I am agree with Shawn, this problem is due to mismatch of charachter set.
first of all thanks for your reply.
I had same problem when loading data from SQL server to Oracle.As you already know this happens due to windows and UTF8 charset. So, here is what i did to fix issue. normally we create external table for the file in Oracle using "RECORDS delimited BY newline CHARACTERSET WE8MSWIN1252 " in ORGANIZATION EXTERNAL options.
Then do below, to read properly,
select cam_name,CONVERT(cam_name, 'WE8MSWIN1252','AL32UTF8' ) from ext_table
Convert will do the trick and all international characters get loaded properly. you normally find problems with chinese/japanse/indian characters. hope this helps.
The blog is not the appropriate place for this question, but most likely your issue is that you are using the Windows-1252 character set in your SQL-Server database. Windows-1252 has a subset of characters which do not map to other more standard character sets and you'll have to replace them with spaces or write some manual code to translate the characters to UTF-8. In the future you should make sure to configure SQL Server to use a standard character set such as UTF8 instead of Windows-1252 and you will not have this type of problem.