It will be better to check after 48 hours transfection and subsequently in 72 hrs as well as 96 hour in culture medium.

I have tried several methods of collecting lentivirus and found in general that daily collection starting 24 hour after transfection out to 96 hours post transfection followed by pooling the collected supernatants and enrichment of virus works well (~80-85% transduction efficiency). This being said you could probably produce higher efficiency virus by testing each of these fractions individually but I suspect this would reduce overall yield (ie you can do fewer transductions with each viral transfection) and increase the variability in the viral supernatant produced. Equally important is the confluence of the HEK cells at the time of transfection as the cells will produce virus best when healthy and able to proliferate so to maximize virus production over a multiple day period (4 days for the method I use) it would be ideal for the cells to reach 100% confluence after 72 hours post transfection (again this is if you are using a similar 4 day method).

I hope this helps

I've done this experiment in the past and the answer actually depends on the transfection reagent and the pseudotype you're making. If you're using CaPO4, virus production peaks at 24-48 hours and decreases at 72h. If you use PEI or Lipofectamine, the best times to harvest are 48 and 72 hours. VSV-G is produced quite quickly but for some pseudotypes you'll get barely anything at 24 hours. Best way to know is to try at different time points and see what you get