You could try replacing the text using the Filter() function. I suggest you experiment with it in a separate field first to see if it does the trick.
It may be CR converted to Char(11) in csv.
I would also use the filter function to clean out the garbage.
In the field definition put an auto enter calculation of Filter ( the_field ; "012345679") if its a number only field or Filter ( the_field ; "012345679abcdefghijklmnopqrstuvwxyxABCDEFGHIJKLMNOPQRSTUVWXYZ") if its an alpha numeric only field.
Set the calc to replace existing values always
The all you have to do then it show all records in the table and replace the contents of the_field with itself.
That's pretty much what I had in mind. You will need to include in your filter list any punctuation that is acceptable (decimal points, etc) and also a space if that is also acceptable. I would still strongly advise using a separate calc field rather than Replace, so you have an escape route if you needed. From what you describe, it sounds like you are working in an interim table at this stage to clean up the data before importing into a new solution, so at that final stage you can import the cleaned up data from the calc field anyway.
I recognize the character . It's an enter value in your original data.
You can fix this with a substitute.
a co-worker of mine had me download and install sublime text 2 (freeware), and it allows you to open a csv file. It highlights all the Unicode characters, and with a few fancy keyboard shortcuts, allows you to pinpoint exactly what you want cleaned out, and it does it fast.
Got me out of a jam.