This task is best done with the aid of some programming language. Here are some clues of what can be done in SQL:
mysql> CREATE TABLE f420270 (s VARCHAR(22) CHARACTER SET latin1);
mysql> INSERT INTO f420270 (s) VALUE (CONCAT('A', UNHEX('92'), 'C'))
mysql> INSERT INTO f420270 (s) VALUE ('This is ok');
mysql> SELECT s, HEX(s) FROM f420270;
+------------+----------------------+
| s | HEX(s) |
+------------+----------------------+
| A'C | 419243 | <-- see the 92
| This is ok | 54686973206973206F6B |
+------------+----------------------+
# Find rows with any 8-bit chars:
mysql> SELECT * FROM f420270 WHERE HEX(s) RLIKE '^(..)*[8-F].';
+------+
| s |
+------+
| A'C |
+------+
# Or look only for your naughty char:
mysql> SELECT * FROM f420270 WHERE s RLIKE UNHEX('92');
+------+
| s |
+------+
| A'C |
+------+
# chars before the first naughty:
mysql> SELECT s, LENGTH(SUBSTRING_INDEX(s, UNHEX('92'), 1)) FROM f420270;
+------------+--------------------------------------------+
| s | LENGTH(SUBSTRING_INDEX(s, UNHEX('92'), 1)) |
+------------+--------------------------------------------+
| A'C | 1 |
| This is ok | 10 |
+------------+--------------------------------------------+
That LENGTH(SUBSTRING_INDEX(...)) is a start at picking out the bad char using UPDATE ... s=REPLACE(...).
If hex 92 is your only problem, then let's continue:
First, I want to add a row with two naughties, so as to verify that both get fixed.
mysql> INSERT INTO f420270 (s) VALUE (CONCAT('a', unhex('92'), 'c', unhex('92'), 'e'));
mysql> UPDATE f420270 SET s=REPLACE(s, UNHEX('92'), '-');
Query OK, 1 row affected (0.12 sec)
Rows matched: 3 Changed: 1 Warnings: 0
mysql> SELECT * FROM f420270;
+------------+
| s |
+------------+
| A-C |
| This is ok |
| a-c-e |
+------------+
So, maybe all you need is
UPDATE tblname SET colname = REPLACE(colname, UNHEX('92'), '');