"+¦" is suppose to represent "ö"? So at least 10 bytes? No one encoding would produce that. Even in UTF-8, ö only takes two bytes. Repeatedly encoding using UTF-8 doesn't result in anything like what you have either. (There would be repetition.)
Between + and | there were actually 4 characters. When I pasted the chars in perlmonks, it further added 4 more characters. Hence you noted 10 characters in total.