EHT dimension problem

[Raffaele Credendino]

Hello, I'm Raffaele and I'm studing a system with:
natoms=471
nshell=4396
nbf(CAO)=9420
nbf(AO)=9106

I use Turbomole 5.10.

I think that dimensions of my system are under limits of Turbomole default version:

   Parameter         |     default version    |    huge version
      ----------------------------------------------------------------
      atoms            |        700             |        1400
    basis functions  |      10000            |       21000
        shells           |       6000             |       16000


I have this problem when I try to use EHT in define:

----------------------------------------------------------------------------------------------------------------------
reading orbital data 1S(DZ)  from file /home/programs/TURBOMOLE-5.10/basen/mg .




  dimension too small
<ehtinp> : nshell =       6594 > ndi9 =       6000


MODTRACE: no modules on stack

fatal error in dimchk
define ended abnormally
define ended abnormally
program stopped.
--------------------------------------------------------------------------------------------------------------------

How should I resolve it?
Thank you.
Have a good day.

[antti_karttunen]

Hello Raffaele,

It's quite common that the EHT guess hits some limits even though the actual modules (dscf, ridft, etc.) still run fine. I suggest that you request the huge version from Cosmologic and use the define_huge binary to create the initial guess.

Regards,
Antti

[Addiw7]

Hi,

I have a similar problem. While using eht I get this information:
    sao summary :
   irrep      number of sao's referring to
                old basis    new basis
    a              69          526

So in control file it says: $rundimensions
   dim(fock,dens)=178528
   natoms=25
   nshell=194
   nbf(CAO)=595
   nbf(AO)=526
   dim(trafo[SAO<-->AO/CAO])=733
   rhfshells=1

But in mos file I have only 40 MO (which cover 80 electrons in my molecule with +1 charge).
What to do?

[antti_karttunen]

Hi,

Do you mean by "similar problem" that Define crashes also for you? Or are you just wondering the low number of MOs in the mos file? Did you converge the SCF with dscf or ridft? Before you converge SCF, the mos file only contains the initial guess from EHT and no virtual orbitals at all.

Antti

[Addiw7]

Hi,

I didn't know about it  . Thank you for advice, you were right of course .

[juhidutta]

Hello
I am using b97-d level of theory and 6-31G* as external basis set for Se systems. Upto the total no. of atoms equal to 80, it was working fine but when the number of atoms increased to 96, it was showing the following error while defining:


dimension too small
<makbas> : ncons =         55 > ndi11 =         52


MODTRACE: no modules on stack

fatal error in dimchk
define ended abnormally


Please, help me out.
Thanking You,
Juhi.

[uwe]

Hi,

that looks like an old Turbomole version. Meanwhile all those limitations are gone.

Nevertheless, the limits are so small  that I guess that you might have tried to manually change some settings in the control file.

Please delete the $rundimensions keyword (by hand or just run kdg rundimensions on the shell) and retry to run define.

Regards,

Uwe

[juhidutta]

Hi Uwe,
Thanks for your reply. But the way you told, it is not working. It is showing the same error. I am using Turbomole 6.5 version. Please just tell me How to use external guess mos avoiding the eht?? Or if anything else is possible, please suggest me that also.
actually the problem is due to the increase of Se atoms not the total no. of atoms. So please help me out.

Thanks,
Juhi.