[CP2K-user] Error: Cholesky decompose failed: the matrix is not positive definite or ill-conditioned.

[Travis]

Hi,

It can often be helpful to center your system at least initially in the box 
when running gas phase calculations,

  &TOPOLOGY
   COORD_FILE_FORMAT XYZ
   COORD_FILE_NAME foo.xyz
   CONN_FILE_FORMAT off

*   &CENTER_COORDINATES   &END CENTER_COORDINATES*
  &END TOPOLOGY

Try both the MT and WAVELET solvers to see if perhaps one is more stable or 
can use a smaller box.

-T


On Tuesday, November 5, 2019 at 11:20:00 PM UTC-5, Yang Liu wrote:
>
> Dear Travis,
>
>     Thanks for your suggestion! I will try a larger box, see if it fixes 
> the problem.
>
>     Best regards,
>     Yang
>
> On Wednesday, November 6, 2019 at 2:00:35 AM UTC+9, Travis wrote:
>>
>> Hi,
>>
>> For gas phase simulations, the box size has to be large enough that the 
>> electron density is essentially zero at the box edges.
>>
>> Refer to the message here, 
>> https://manual.cp2k.org/trunk/CP2K_INPUT/FORCE_EVAL/MM/POISSON/MT.html
>>
>> -T
>>
>>
>> On Tuesday, November 5, 2019 at 8:43:11 AM UTC-4, Yang Liu wrote:
>>>
>>> Dear CP2K users,
>>>
>>>     I'm trying to run NVT using cp2k, but this error occurred. Could you 
>>> please help me solve this problem? Thank you so much!
>>>
>>>     attached with my input and xyz file.
>>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20191105/62981a5e/attachment.htm>