The answer is that the second picture was taken with a Type 2+ rotor installed in the permanent magnet alternator (pma), - a Type 2+ rather than a Type 2.
So what is a Type 2+ rotor and why should it increase the power whilst reducing the rpm? The standard rotor for a 42 pole Smart Drive is the Type 2 and it's capable of generating 0.75 W / rpm. The Type 2+ is magnetised to a higher flux strength and is capable of making 1 W / rpm.
For most purposes, which is to say for most operating conditions of head and flow that characterise an installation, the Type 2 will output as much electrical power as the pelton is capable of extracting from the flow available, and it will do this whilst keeping the speed of the pelton at, or near enough at, optimum speed.
But at the very top end of an installation's design flow, when the power put out by the water side of the turbine is greatest, the Type 2 may not be capable of converting all that 'hydraulic-cum-shaft' power to electrical power without rpm's increasing significantly. Not that there's any problem with this: electrical power WILL increase with more water flow because the turbine will simply spin faster - but it won't increase linearly and in the extreme situation power output can actually DECREASE with more flow.
All this is best understood by reference to the following diagram which relates the power output of a pelton to its speed:
In the situation illustrated by the before and after pictures above, the difference between them is represented by a shift of the point on the curve at which the Powerspout is operating from A to B: speed was too HIGH in the before picture, which is to say the turbine was operating to the RIGHT of the optimum speed point, and by REDUCING it (point moves leftward toward Speed/Optimum speed = 1), power output as a proportion of max power increases. In effect, there is an efficiency improvement being gained which produces more watts output.
For those who read my earlier diary entries about Rotor packing, the above diagram also helps in giving understanding there: in that situation, power output was lower than it could have been because speed was too LOW: - the Powerspout was operating at a point on the LEFT side of the curve; by changing the magnetic environment of the pma, (which is achieved by reducing the flux linkage by packing the rotor away from the stator coils), the speed is allowed to INCREASE, and with that increase comes an increase in watts output as the operating point moves rightward up the slope of the hump. On the diagram this is a move from C to D.
For the owner / operator of a Powerspout who really wants to maximise output under all flow conditions by playing around with the strength of the magnetic field, the challenge is knowing what the optimum speed for their particular installation is. For me, and as a result of this experiment using a Type 2+ rotor, I'm settling on a figure somewhere between 950 and 1000 rpm. But take note, this is for my site only; it won't hold true for somewhere else.