Hi, I don't think you need to touch hpts(). Just compute whatever you're after in userchk() and store the results in the T-array. After that call hpts(). Also, make sure that you dump the right fields (check OUTPUT section in .rea file) Stefan -----Original Message----- From: [email protected] on behalf of [email protected] Sent: Mon 7/19/2010 09:30 To: [email protected] Subject: [Nek5000-users] Total energy and wall gradients Hi Nek's, I am about to run a 3D-stenotic flow DNS, and I would like to compute the total energy in the computational domain. So far, I have only a routine for 2D (surf_int). Does such a 3D routine exists? My second question concern the velocity gradient normal to the wall. Is there any way to output its value at various wall points for each time step? I was thinking ofadding a few lines to the hpts routine, but I am sure about what to write exactly. Could anyone give me a clue? Sincerely, JC _______________________________________________ Nek5000-users mailing list [email protected] https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users