The concentration needed will depend on the sensitivity of your instrument, and the extinction coefficient of the chromophore you are interested in. Compared to absorbance spectroscopy, CD is a bit more challenging as the extinction coefficients are modulated by the asymmetrical environment (conformation) making them more difficult to estimate (than say for UV/Vis absorbance). But similar to UV/Vis absorbance, you will need to keep the total signal below the non-linear threshold of your photomultiplier (usually
concentration times thickness of the cell is the determining factor which has to be chosen that the extinction is not too high (the allowed value depends on the sensitivity of the instrument used and here on the chosen spectral rergion). Often the given values for the allowed extinction are over-estimated - especially for older instrument where the optical elements are no longer very clean.
A good possibility to check whether the chosen values are correct one should repeat the measurement with the same concentration and a smaller path length of the cell (1/2 e.g.) and compare the band shape of the CD spectrum: for a good measurement the band shape has to be equal within the experimental accuracy.
The answer given above ist correct. The important quantity for the check of the measurement is the product of concentration time the thickness of the sample or the dissymmetry factor = delta epsilon divided by the extintion. It is true that the given values for instruments - especially if the are not new and possess not very clean optical elements (mirrors, prism etc.) - are too high. Changing the c times d ist a good prove.in the sprectral region below 220nm I would never choose extinction hogher than 0,8 t0 1.