Velocity gradient tensor

[Bhargav Mantravadi]

Dear All,

I am computing the velocity gradient tensor on the fly, followed by a volume average. However, the components of the velocity gradient are all turning out to be zero, despite the flow being developed. Could you please advise on what I might be doing incorrectly? I have included the relevant code below.

I am using the subroutine lambda2(l2) in ./core/postpro.f as a reference. 

Thank you in advance for your help.

Regards,

Bhargav

-------------------------------------------

   subroutine userchk()

      include 'SIZE'
      include 'TOTAL'

      parameter (lxyz=lx1*ly1*lz1)
      real gije(lxyz,ldim,ldim)
      real visc_dsp_l(lxyz,nelv)
      real visc_dsp

      nxyz = lx1*ly1*lz1

      do e=1,nelv
      !compute velocity gradient tensor
      call comp_gije(gije(1,1,1),vx(1,1,1,e),vy(1,1,1,e),vz(1,1,1,e),e)
      do i = 1,nxyz
      visc_dsp_l(i,e) = 2.0*gije(i,1,1)**2 + 2.0*gije(i,2,2)**2
     &  + 2.0*gije(i,3,3)**2
     &  + (gije(i,1,2) + gije(i,2,1))*2
     &  + (gije(i,1,3) + gije(i,3,1))*2
     &  + (gije(i,2,3) + gije(i,3,2))*2
      enddo
      enddo
      visc_dsp = glsc2(visc_dsp_l,bm1,n)
      visc_dsp = visc_dsp/volvm1

      return
      end

[YuHsiang Lan]

Hi Bhargav,

It looks like you miss to assign the variable "n".
Also, the default implicit for "e" is real, not the integer you need.

For debugging, I suggest using "implicit none" to make sure every variable is declared.

While the compiler tells some variable need declaration, it's also a good time to check whether its value is assigned.

ChatGPT (https://chatgpt.com/g/g-68a731e21f7c8191bcc684cc265a1d79-nek5000-assistant) is able to catch "n" but not "e":

https://chatgpt.com/share/693dc6ca-f23c-8009-99a5-2ca128c76885

Hope this helps,

Yu-Hsiang

--

[Bhargav Mantravadi]

Hi Yu-Hsiang,

Thank you very much for your reply and the helpful input. I have noted your suggestions, and the problem is now resolved.

Regards,

Bhargav

--
You received this message because you are subscribed to the Google Groups "Nek5000" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nek5000+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/nek5000/060e0071-6320-4e1c-810d-b32da8f82c61n%40googlegroups.com.