone concepts as a basis to analyze what this means. The spec of 20 degrees for a 2d cone is measured at a specific strength. If you half that strength the cone will not be 20 degrees. A weak area in the range of the LVS would mean that the coverage area at that distance would be smaller. I have confirmed to my satisfaction that there is not a dead zone and posted the methods used on this board and performed the experiments using ordinary materials and methods that anyone can replicate to either confirm or challenge my results. What these experiments did not show though was how wide the coverage area is at the different distances. I have observed that the width at 60 ft is wider than it is at 30 ft when in 30 ft of water. The pointing gets so delicate at some distances that even the slightest rotation makes the target disappear. That's the weak area. I doubt if most mechanical or electrical rotation system can produce the finesse necessary to stay on target. A hand rotated system would. The Force will, when pointed forward. It's a design feature of the rotation. When the target is at that weak distance, it will only be picked up dead center. I have been experimenting lately with attempting to accurately measure the coverage area at different ranges. That's still a work-in-progress and I have no idea when I will be able to report results. This is such a controversial topic I don't want to get it wrong and I have run into several complications.
I do not understand since it is so widely reported on here that v2.20 was so good why people don't just roll back to that version. I don't agree that v2.20 is any better than v2.41 but that's not relevant to the point. I have confirmed via Garmin engineers that if you roll back to that version that there are no remnants of the later version hanging around. Just don't expect canted performance from a non-canted mount.