UNIX for Dummies Questions & Answers

I have a series of filenames of the format junk_x_###.txt where x and ### are numbers. I need to detect when at least 30 consecutive files (e.g. junk_1_001.txt, junk_1_002.txt.....) are in the directory. There are sets of files with similar names that need to be grouped together and counted separately (e.g. junk_2_001.txt..., junk_3_001.txt....) and whichever cluster has more than 30 selected. I also need to allow for the consecutive filenames not beginning at 001 (e.g. junk_1_100.txt, junk_1_101.txt.....).

In essence I need to identify 30 consecutive files with identical filenames as far as junk_x_
What is the best strategy to take here? I'm trying to cluster the root of each filename, list them and then pipe to wc -l to count them but not getting the initial clustering to work well. Any suggestions for this step?
thanks