Polypeptide alpha-helix with Amber 3.0

[apa ^at^ ccadfa.cc.adfa.oz.AU (Alan P Arnold)]

Thanks to all those who replied to my earlier request for
 creating MIN coordinate files from MD trajectories.
 Now, another problem.  I have created an 8-mer of Alanine
 with neutral terminii, using Amber.  The starting 'conformation'
 created by LINK/EDIT/PARM is an extended chain.  I want to
 get the (Ala)8 into an alpha-helix.  Naively, I thought that
 using constraints on the H-bond distances between the ith NH
 and i+3'th carbonyl-O (of about 2.0A with 'energy' of 10kcal)
 should spring the peptide pretty close to alpha-helical.  No way.
 Further contstraining the H-bond angles to near 180deg as well
 as H-bond distance constraints still doesn't work.  All this
 'constraining' is done in PARM.
 I am obviously doing something wrong here.  How do I force such
 a peptide into an alpha-helix?
 A secondary question obviously arises - what is the likelihood
 that MIN (or any minimiser) will find the alpha-helical conformation
 starting from an extended chain?
 Thanks again.
 ----
 Alan Arnold             	     |  Phone: +61 62 68 8080
 Chem. Department,University College  | ACSNET: apa ^at^ ccadfa.oz
 Australian Defence Force Academy     |   UUCP: ...!seismo!munnari!ccadfa.oz!lpb
 CANBERRA  ACT 2600 Australia         |   ARPA: apa%ccadfa.oz ^at^ SEISMO.CSS.GOV