I have no idea really, try setting the Travel Mode to can go down only. If you
know how, you might want to check the responses the web server gives for
requesting the files, what is the date it gives for the file? (telnet to the
http port of the server and issue the GET command manually, I'm afraid I
cannot give more precise help than this)

A misconfigured web server might give the revision time for the file as
current time, whether this would be a misconfiguration or done on purpose hard
to say.

This would cause Httrack to download the file always again because it would
seem that the file has changed. If there is no way to limit the download based
on filters or depth, you might be out of luck.

Sometimes I find that in order to download a certain site I have to enter
multiple subdirectories into the URL box instead of the server root, because
of the way the site is built - much more easy than writing separate scan rules
for all possible instances.

Very often 'when there's a will there's a way', except with the sites using
URLs obfuscated by complex javascript functions. Wish someone will tackle
these some day.