Re: [Ifeffit] Error in second shell fit
On Thursday, February 21, 2013 09:05:51 AM Ivan Stijepovic wrote: Just a thought and one question regarding freeze option in Athena. Would it be better to set rmax in Athena to some higher value in the first place, e.g. 7, and then use freeze group option? The error in Artemis would be avoided and one could still use freeze. Does it sound right? Hi Ivan, Your suggestion avoids Dev's immediate problem, but introduces two more problems. The best solution is for me to change the behavior of Artemis so that it turns off the freeze flag when data are imported from an Athena project file. The problems are (1) this just freezes the rmax value to a different number. One still will want to change that value in Artemis. (2) Rmax is used to compute Nidp (number if independent points). Set to 7, Nidp will be computed wrongly, as will all the statistical parameters. B -- Bruce Ravel bra...@bnl.gov National Institute of Standards and Technology Synchrotron Methods Group at NSLS --- Beamlines U7A, X24A, X23A2 Building 535A Upton NY, 11973 Homepage:http://xafs.org/BruceRavel Software:https://github.com/bruceravel ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Re: [Ifeffit] Error in second shell fit
I see. Thanks for clarification! On Thu, Feb 21, 2013 at 7:31 PM, Bruce Ravel bra...@bnl.gov wrote: On Thursday, February 21, 2013 09:05:51 AM Ivan Stijepovic wrote: Just a thought and one question regarding freeze option in Athena. Would it be better to set rmax in Athena to some higher value in the first place, e.g. 7, and then use freeze group option? The error in Artemis would be avoided and one could still use freeze. Does it sound right? Hi Ivan, Your suggestion avoids Dev's immediate problem, but introduces two more problems. The best solution is for me to change the behavior of Artemis so that it turns off the freeze flag when data are imported from an Athena project file. The problems are (1) this just freezes the rmax value to a different number. One still will want to change that value in Artemis. (2) Rmax is used to compute Nidp (number if independent points). Set to 7, Nidp will be computed wrongly, as will all the statistical parameters. B -- Bruce Ravel bra...@bnl.gov National Institute of Standards and Technology Synchrotron Methods Group at NSLS --- Beamlines U7A, X24A, X23A2 Building 535A Upton NY, 11973 Homepage:http://xafs.org/BruceRavel Software:https://github.com/bruceravel ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit -- Best wishes, Dr Ivan Stijepović Department of Materials Engineering Faculty of Technology University of Novi Sad Bul. cara Lazara 1 21000 Novi Sad Serbia tel: +381 21 485 3750 fax: +381 21 450 413 e-mail: ivan.stijepo...@gmail.com ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Re: [Ifeffit] Error in second shell fit
Dev, I work with older Ifeffit version thus I can not help you much - I can not open (Artemis) or work (Athena) with projects... I have only one question - do you need to fit paths about 4A? I hope that someone on this mailing list will check your files regards kicaj W dniu 13-02-19 17:47, Devender pisze: Dr. Kicaj, I tried that too, went till 8 A, but it is still showing same error whereas R effective for second shell paths are 4.4 and 4.7 A. I have also attached athena file and artemis project file for my data for reference. -Dev On Tue, Feb 19, 2013 at 10:57 AM, Dr. Dariusz A. Zając ki...@ifj.edu.pl mailto:ki...@ifj.edu.pl wrote: Hi, could you check if the same error appears if you give the Rmax of, let's say, 8A or even 6.5A? For higher R, the paths sometimes does not look like a single gaussian peak, but has some kind of satellites. You have to also remember that each path consists of the real and imaginary part. Regards kicaj W dniu 13-02-19 16:42, Devender pisze: Hi, I am trying to fit Bi2Te3 exafs data using artemis. I am able to fit first shell (screenshot of fit is attached, I have also attached 'atoms' and 'feff' files for reference) but when I am trying to include second shell, I am getting the error (attached file- Fit_error.log). Error reads that- The R effective for this path is well beyond the rmax value of its Data object. but I am already using Rmax of 6 A and Reff of these path are less than 5 A in my fit. I will appreciate if anyone can guide me or point me where I am making mistake. Regards, -- Devender Graduate Student, Materials Science and Engineering Rensselaer Polytechnic Institute, Troy, NY Website https://sites.google.com/site/devendermaun/ ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov mailto:Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov mailto:Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit -- Devender Graduate Student, Materials Science and Engineering Rensselaer Polytechnic Institute, Troy, NY Website https://sites.google.com/site/devendermaun/ ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Re: [Ifeffit] Error in second shell fit
