CD is fairly precise for a-helices of moderate length (> 5 I think). It is less accurate for short helices and other types of helices besides alpha such as 310 helices.

To get an accurate CD measurement it important that the concentration be known exactly and the absorption from 190 nm onwards is in the detectable range.

Dear Peter V. Dubovskii ,

Indeed the 220 nm estimation method is a rough method and mostly used to give an indication of the relative changes between different mutants or different circumstances (with or without anionic phospholipids for example). See for an example:

Article Anionic phospholipids are essential for α-helix formation of...

To check whether the CD analysis makes any sense, the sequence can be used for secondary structure prediction:

Like SOPMA:

https://npsa-prabi.ibcp.fr/cgi-bin/npsa_automat.pl?page=/NPSA/npsa_sopma.html

gives 46 % alpha helix.

Or Porter5:

http://distilldeep.ucd.ie/porter/

gives 77% alpha helix.

While JPred4:

http://www.compbio.dundee.ac.uk/jpred/

Gives 65-73 % alpha helix.

A little comparison made by myself indicated that SOPMA sometimes underestimates the helical content, see for more details (Table 2):

Article The role and significance of potential lipid-binding regions...

Anyway for those still interested in these type of matters.