Closed Captioning Workflow

I am trying to add Closed Captions to a project for CCTV, using compressor. Here's the output standards I'm told the TV should use:

MPEG2 Transport stream with AC3 audio
MPEG-2 Video
3600 kb/s
23.976 fps (use 3:2 pulldown if source is 29.97)
128 kb/s audio rate

I created a preset in Compressor to do this, ended up with a m2v and ac3 file, then used a freeware program called MPEG Streamclip to get a .ts file as the final output (the reference file that is known to work was also in this format).

I was concerned that I might lose the caption file when I used MPEG Streamclip to mux the audio and video, but I tested with VLC and captions appeared in the file with no problems. However, on the CCTV network, captions are not showing up whatsoever.

Is there something that I could be overlooking? When I view info for the reference file, it's nearly identical to the file I created (difference in resolution, reference file is 1080, but don't think that should matter for captions?)

Any insight is appreciated.

http://www.orangeslicevideo.com

The sample provided was from the tv series 'Chuck'. It works correctly. Is there a way to reverse engineer the file to find what was used?

http://www.orangeslicevideo.com