Dev, Rmax in the Feff card limits the size of the cluster, it means no. of atoms, used in calculation of paths, whereas Rmax in the Fourier transform window limits the fitting region in the R space. Could you tell which values of delr for these paths you obtained in the fit? Is the same variable for all paths? regards kicaj W dniu 13-02-20 16:54, Devender pisze: Dr. Patridge, Thanks for suggestion. I am not sure if I get completely, I do understand the difference of Rmax in artemis and feff. You mean make changes in rmax in feff input file? I did that and attached are screenshots for that but I am still getting the same error. Let me know if I interpreted your suggestion wrongly. Dr. Kicaj, I am studying doping problems and it's important I get fitting from second and subsequent shells as there are new peaks that shows up at higher radial distances. Regards, Dev On Wed, Feb 20, 2013 at 8:26 AM, Christopher Patridge patri...@buffalo.edu mailto:patri...@buffalo.edu wrote: Dev, I don't know if my email was received. The think the problem is what you interpret as rmax. Rmax in artemis is the maximum value of R that you fit the data with, whereas rmax in feff is the largest distance from the center absorbing when calculating all the paths. The reason changing the rmax in feff does remove the error is that you must change the fitting range in (D)Artemis. I think the default vaue is ~ 3 A. Chris Christopher J. Patridge, PhD NRC Post Doctoral Research Associate Naval Research Laboratory Washington, DC 20375 Cell:315-529-0501 tel:315-529-0501 On 2/20/2013 3:13 AM, Dr. Dariusz A. Zając wrote: Dev, I work with older Ifeffit version thus I can not help you much - I can not open (Artemis) or work (Athena) with projects... I have only one question - do you need to fit paths about 4A? I hope that someone on this mailing list will check your files regards kicaj W dniu 13-02-19 17:47, Devender pisze: Dr. Kicaj, I tried that too, went till 8 A, but it is still showing same error whereas R effective for second shell paths are 4.4 and 4.7 A. I have also attached athena file and artemis project file for my data for reference. -Dev On Tue, Feb 19, 2013 at 10:57 AM, Dr. Dariusz A. Zając ki...@ifj.edu.pl mailto:ki...@ifj.edu.pl wrote: Hi, could you check if the same error appears if you give the Rmax of, let's say, 8A or even 6.5A? For higher R, the paths sometimes does not look like a single gaussian peak, but has some kind of satellites. You have to also remember that each path consists of the real and imaginary part. Regards kicaj W dniu 13-02-19 16:42, Devender pisze: Hi, I am trying to fit Bi2Te3 exafs data using artemis. I am able to fit first shell (screenshot of fit is attached, I have also attached 'atoms' and 'feff' files for reference) but when I am trying to include second shell, I am getting the error (attached file- Fit_error.log). Error reads that- The R effective for this path is well beyond the rmax value of its Data object. but I am already using Rmax of 6 A and Reff of these path are less than 5 A in my fit. I will appreciate if anyone can guide me or point me where I am making mistake. Regards, -- Devender Graduate Student, Materials Science and Engineering Rensselaer Polytechnic Institute, Troy, NY Website https://sites.google.com/site/devendermaun/ ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov mailto:Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov mailto:Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit -- Devender Graduate Student, Materials Science and Engineering Rensselaer Polytechnic Institute, Troy, NY Website https://sites.google.com/site/devendermaun/ ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov mailto:Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov mailto:Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov
Re: [Ifeffit] Error in second shell fit
On Wednesday, February 20, 2013 10:54:02 AM Devender wrote: Dr. Patridge, Thanks for suggestion. I am not sure if I get completely, I do understand the difference of Rmax in artemis and feff. You mean make changes in rmax in feff input file? I did that and attached are screenshots for that but I am still getting the same error. Let me know if I interpreted your suggestion wrongly. Dr. Kicaj, I am studying doping problems and it's important I get fitting from second and subsequent shells as there are new peaks that shows up at higher radial distances. Regards, Dev Dev, I get the impression from your emails that you have not much availed yourself of the resources that are already available for learning how to use the software. At http://bruceravel.github.com/demeter/ you will find the Artemis manual. I'll admit, it's still pretty mediocre, but it's a start. You will also find some videos that go a long way towards explaining how the software works. You may find it useful to spend some time with those things. I certainly think some time spent with those resrouces would obviate a problem like I am not sure if I ... completely ... understand the difference of Rmax in artemis and feff. That, frankly, isn't a very good question. While it is certainly true that Artemis suffers from being a program written by a physicist who has no training in computer science, user experience engineering, or design, you would have found the answer to that question even in the fairly crappy Artemis manual. B -- Bruce Ravel bra...@bnl.gov National Institute of Standards and Technology Synchrotron Methods Group at NSLS --- Beamlines U7A, X24A, X23A2 Building 535A Upton NY, 11973 Homepage:http://xafs.org/BruceRavel Software:https://github.com/bruceravel ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Re: [Ifeffit] Error in second shell fit
