Molcas Forum

chipmunk

Member

Registered: 2015-11-12

Posts: 23

ERROR: Making explicit Hamiltonian failed

Dear Molcas Developers and Users,

My calculations suffer from the error below. I tried to increase memory but the problem still happens. Could you please give some advice on this problem?

Thank you in advance,

Kind regards,

Nhan

 ###############################################################################
 ###############################################################################
 ###                                                                         ###
 ###                                                                         ###
 ###    ERROR: Making explicit Hamiltonian failed.                           ###
 ###                                                                         ###
 ###                                                                         ###
 ###############################################################################
 ###############################################################################
  An unforeseen catastrophic failure occurred
  in the CI solver. The size of the explicit
  part of the CI Hamiltonian matrix was not
  sufficient. Suggested fix: Change the size
  by adding 'SDAV=XXXXX' to the rasscf input.
  XXXXX is some integer at least            23622320328
  Sorry about this. Consider telling the
  Molcas group about this failure.

Ignacio

Administrator

From: Uppsala

Registered: 2015-11-03

Posts: 1,085

Re: ERROR: Making explicit Hamiltonian failed

The error message already contains a suggestion: add 'SDAV=XXXXX' to the rasscf input. However, the suggested minimum value looks terribly wrong. It could be something wrong in your input or active orbitals, or an inherent difficulty in your system, or a bug. Hard to say without a complete input.

chipmunk

Member

Registered: 2015-11-12

Posts: 23

Re: ERROR: Making explicit Hamiltonian failed

Hi Ignacio,

Thank you for your quick reply.

This is my input. The ExoOrb is about 43 MB. I will upload it to server and then share the link here later.

&GATEWAY
  Coord=LCuI.xyz
  Group = NoSym  
  Basis Set = O.ANO-R2,H.ANO-R1, C.ANO-R2, N.ANO-R2, P.ANO-R2, Cu.ANO-R3
  RICD 
  RF-input
  PCM-model
   solvent
   acetonitrile 
   AAre
   0.2
  End  of  rf-input 

&SEWARD 


&RASSCF  
FileOrb=$CurrDir/$Project.ExpOrb
Spin = 1 
Nactel
22 2 0 
Orbappear
*Full
Compact
OrbListing
ALL
CiRoot
1 1 1

>> COPY $Project.RasOrb  $CurrDir/$Project-1Root.INPORB 

&CASPT2
Multistate
1 1     
Maxiter
 100
Imaginary = 0.1
RFpert


&RASSCF  
Spin = 1 
Nactel
22 2 0 
Ras1
6
Ras2
11
Ras3
0
Orbappear
*Full
Compact
OrbListing
ALL
CiRoot
4 4 1
RFRoot= 4
NONEquilibrium
End of Input

>> COPY $Project.RasOrb  $CurrDir/$Project-4Roots.INPORB 

&CASPT2
  Multistate= 1 4
  RFPert
  End of Input

and this is the xyz file

I hope that developers have had the coordinate data. 
This is our on-going project in collaboration with several experimental partners.
Due to confidential concern, the system is hidden. Coordinates will be provided upon request. 
I am so sorry for this inconvenience.

Last edited by chipmunk (2022-07-07 13:41:28)

chipmunk

Member

Registered: 2015-11-12

Posts: 23

Re: ERROR: Making explicit Hamiltonian failed

Hi Ignacio,

I just updated the input and this the ExpOrb used in the input. In this orbital file, we can see the active space.

But basically, ras1 6, ras2 11, and ras3 0.

https://www.swisstransfer.com/d/e98ee3a … 1652658747

Thank you for your help.

Cheers,

chipmunk

Member

Registered: 2015-11-12

Posts: 23

Re: ERROR: Making explicit Hamiltonian failed

Hi Ignacio and developers,

Now, I am sure that this is a bug in the new code (version: 22.06, tag: 14-gf044a5e4c). This annoying error has not happened in the older version (version: 19.11, tag: 1167-gecad0298). Also, I have to note that something weird related to PCM model also happened in the newest version (version: 22.06, tag: 14-gf044a5e4c). I got kind of long and tedious warning about iteration as follows. This annoying PCM model warning did not happen in the older version too.

Cheers,

      Polarizable Continuum Model (PCM) activated
      Solvent: acetonitrile
      Version: Dielectric
      Average area for surface element on the cavity boundary: 0.2000 Angstrom^2
      Minimum radius for added spheres: 0.2000 Angstrom
      Calculation type: equilibrium
 

           INTER: too many iterations

           INTER: too many iterations

ATTENTION: THE SURFACE OF A TESSERA IN SPHERE   4 IS NEGLECTED

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

ATTENTION: THE SURFACE OF A TESSERA IN SPHERE   4 IS NEGLECTED

           INTER: too many iterations

           INTER: too many iterations

ATTENTION: THE SURFACE OF A TESSERA IN SPHERE  35 IS NEGLECTED

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations

           INTER: too many iterations
       Reaction field from state:            1
--

Last edited by chipmunk (2022-07-07 03:47:02)

chipmunk

Member

Registered: 2015-11-12

Posts: 23

Re: ERROR: Making explicit Hamiltonian failed

Today, I spent my day to check several versions of OpenMolcas, and confirm that this error happened in all versions of 22.xx.

Ignacio

Administrator

From: Uppsala

Registered: 2015-11-03

Posts: 1,085

Re: ERROR: Making explicit Hamiltonian failed

Please try to reduce your calculation to something smaller that shows the same kind of issue and that you can share.

chipmunk

Member

Registered: 2015-11-12

Posts: 23

Re: ERROR: Making explicit Hamiltonian failed

Please try to reduce your calculation to something smaller that shows the same kind of issue and that you can share.

Sure! Give me some time and I will do. Thanks Ignacio.