I need to subtotal times from a list where they overlap.

I have a column (M) that the contents are the result of subtracting a cell in column C from Column K. Now some of these numbers are positive and some are negative (#####). They are all the difference in start and stop times. These times are the start and stop of computer jobs and some run concurrently and some run after a break. So, you might have Job1 that runs from 09:00 to 09:15 while Job2 runs from 09:10 to 09:20. Essentially they took together 20 min to run (what I want) while if you added the times of each you would come up with 25 min (not what I want). Then you might have a 2 hour break and Job3 runs for 30 min, then and hour later Jobs 5-6 run from 12:00 to 14:30, 2.5 hours. These jobs are grouped in any one of 6 categories.*

What I want to get is the "real" time it took to run all the jobs in each category and I can't get my mind around how to do it. In the example above the correct answer would be 2:30 + 0:30 + 0:20 or 3 hours and 20 minutes. I have all this information in a spreadsheet output by the software that runs it, so I can sort by category, then start time to get a baseline. I have to do this every day at end of day, so I really don't want to sift through 650 rows or so to figure it out manually.

Thanks for any help, I'm lost

Here is an example:

Capture.JPG

Thank you.

Sorry, should have posted a completed one:

Capture with total.JPG

So the answer in this case is 1:08.

HA! that was a test! Yes, my mistake, but you have the idea exactly. So, essentially, the way I see it if cell Ax is larger then the value in cell Bx-1, then Cell Bx-1 is the end of the series. Then cell Ax begins the start of a new series and so on. Thank you for the reply.