Spartfxr, I think you should re-time your DR cam to the factory marks.  The DR cam closes the intake earlier than the stock cam.  Advancing the timing aggravates that condition.  Closing the intake so early is limiting your powerband, causing the engine to fall flat on it's face at 4500 rpm.  

With the DR cam, my engine pulls hard all the way to 6500 rpm.  The cam is set to the stock timing marks.   Check the dyno curve I have attached to this post.  You will see that the DR cam results in a nice broad power band that does not sign-off at 4500.  

I wouldn't use the ProCycle timing data (that you discussed in an earlier post) because the cam they offer is for a 1996 and later DR650, which uses a different cam (it won't fit our LS650 engine).

I procured two more Web cams, the 402 grind and the 340b grind.  They are both aggressive camshafts and are not bolt in cams.  As a minimum, they will require cutting/shortening the valve guides to achieve adequate clearance between the spring retainer and guide seal, and installation of stiffer springs with more available travel.  The stock springs approach coil-bind at the higher lifts these cams provide.

There has been some discussion recently about the merits of the higher lift.  I thought I would elaborate on that in this update.  There is a rule of thumb that maximum flow is achieved once a valve reaches a lift that equals 25% of its diameter.  Our intake valves are 1.3” diameter, so that rule of thumb would dictate that max flow should be about .325” lift.  My flow bench tests confirm that.  The flow starts to flatten out at .325” lift.  It increases a bit from .325” up to .400”, but it’s a very shallow curve.  I think its safe to say that the rule of thumb is accurate.  You’re not gonna gain much by opening the valve farther than .325” (25% of valve diameter).

The benefit of the additional lift isn’t the increase in valve opening, it is the length of time (in degrees crankshaft rotation) that the valve is at or above the max flow lift.

It would be nice if the cam simply held the valve at .325” for a longer period.  That way, you wouldn’t need to fiddle around with the guides, and the stock springs wouldn’t go into coil-bind.  Alas, its not that simple.  The cam must incorporate acceleration & deceleration ramps.  The rocker arm and valve motion just can’t be stopped abruptly.  Just as you can’t allow the valve to slam into the seat when it closes, you also can’t expect it to change direction abruptly from opening to closing.  The cam must start slowing things down as the valve approaches full open, otherwise the rocker arm will loft off the cam lobe and you will lose control of the valve.  In order to keep the valve at max flow lift for a longer period, the cam has a deceleration ramp.  The deceleration ramp is in the portion of the lobe profile that opens the valve past max flow lift.

Here is the Web 402 timing data taken at the lobe.

Here is a graph of the Web 402 intake compared to the stock DR650 intake.

Here is the Web 340b timing data taken at the lobe.

Here is a graph of the Web 340b intake compared to the Stock DR650.

I think these two additional cams provide us with some interesting options.  They are not bolt in cams.  The cylinder head must be removed in order to install either cam properly.  All clearances must be verified.  They will require some minor machining and the added expense of stiffer springs.  In addition, the DR650 spring kit is not directly applicable to the LS.  The spring retainers and cotters can't be used on the LS.  With a little research and patience, I'm sure the problems can be solved.

The timing and lift data I have provided should be helpful if you are considering a big cam upgrade.  I hope you find it useful.

Knowledge is power.