Calculating delta t from a dataset? Diff command.

I have a very large dataset named Current that contains over a million rows. Here is an example of the first 15 lines:

 Hour Min   Second      Line1   Line2   Power  Dt
0  0  3  1.94  3.71  678  6
0  0  9  1.93  3.71  676.8  5
0  0  14  1.93  3.71  676.8  6
0  0  20  1.94  3.72  679.2  5
0  0  25  1.93  3.72  678  6
0  0  31  1.93  3.72  678  5
0  0  36  1.93  3.73  679.2  6
0  0  42  1.93  3.72  678  5
0  0  47  1.93  3.72  678  6
0  0  53  1.92  3.72  676.8  5
0  0  58  1.92  3.72  676.8  -54
0  1  4  1.92  4.99  829.2  5
0  1  9  1.92  5.02  832.8  6
0  1  15  1.92  5.02  832.8  11
0  1  26  1.92  5  830.4  6
I am trying to fill Dt, Delta T, with the change in time using 

    Current.Dt = diff(Current.Second);

There are two problems with this. 1. The Dt's need to be shifted down one. For example the second row should be 6 instead of 5 and the third 5 instead of 6. 2. The second problem is when the diff takes the value 4-58 it gets a the value of -54. This is incorrect. The time data was collected by the min so the value should actually be equal to 64-58 or 6.

I am not sure how to fix this and any help will be appreciated

Answer 1

[Star Strider]

I suggest you consider the etime function. You’ll have to pad each 3-element time vector with leading zeros to get the six-element date vector, but it is probably the easiest solution to implement.