Sum values distinct in time

In the attached sheet, I have listed downtime in seconds by station number on my production line. If I look at total downtime for the day by simply totaling up column F, I include values that overlap in time, thereby inflating actual seconds down. For example, the downtime event on row 5 occured within the duration of the downtime event in row 4, therefore I would only want to include the 21 second duration of the event from row 4. Is there a formula or table I could create that would omit these values that occur in the same moments in time and only count distinct values?

Perhaps
in g2=d2
in h2 =MAX(ROUND((E2-G2)*24*60*60,0),0)
in g3 =IF(MAX($E$2:E2)>D3,E2,D3)

copy h2 and g3 down to the bottom of your data

Any closer?

main formula
=SUM(IFERROR((IF(E3>$E$3:$E$2635,$E$3:$E$2635,E3)-IF(D3>$D$3:$D$2635,D3,$D$3:$D$2635))^0.5,)^2)-G3 as array formula

Very close, but doesn't seem to work when there are more than 2 overlapping values. Thanks!

Looks like a winner! Thanks!

Actually may have spoken too soon. This method also seems to have an issue when dealing with more than two occurences during the same time period.

=SUM(IFERROR((IF(E3>$E3:$E$2635,$E3:$E$2635,E3)-IF(D3>$D3:$D$2635,D3,$D3:$D$2635))^0.5,)^2)-G3
and just
=SUBTOTAL(9,F3:F6000)
but rslt is the same