That makes alot more sense.
but still the reading from the cross correlation is 12 differentce between the 2, so 12ms .. which isant correct because the real delay is 3ms. I dont suppose you know why this is the case? if not no worries you've helped alot :)