It is possible but you will have incredible peak broadening due to longitudinal molecular diffusion unless you use small lenght column (ie 100mm), smaller particules size and low void volume (from the unions especially).
SALAAM AND GREETINGS. YOU CAN USE 3 COLUMS IN SERIES PROVIDED THEY ARE SHORT COLUMNS e.g. 5cm in length or may two 5cm columns + one 15 cm column to make the total length 25 cm or other combinations in length so it does not exceed 25 to maxium 30cm other wise you will have peak broadening. Also it is wise to use small particle size columns 3 micron and /or preferably solid core columns.
I would not recommend this approach at all and not just because of back pressure concerns. With most chiral columns, they either work to resolve your sample apart or they do not. They do not all contribute to the separation as is the case with some other types of columns. Coupling different chiral columns in-series provides no gain in efficiency or performance at all (unlike with SEC/GPC columns). Each column will only add dispersion to the sample and make it harder for the next column to show any separation (if there is any). The result will be that you will miss finding all possible chiral separation methods. With NP/RP chiral separations, it is best to use one chiral column at a time (largest column you can get to maximize the available chiral selector present).