Question:
I wanted to know that can we have contact between Beam Elements and Shell elements and 
what Card is to be used. I have tried all "CONTACT_AUTOMATIC" cards but was not able to get the 
contact. The part with Beam elements passes through the wall of Shell elements

Answer:
*contact_automatic_single_surface, *contact_automatic_general, or 
*contact_automatic_nodes_to_surface will handle some, but not all, 
beam-to-shell-surface contact situations. 
All of these contact types take into account thickness offsets.  
By this I mean that the contact surface is offset from the shell midplane by half the shell 
thickness and offset from the beam centerline by a radius of the 
beam's equivalent circular cross-section.  
Of course, the contact thickness can be modified by the user on card 3 of *contact...
or by using *PART_CONTACT instead of *PART.
The first two contact types mentioned above are single surface 
contacts and so both the shell and beams parts should be included on the slave 
side with the master side being null.
For an automatic_nodes_to_surface contact, the beam part (or its nodes) should be slave, the 
shell part (or its segments) should be master.  
For any of the above, a search is made for penetration of beam nodes 
(or more precisely, a sphere around each beam node) through shell surfaces.

For beam to shell edge contact, one could use *contact_automatic_general AND add null beams (low 
density beams utilizing *mat_null) along (merged to) the outer edges of your shells.  
Both the beam part and the null beam part should be included on the slave side of 
the contact.  No master side is given.  With this contact type, the contact surface 
is offset from the beam midlines.  See the interaction of parts 16 and 116 in 
http://ftp.lstc.com/anonymous/outgoing/jday/beam-to-hole-edge.k
_________________________________________________________________________

***  NEW CONTACT TYPE ***

*contact_automatic_beams_to_surface is available in 971 R5.0.
http://ftp.lstc.com/anonymous/outgoing/jday/beam_thru_hole.k is
a small example illustrating the contact in a beam-to-solid-surface application.
Unlike *contact_automatic_nodes_to_surface, it is clearly able to detect contact 
anywhere along the beam length.   
I notice that d3hsp calls the contact "a5" contact which it is not (Bugzilla #3666). 

jpd