Thanks.  However, I see the same problem even putting both characters
in the same font.  With S-mouse-1 "Change Buffer Font..." I select
Norasi.  M-x describe-char then reports

         character: ◌ (9676, #o22714, #x25cc)
preferred charset: unicode-bmp
 		   (Unicode Basic Multilingual Plane (U+0000..U+FFFF))
        code point: 0x25CC
            syntax: w 	which means: word
          category: .:Base
       buffer code: #xE2 #x97 #x8C
         file code: #xE2 #x97 #x8C (encoded by coding system utf-8-unix)
           display: by this font (glyph code)
     xft:-unknown-Norasi-normal-normal-normal-*-13-*-*-*-*-0-iso10646-1 (#x146)

and

         character: ึ (3638, #o7066, #xe36)
preferred charset: unicode-bmp
 		   (Unicode Basic Multilingual Plane (U+0000..U+FFFF))
        code point: 0x0E36
            syntax: w 	which means: word
          category: 2:upper diacritic, t:Thai
       buffer code: #xE0 #xB8 #xB6
         file code: #xE0 #xB8 #xB6 (encoded by coding system utf-8-unix)
           display: by this font (glyph code)
     xft:-unknown-Norasi-normal-normal-normal-*-13-*-*-*-*-0-iso10646-1 (#x109)

The characters remain uncombined.