source.f & mgdraw.f - input and output my own particles

Dear Fluka experts,
I'd like to "include" fluka inside a kit of programs that we use to
simulate our beamline. To do so, I must be able to use as an input an
ASCII files which describes our protons, which structured in this way [6
columns]: x (micrometers), y, z, px (gamma*beta), py, pz, with z as the
longitudinal direction of our beam.
To do so, I tried to learn from manual and previous question in this
mailing list and I prepared the source.f attached.
In my initial tests, I'm using very small files with just 997 protons,
but this number will grow to maybe one million. So I must put just one
particle at a time on the stack. I hope that a source.f done in this way
is correct. Also, I put a START 997 scorecard in the input file.
At the end of the interaction of my protons with this small slab of
material, in which they will not be stopped when we will finalize the
thickness, I will have to obtain a similar file with the new positions
and momenta of just the protons (if not position and momenta, position,
angular direction and energy can be of course accepted, anyway the full
phase space), so I though that I have to use mgdraw.f.
I tried to write that too, and enabled through the USERDUMP scorecard.
It compiles and runs, but unfortunately it's just writing a lot of trash
on the final phase-space file. Also, I'm not sure if I'm reading my
particles correctly.

Any help will be very well appreciated.
Best regards,

       Stefano

ps: i put a .txt extension at the end of the filenames attached to
prevent a problem with mimetype I got in the past. You have to rename
them and remove those .txt of course if you'd like to test them.