I think Johan Liljencrants has won the prize for being the first
person to do this in practice, although he does it in a different way.
He does it by spilling the feeder air back into another feeder which
is opening, netting zero supply.  This, in turn, will reduce the torque
at the crank.

I agree with Mike Knudsen that it may be difficult to maintain
a constant rotational speed so perhaps Johan can tell us how he got on,
or was his organ motor powered?  A skilled organ cranker can compensate
for a two feeder crank giving two 'jerks' per rev so maybe it's not
that difficult after all.  Also, he will know what air demands are on
their way as he will be watching the roll going over the tracker bar
and can be prepared.

I agree with Mike about names for instruments and everybody who
calls a mechanical organ a hurdy-gurdy should be severely reprimanded!
I also get fed up with barrel pianos being called barrel organs.  When
I am out with any of my organs I try to gently educate the public as to
the correct names of our instruments.

When I set the challenge I never thought about keeping the feeder
stroke constant and spilling the contents of the feeders rather than
spilling the reservoir.  This is an elegant solution and probably less
space consuming than doing it with a direct mechanical system.  John
Farmer hits the nail on the head.  Steam locomotive valve gear does
exactly the same thing.  Both common types, Stephenson's and
Walschaert's, use a sliding link, although the challenge is to design
it within the confines of a hand cranked street organ, and to actually
make one.

Art's response shows that its all been done before in an associated
field, and that the added mechanical complexity far outweighs any real
benefit.  I think the same is probably true for hand cranked organs.

Many thanks,  Nicholas Simons, GB