Hey Holger!

Looking at a cp437 table and the characters you seem to be most interested in,
it occurs to me that all of them once converted to utf-8 become a pair of 8-bit
characters starting with hex C3 (decimal 195).  That happens to coincide with a
line/box drawing character (not text).  The second byte or character will vary
depending on what unicode character the pair of bytes actually is and may or
may not be reproducable in cp437.  A perfect example is, "A Møøse once bit my
sister ..." which contains the slashed 'o' character (Norwegian/Danish vowel). 
In this case you should see the two byte sets as hex C3 B8.  Hex B8 (decimal
184) is another line/box drawing character in cp437.  Bottomline is that you
should be able to replicate any utf-8 character using any 8-bit editor once you
know what unique pair of 8-bit characters in your editor represent, and in your
case all the pairs should start out with hex C3 (decimal 195) for ALL the
characters you have asked about.  I've seen exceptions but in those cases it is
usually some long dead language or non-Latin based.

Does any of the above help?  It should be really easy for your purpose to
convert what little you need to convert since there are only 6 characters in
the Swedish language you need to worry about although there are a few others
that sneak in, such as some Germanic or Nordic characters every now and then
but those should be readily identifiable.  The slashed 'o' one isn't doable
though which is a very good reason to ditch cp437.  ;-)

Life is good,
Maurice

... Don't cry for me I have vi.
--- GNU bash, version 4.3.30(1)-release (x86_64-atom-linux-gnu)