How can I successfully read unicode with special characters from Oracle db to SQL db through SSIS?

I created a dataflow task which has a source oledb connection to Oracle and a destination oledb to SQL server. While testing the data I noticed that special characters in Oracle are being inserted as ? in SQL server, next I filtered the data of the source query to check only one record and when I did a preview in SSIS it showed with the ? instead of the special character. I tried several things: - Changed the code page to 65001 instead of 1252 but that didn't work. - Changed the source data type from DT_STR to DT_WSTR but still didn't show correctly. - Changed the source driver from "Microsoft OLE DB Provider for Oracle" to "Oracle Provider for OLE DB", however the processing is SO slow it took over 2 hours to process around 3 million records. P.S. The source data type is varchar2 Any suggestions? More info: The column in question is ITEM_DSCR. Source (Oracle): COLUMN_NAME DATA_TYPE DATA_LENGTH DATA_PRECISION NULLABLE ITEM_DSCR VARCHAR2 120 NULL Y Source OLEDB Connection (SSIS): I changed it to DT_WSTR from advanced editor for OLEDB connection, under OLEDB Source Output (Output Columns). P.S. every time I change the same under OLEDB Source Output (External Columns) it gets changed back automatically to DT_STR. Examples of the special characters involved: ¹ ¬ ¦ ¿