[Pw_forum] convergence stops after 100 iterations

Feng, Zimin Feng.Zimin at ireq.ca
Tue Dec 24 18:17:16 CET 2013


in order to allow more than 100 iteration you need to change the code and recompile as this limit is hard coded.
but as is mostly the case, there is probably something else that's bothering you. you might want to plot the atomic structures at each of the steps and see if everything's alright.
fzm


-----Original Message-----
From: pw_forum-bounces at pwscf.org on behalf of manchugh at iitk.ac.in
Sent: Tue 12/24/2013 9:00 AM
To: PWSCF Forum
Subject: Re: [Pw_forum] convergence stops after 100 iterations
 
Hi Nguyen Chuong

I have used PBE (GGA)
electron_maxstep in my calculations is 100 by default. It means 1 scf
cycle will take 100 iterations at max,  nstep = 100 means number of ionic
+ electronic steps. It does not mean convergence will stop after 100
iterations.
I have convergence problem with different vacuum gaps but same number of
atoms in two cases I mentioned in my previous post.. why so ?

thanks

 Manjusha
 Research Scholar
 Department of Chemistry
 Indian Institute of Technology,Kanpur
 India

> and I have a question, in your file, which method you used for
> calculation?
> PBE (GGA) or PZ (LDA)
>
>
> On Tue, Dec 24, 2013 at 5:30 AM, Nguyen Chuong
> <chuongnguyen11 at gmail.com>wrote:
>
>> in your input file, nstep = 100, its mean convergence  will stopping in
>> 100 iterations.
>>
>>
>> On Tue, Dec 24, 2013 at 3:30 AM, <manchugh at iitk.ac.in> wrote:
>>
>>> Hi Quantum Espresso users
>>>
>>> I am trying to perform 'relax' calculations of GaN surface and trying
>>> to
>>> optimize the vacuum gap keeping the slab thickness same. One side of
>>> the
>>> slab (the N-terminated one) is passivated by pseudohydrogens of charge
>>> 0.75. For some values of the vacuum gap, relaxation proceeded
>>> successfully, (got Final energy in 12 scf cycles) but for some other
>>> values of vacuum gap, the 100 iterations in the Ist scf cycle itself
>>> are
>>> not sufficient and the following error appears:
>>>
>>> convergence NOT achieved in 100 iterations: stopping
>>>
>>> I am using  espresso-5.0.2 and my input file is as follows :
>>>
>>> &CONTROL
>>>     calculation = "relax",
>>>     title = '1x1x4_7gap'
>>>     verbosity = 'high',
>>>     restart_mode = 'from_scratch',
>>>     prefix = '1x1x4_7'
>>>     pseudo_dir = '/home/.....',
>>>     outdir = '/home/....../',
>>>     tprnfor = .TRUE.,
>>>     etot_conv_thr = 1.0D-04,
>>>     forc_conv_thr = 1.0D-03,
>>>     nstep = 100
>>>  /
>>> &SYSTEM
>>>     ibrav = 0 ,
>>>     celldm(1)= 6.088,
>>>     nat = 19 ,
>>>     ntyp = 3 ,
>>>     ecutwfc = 60.0 ,
>>>     ecutrho = 480.0 ,
>>>     input_dft = 'PBE' ,
>>>     occupations = 'smearing',
>>>     smearing = 'methfessel-paxton' ,
>>>     degauss = 0.05
>>>  /
>>>
>>> &ELECTRONS
>>>   diagonalization='cg',
>>>   conv_thr = 1.D-6,
>>>   mixing_mode = 'local-TF',
>>>   mixing_beta = 0.2D0
>>> /
>>>
>>> &IONS
>>>   ion_dynamics = 'bfgs'
>>> /
>>>
>>> CELL_PARAMETERS {hexagonal}
>>>  1.00  0.00   0.00
>>>  0.50  0.866  0.00
>>>  0.00  0.00   17.886
>>>
>>> ATOMIC_SPECIES
>>>   Ga  69.723  Ga.pbe-n-van.UPF
>>>   N   14.006  N.pbe-van_ak.UPF
>>>   H   1.0079  H.pz-vbc_075.UPF
>>>
>>> ATOMIC_POSITIONS {crystal}
>>> H   0.3333  0.3333  -0.029195     0  0  0
>>> N   0.3333  0.3333  -0.0111818    0  0  0
>>> Ga  0.000   0.000   0.000         0  0  0
>>> N   0.000   0.000   0.03427272
>>> Ga  0.3333  0.3333  0.04545454
>>> N   0.3333  0.3333  0.07972727
>>> Ga  0.000   0.000   0.09090909
>>> N   0.000   0.000   0.12518181
>>> Ga  0.3333  0.3333  0.13636363
>>> N   0.3333  0.3333  0.17063636
>>> Ga  0.000   0.000   0.18181818
>>> N   0.000   0.000   0.21609090
>>> Ga  0.3333  0.3333  0.22727272
>>> N   0.3333  0.3333  0.26154545
>>> Ga  0.000   0.000   0.27272727
>>> N   0.000   0.000   0.30699999
>>> Ga  0.3333  0.3333  0.31818181
>>> N   0.3333  0.3333  0.35245454
>>> Ga  0.000   0.000   0.36363636
>>>
>>> K_POINTS (automatic)
>>> 4  4  1  0  0  0
>>>
>>>
>>> For further information, this is a system with vacuum gap equivalent to
>>> 7
>>> GaN unit cells, the system with vacuum gap equivalent to 6 GaN unit
>>> cells
>>> worked fine. I am not getting why this problem is occurring only for
>>> few
>>> cases?
>>>
>>> Any kind of help or suggestions will be very valuable to me.
>>>
>>>
>>> Thanks
>>> -
>>> Manjusha
>>> Research Scholar
>>> Department of Chemistry
>>> Indian Institute of Technology,Kanpur
>>> India
>>> _______________________________________________
>>> Pw_forum mailing list
>>> Pw_forum at pwscf.org
>>> http://pwscf.org/mailman/listinfo/pw_forum
>>>
>>
>>
>>
>> --
>> Nguyen Van Chuong
>> PhD Student,
>> Department of Physics,
>> Don State Technical University.
>> Rostov on Don, Russia
>> Group Graphene
>> Email: chuongnguyen11 at gmail.com or cpuphys86 at gmail.com
>> Website: http://cpuphys86.webs.com/
>> Phone mobile: +7 905 45 94 888
>>
>>
>
>
> --
> Nguyen Van Chuong
> PhD Student,
> Department of Physics,
> Don State Technical University.
> Rostov on Don, Russia
> Group Graphene
> Email: chuongnguyen11 at gmail.com or cpuphys86 at gmail.com
> Website: http://cpuphys86.webs.com/
> Phone mobile: +7 905 45 94 888
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum



_______________________________________________
Pw_forum mailing list
Pw_forum at pwscf.org
http://pwscf.org/mailman/listinfo/pw_forum

-------------- next part --------------
A non-text attachment was scrubbed...
Name: winmail.dat
Type: application/ms-tnef
Size: 5386 bytes
Desc: not available
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20131224/d4d987ec/attachment.bin>


More information about the users mailing list