From: Abhijit Gogoi (gogoiabhijit10_at_gmail.com)
Date: Wed Feb 19 2020 - 02:25:32 CST
One of the reasons may be "not minimizing the system long enough". You can
check the convergence of the total energy (or volume) in NAMDPlot.
Another reason may be "not properly specifying the periodic boundary". You
can check the dimensions of the simulation box in VMD and accordingly
specify the periodic boundary.
If you are heating the system (or if there are some implicit internal
forcing is required within the simulation box) then it is good to use a
lower value of time step at least during the equilibration run.
Best wishes
Abhijit
On Tue, Feb 18, 2020 at 10:03 PM Josh Vermaas <joshua.vermaas_at_gmail.com>
wrote:
> Minimization was not successful. You've got a long bond somewhere or bad
> geometry that splits your geometry past adjacent patches, or at least that
> is why I get global count errors. You'll see warnings in the log to this
> effect.
> -Josh
>
> On Tue, Feb 18, 2020, 1:34 PM Francesco Pietra <chiendarret_at_gmail.com>
> wrote:
>
>> Hello
>> With a large system (RNA-protein in TIP3, Atoms: 1352778) after
>> successful minimization, first by restraining RNA and the protein, while
>> minimizing TIP3 MG ZN2 SOD, then without any restraint, at heating, with
>> either ts=0.5fs or 0.1fs the error arises
>>
>> TCL: Running for 500000 steps
>> REASSIGNING VELOCITIES AT STEP 0 TO 1 KELVIN.
>> FATAL ERROR: Bad global bond count! (984320 vs 984325)
>>
>> FATAL ERROR: See http://www.ks.uiuc.edu/Research/namd/bugreport.html
>>
>> There is some precedent for such curious error, however unresolved
>> *From:* Peterson J (*peterson.jjjj_at_gmail.com*
>> <peterson.jjjj_at_gmail.com?Subject=Re:%20FATAL%20ERROR:%20Bad%20global%20bond%20count!>
>> )
>> *Date:* Wed Jun 12 2013 - 11:01:23 CDT
>>
>> thanks for advice
>> francesco pietra
>>
>>
This archive was generated by hypermail 2.1.6 : Thu Dec 31 2020 - 23:17:12 CST