Re: [Wien] Potential bug with initso_lapw in version 18.2

2019-07-03 Thread Peter Blaha
After some additional info I can verify the problem. It happens in cases WITH inversion symmetry, when you run init_so several times and the number of equivalent atoms has been changed by a previous init_so. The problem is in the init_so script, which does not update the original case.in2

Re: [Wien] Potential bug with initso_lapw in version 18.2

2019-07-03 Thread Peter Blaha
Hi, I cannot verify this problem. In my tests I always get a proper and complete in2c file. Can you send me privately your struct file and describe your magnetization direction, where the problem occurs. Am 03.07.2019 um 04:51 schrieb Huseyin Ucar: Dear Developers, I was having similar

Re: [Wien] potential

2016-09-02 Thread Yundi Quan
I'm doing spin polarized calculation. So I should delete case.clmcorup and and case.clmcordn and case.clmsum_old. Does it seem right? On Fri, Sep 2, 2016 at 6:02 AM, Peter Blaha wrote: > Yes. You can "miss use" mixer for that. > > Make sure there is no

Re: [Wien] potential

2016-09-02 Thread Peter Blaha
Yes. You can "miss use" mixer for that. Make sure there is no case.clmcore nor an clmsum_old file, just a case.clmval. x mixer will just convert the clmval file into clmsum. On 09/02/2016 02:58 PM, Yundi Quan wrote: Is there a way to automatically convert case.clmval into case.clmsum

Re: [Wien] potential

2016-09-02 Thread Yundi Quan
Is there a way to automatically convert case.clmval into case.clmsum format? case.clmsum has slightly different Y00 term from case.clmval. On Fri, Sep 2, 2016 at 1:26 AM, Víctor Luaña Cabal < vic...@fluor.quimica.uniovi.es> wrote: > On Fri, Sep 02, 2016 at 12:21:24AM -0700, Yundi Quan wrote:

Re: [Wien] potential

2016-09-02 Thread Víctor Luaña Cabal
On Fri, Sep 02, 2016 at 12:21:24AM -0700, Yundi Quan wrote: > The lowest value is -0.35234088194673E-02. I'm using modified Beck-John. I should not consider that numerical inaccuracy. Something else is at work. As Peter Blaha, says converging RMT's and kpoints can be quite important. If you can't

Re: [Wien] potential

2016-09-02 Thread Peter Blaha
Vcxoul can have pos and neg. values. Rho should be positive everywhere. However, when you plot the density of only one state, rho can get zero easily in large parts of the unit cell and thus by numerical inaccuracies even small !!! negative densities may pop up, in particular if you are not

Re: [Wien] potential

2016-09-02 Thread Yundi Quan
I wanted to plot only the charge density of a particular state. So I changed the k-mesh to only one point and then used x lapw1 -up -c x lapw1 -dn -c x lapw2 -vresp -c -all emin emax The resulting case.clmval is used as the input charge density. I know that the Y00 term in case.clmval is different

Re: [Wien] potential

2016-09-02 Thread Yundi Quan
The lowest value is -0.35234088194673E-02. I'm using modified Beck-John. Yundi On Thu, Sep 1, 2016 at 11:30 PM, Víctor Luaña Cabal < vic...@fluor.quimica.uniovi.es> wrote: > On Thu, Sep 01, 2016 at 11:15:32PM -0700, Yundi Quan wrote: > > What's the sign convention used in the output files of

Re: [Wien] potential

2016-09-02 Thread Víctor Luaña Cabal
On Thu, Sep 01, 2016 at 11:15:32PM -0700, Yundi Quan wrote: > What's the sign convention used in the output files of critic2? I notice > that there are positive and negative values in both vcoul.cube and > rho.cube. Maybe it is OK to have both positive and negative vcoul.cube. But > what does it

Re: [Wien] potential

2016-09-02 Thread Yundi Quan
gt; > Date: Thu, 1 Sep 2016 12:56:49 >> From: Yundi Quan <quanyu...@gmail.com> >> Reply-To: A Mailing list for WIEN2k users <w...@zeus.theochem.tuwien.ac. >> at> >> To: A Mailing list for WIEN2k users <wien@zeus.theochem.tuwien.ac.at> >> Subject: Re

Re: [Wien] potential

2016-09-01 Thread tran
yes On Thursday 2016-09-01 12:56, Yundi Quan wrote: Date: Thu, 1 Sep 2016 12:56:49 From: Yundi Quan <quanyu...@gmail.com> Reply-To: A Mailing list for WIEN2k users <wien@zeus.theochem.tuwien.ac.at> To: A Mailing list for WIEN2k users <wien@zeus.theochem.tuwien.ac.at>

Re: [Wien] potential

2016-09-01 Thread Yundi Quan
Does the procedure stay the same if I use modified Becke-John potential? Yundi On Wed, Aug 31, 2016 at 5:12 PM, Yundi Quan wrote: > Thanks. > > Yundi > > On Wed, Aug 31, 2016 at 11:27 AM, Peter Blaha < > pbl...@theochem.tuwien.ac.at> wrote: > >> As mentioned before, you

Re: [Wien] potential

2016-08-31 Thread Yundi Quan
Thanks. Yundi On Wed, Aug 31, 2016 at 11:27 AM, Peter Blaha wrote: > As mentioned before, you should not use case.vsp/vns. > > But there is a more or less "standard" method: > > Instead use the R2V option in lapw0 and create case.vtotal. > > The potential in this

Re: [Wien] potential

2016-08-31 Thread Peter Blaha
As mentioned before, you should not use case.vsp/vns. But there is a more or less "standard" method: Instead use the R2V option in lapw0 and create case.vtotal. The potential in this format can be used in lapw5 (see UG) to create V on a grid in a single plane. You can also use

Re: [Wien] potential

2016-08-31 Thread Alberto Otero de la Roza
Hi Yundi, You can write real-space grids of WIEN2k potentials, densities, or combinations of them using critic2 (https://github.com/aoterodelaroza/critic2). Use this input: crystal case.struct load wien case.vcoul case.struct vnorm id vcoul load wien case.clmsum case.struct id rho cube cell 50

Re: [Wien] potential

2016-08-31 Thread Yundi Quan
Thanks Professor Marks. What I'm trying to do is to visualize \psi(r)^* V(r) \psi (r) in real space for a layered material so as to show how the potential changes from layer to layer. On Wed, Aug 31, 2016 at 7:50 AM, Laurence Marks wrote: > I am 99% certain that

Re: [Wien] potential

2016-08-31 Thread Laurence Marks
I am 99% certain that there is nothing which will do this automatically for you, some work will be required. I am pretty certain that you can use the R2V option in case.in0 to output vtotal. You would then need to do something like use lapw5 to generate a number of different slices, or use lapw3

Re: [Wien] potential well depth

2015-02-26 Thread Peter Blaha
Now we start to understand what you want. You have to create manually some distorted structures (corresponding to the potential well you want to investigate) and calculate E-tot. Then plot E-tot vs. distortion. It is your task, to define the distortions (eg. in SrTiO3 it could be a simple rigid

Re: [Wien] potential well depth

2015-02-26 Thread Mohammed Abujafar
Dear WINE2k developers and users,Hi,I am concerning about reading the  value of the potential depth well from the DFT results ,for an example SrTiO3 compound in perovskite structure.How can I get it?Your help is highly appreciated.Thanks a lot in advance.With best regardsMohammed On