DBM, you might try turning the elbow 90 degrees , that might help balance the air flow through the throttle plate ,  top to bottom,  as the flow tends to hug the outside curve of the 90 and the roof of the intake passage , more so at lower throttle setting when intake velocity is lower and atomization of the fuel on the short turn is much lower ?  That or adding a straight section of intake between the filter and carb ,either might allow easier jetting /smoother running lower speeds?

 The air mass inside the intake pipe , may be slow to accelerate ,if the pipe is oversized . Use of a smaller pipe should increase velocity at lower throttle openings , that may be causing a lean condition , but may limit performance at higher rpms , but if DBM is turning rpms over 8000, he may well be able to except  that loss to have a smoother running motor.

Batman, the Fastman is correct.  The new filter is located in the airbox.  The flow path is similar to the stock airbox setup.  Air enters the filter element, turns vertical and moves up through a 2 inch 90 degree elbow, then flows horizontal straight through a 2 inch hose into the carburetor.  The flow path is almost identical to the stock airbox except it's smoother, no ridges, ledges, edges, etc.  The idle mixture port and transition ports are in the top of the carburetor bore, just like the stock carburetor.  That would confine the air mass to the side of the butterfly where the idle mixture and transition ports are located during high-flow conditions (WOT).  At low flow conditions, the air mass is sort of piling up against the throttle plate.  I don't see it being much different than the stock geometry, but who knows?

Lancer I know my pipe is less than ideal and a 1.62" would be better, but out here in the middle of the ocean, custom exhaust headers are few and far between.  I have some plans for diagnostic testing that should help me narrow the problem down.  Would you agree that if I installed the stock header and the light-load issue went away then its a pretty clear indication that the exhaust pipe might be my problem?....or.....If I installed the stock carburetor and the light-load issue went away, wouldn't that indicate that the S&S carburetor might be my issue?....or....If I installed a carburetor more suited to a single-cylinder engine (like a VM) and the light-load issue went away, wouldn't that tell us something?

Regarding your questions about RPM and throttle position.  It seems to be more load dependent than rpm dependent.  It only occurs when the load is extremely light.  It's most pronounced on the freeway, where you can get the thing into a situation where you are running completely steady-state at extremely small throttle opening.  Couldn't pick a more irritating condition.  The AlF goes waaaaaay lean.  Open the throttle just a hair and start to load the engine and the mixture returns to 11-12 and away you go.  Put any load on it at all and the issue clears right up.  It's right where the throttle plate is sweeping across the transition ports.  Go WOT from any speed or rpm and it pulls great, clean, nice & rich, all the way to 8K+.

For now, I'm staying away from the diagnostic header change.  The way I have my exhaust set up it's a royal pain to separate the header from the muffler, so I want to try some other stuff first.  This is a good exercise.  Frustrating but very interesting.

I appreciate all the comments.  Knowledge is power.  Stay tuned.  

The ID of the stock airbox nipple is 2.260".  My 2" ABS elbow has an ID of 2.030" so its just a tad smaller than stock.  My carburetor inlet has an ID of 2.190".  I would prefer having a converging inlet track starting from about 2.5" at the airbox and reducing to 2.2" at the carb inlet, but you takes what you gets.  The selection off the shelf at the hardware store is limited.

I suspect my issue is more closely related to carburetor application.  All along I knew that installing a fixed-venturi carburetor designed specifically for a large displacement V-Twin would be a challenge.  I'm actually surprised that I got it as good as it is.  I'll leave the intake tuning to the scientists but I'm sure it wouldn't hurt to study up a bit.  

Thanks for the suggestion.  You got any links to reference material (not on the grooves but on the intake tuning)?  

