Yes, I just found a page explaining that, and using it like you specified
now it outputs UTF-8 correctly, with the charset attribute in the META tag
set to UTF-8 as well.

Now I have a follow up question. I get an UTF-8 encoded XML document that I
load into a DOM object and then transform (the "domXML" below).
The xml is this:
http://gabbah.no-ip.org/mirakel/default.asp?ctype=STP&cdoc=object&cnMaxRows=
10&cnBookmark=H2856&cpKNP=*&transform=no

So everything looks fine, the XML is UTF-8, the transformation outputs
UTF-8. BUT, the characters E D V in the xml-file are encoded and I can't get
them to look correct no matter what encoding I choose in the browser.
I don't understand why this is happening in the transformation! I mean if
everything is UTF-8 it should work fine right?

The result is this:
http://gabbah.no-ip.org/mirakel/default.asp?ctype=KSP&cdoc=searchList&cpKNP=
*%C3%85*&cpADR=&cpSTN=&cnMaxRows=10