I think the design of the flapper valve is sub optimal.  If you imagine the
flapper position under full flow it would be half way between the two.  Fron
the pics it would seem that the walls and the flapper in this position form
a restriction.

I am guessing enlarging the "outflow" portion should increase flow
dramaticaly.

FWIW


> Last year I tried an experiment placing two EWP's in series, and found
that
> as expected there was a ~30% decrease in flow when only using one pump and
> flowing through the unpowered pump (versus a single pump installation) and
> an ~50% increase in flow when using both pumps. (data is from my memory).
<Snip>