> (files on the same page)
> <http://www.phoenix-mt.com/english/Products/SightGlas>
> sLevelGauges/18-Datasheet70012.pdf
> <http://www.phoenix-mt.com/deutsch/Produkte/Schauglas>
> anzeiger/17-Datenblatt70012.pdf

Telling us the links aren't helpful. Tell us the page the links are on.  Also
what version of httrack. Any non-default settings.

> Other example:
> ...
> href=http://cms.hlplanar.de/data-live-planar/docs/p
> df/Datasheets_eng/NS-25DQL2-IXAeng.pdf" ...
> ...
> href="../../../cms.hlplanar.de/data-live-planar/docs
> /pdf/Datasheets_ger/NS-25DQL2-IXAger.pdf" ...
> 
> The first one wasn't downloaded, while both file
> names looks vary similar.

I assume you have options -> Links -> Get non-HTML = checked or added +*.pdf
in the filters.

If the link inside javascript, options -> Links -> Attempt to detect =
checked. Otherwise perhaps the simple JS parser can't handle this.