   Perhaps telling of my setup may give you a better idea of what can be done for intake tuning, my bike is stock except for a Dyna muffler and my tuned intake . I based my intake length on the rpms found for 17 inches.  The actual length being 16 1/4 inches as the length of the intake needs to be shortened by 1/2 the diameter of the ID of the tube ( 1.57 inches in the case of my 2inch PVC s-40 pipe ) , as the pressure wave has to leave the end of the tube before it reverses and moves back into the tube toward the carb.
      based on this the numbers (ballpark) look like this ;
 
                         start

   4% boost    4048 rpm/61mph
                                         max **
                                  4308/64.5mph
                                                         end
                                                     4527/68mph

   7% boost   5192/78mph
                                 5577/83.5mph
                                                      5934/89 mph
                                                         ***
  10% boost  6911/104mph  
                                   7682/115mph
    ****                                              8386/126mph( 8000/120mph)


 **  boost is not neck snapping it eases in and out ,peaking at the mid range

***  the numbers are ball park , as my bike being stock , is only able to  use the 3rd and 4th waves , but top speed is 90mph -real world speed (6012rpm)

**** You should be able to use these numbers (2nd wave)  the speeds shown are more the type you would see on a dyno , not on the road with rider weight and frontal area wind drag.  your results my vary.

 The pressure waves could care less about the diameter of your pipe only it's length. But the pipe you use should be sized more toward the size of your throttle plate rather than the mouth of the carb ,ex; my cv carb throttle opening is about 1.5 inches in area . my PVC intake 1.57 inches in area.  I believe what your using is to large and killing the velocity in your intake , it is better to error on the small side as you will see gains in the low to midrange rpms and only a slight loss at high rpms, if to large you won't see any gain across the whole rpm range.

   All rpms above are in 5th gear ,as you can see these match up fairly well with my intend highway speeds , If I need more speed say entering an expressway on ramp or passing leaving the bike in 4th gear or down shifting,  will bring  the boost on at lower speeds ex;
4th gear -7% boost -
                         5192/71 mph
                                          5577/77.2mph
                                                               5934 /82mph
        4th gear   71 to 82
       5th gear         78 to 89    
              So traveling a 70mph  needing to pass, would be a down shift to 4th and upshift to 5th @ about 80 to stay in the 7% area of boost from 71 to 89 mph .
           I estimate to achieve 90 mph top speed I need to produce somewhere between 33and 34 hP , about equal to a stock bike with a DR650 cam ,without having to buy or install one , and yes my tuned intake is also inside my stock air box, use of street elbow ,and short section of pipe allows me to retain the rubber tube between the box and carb , I moved my Oldfeller's style filters to the outside rear of the box, between the frame posts just in front of the rear fender , retaining the side covers and the bike appears completely stock .

DBM, is your definition of “entire power band”  based upon rpm or throttle position ?

Lancer, my evaluation of the power band relates to engine rpm.  If I open my throttle anywhere past 3000 this thing takes off, and just keeps accelerating hard all the way past 7000.  It just seems to have killer torque everywhere.  It runs just fine below 3000 too, but I'm very rarely in that rpm zone.  

I'm very surprised how broad the useable power is.  I figured the long duration cam, with 5 degrees of timing pulled out, would make the thing a peaky.  I'm OK with peaky if there's a huge kick to put a grin on my face.  But this thing is as far from peaky as it gets.  I have to assume it's the compression, but I also think that the epoxy buildup on the floor of the intake port has something to do with it.  I think this big filter unleashed the beast and let the other components do their job.

It just has this irritating little glitch right at the edge of the transition ports, right where my throttle plate lines up exactly with the 3rd port.  It's drivin me nuts.

Batman, before I start messing with intake tract tuning, I've got to figure out this little hicup.  Every day I mess with it I become more convinced that its related to the application.  That carb was specifically designed for a large displacement V-Twin.  The transition ports are manifold vacuum dependent.  They don't care about flow because they are well forward of the venturi.  The big cam with retarded timing killed the little bit of manifold vacuum that this single-cylinder engine develops.  I think I have met my Waterloo.