file splitting

I have an ebcdic file of 8gig into multiple chunks

using the following script

split -b 1024m 04122004.BIG.txt RMS

the first file RMSaa looks fine. But in the second file RMSab,

the first letter starts at 9th byte instead of 1st byte and therfore it chopps everynting else that falls after 152bytes.

The records size is 152bytes.

Why does it split like that.

Any other ways i can split the file using unix

split -b 912000000 ... 

use -b with a mutliple of the file record length