Well that is interesting, As was asked earlier, what is the fitted reff from the model? It would be rather non-physical and probably suggest a poor model, but perhaps the bond length is shifted well beyond the rmax which you show sets as 7 A and that is why you are receiving the error. Chris Christopher J. Patridge, PhD NRC Post Doctoral Research Associate Naval Research Laboratory Washington, DC 20375 Cell: 315-529-0501 On 2/20/2013 1:11 PM, Devender wrote: Bruce and Chris, Thanks for follow-up. I understand the frustration on your part. Reason for doubt is I am doing exactly the same thing Chris suggested even before posting the question here i.e. changing the rmax. I have changed rmax to even 9 A whereas shell I am trying to fit is only 5 A (please see attached screenshot) but still getting error log that reff is larger than rmax. I know by default it is set to 3 A but I have tried to change it values more then reff of path but still getting the same error. Please see attached screenshot where rmax I am using is 7 A where as Reff is only around 5 A but still getting the error log. Regards, Dev On Wed, Feb 20, 2013 at 12:49 PM, Bruce Ravel bra...@bnl.gov mailto:bra...@bnl.gov wrote: On Wednesday, February 20, 2013 10:54:02 AM Devender wrote: Dr. Patridge, Thanks for suggestion. I am not sure if I get completely, I do understand the difference of Rmax in artemis and feff. You mean make changes in rmax in feff input file? I did that and attached are screenshots for that but I am still getting the same error. Let me know if I interpreted your suggestion wrongly. Dr. Kicaj, I am studying doping problems and it's important I get fitting from second and subsequent shells as there are new peaks that shows up at higher radial distances. Regards, Dev Dev, I get the impression from your emails that you have not much availed yourself of the resources that are already available for learning how to use the software. At http://bruceravel.github.com/demeter/ you will find the Artemis manual. I'll admit, it's still pretty mediocre, but it's a start. You will also find some videos that go a long way towards explaining how the software works. You may find it useful to spend some time with those things. I certainly think some time spent with those resrouces would obviate a problem like I am not sure if I ... completely ... understand the difference of Rmax in artemis and feff. That, frankly, isn't a very good question. While it is certainly true that Artemis suffers from being a program written by a physicist who has no training in computer science, user experience engineering, or design, you would have found the answer to that question even in the fairly crappy Artemis manual. B -- Bruce Ravel bra...@bnl.gov mailto:bra...@bnl.gov National Institute of Standards and Technology Synchrotron Methods Group at NSLS --- Beamlines U7A, X24A, X23A2 Building 535A Upton NY, 11973 Homepage: http://xafs.org/BruceRavel Software: https://github.com/bruceravel ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov mailto:Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit -- Devender Graduate Student, Materials Science and Engineering Rensselaer Polytechnic Institute, Troy, NY Website https://sites.google.com/site/devendermaun/ ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Re: [Ifeffit] Error in second shell fit
On Tuesday, February 19, 2013 11:47:25 AM Devender wrote: I tried that too, went till 8 A, but it is still showing same error whereas R effective for second shell paths are 4.4 and 4.7 A. I have also attached athena file and artemis project file for my data for reference. Dev, I finally got to sit down and look at your problem in detail. I now appreciate why you are so confused. In Athena, you used the freeze group feature. That's good. Frozen groups are an excellent way of making sure that you do not accidentally change something you don't want changed. I am glad you are using that feature of Athena and I encourage you to continue doing so. Unfortunately, in Artemis, I did not take care to unset the flag that freezes a group. When you attempted to change the value of Rmax for the fit, as Darek, Chris and I were suggesting, it didn't work because I had forgotten to have Artemis unset that flag freezing the data group. Oops! I have already fixed this problem and this fix will be in the ne4xt release. As the immediate work-around, what you should do is open Athena, untick the freeze button for the data you want to analyze, save the Athena project file, then re-import the data into Artemis. Afterwards, I think you will find that changing the Rmax value will have the effect that we were talking about. Thank you for bringing this problem to my attention. It's a very serious problem that needs to be fixed. And now it will be! Cheers, B -- Bruce Ravel bra...@bnl.gov National Institute of Standards and Technology Synchrotron Methods Group at NSLS --- Beamlines U7A, X24A, X23A2 Building 535A Upton NY, 11973 Homepage:http://xafs.org/BruceRavel Software:https://github.com/bruceravel ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Re: [Ifeffit] Error in second shell fit
Hi Bruce, I tried that quickly, and no more error log! Fitting for second and third shells is working. Appreciate your help and going over my files. Freeze option really helps when I am trying to analyze systematic data like doping, where I have added incremental amount of dopants, and want to compare them together to notice changes without letting athena vary parameters for different scans. Regards, Dev On Wed, Feb 20, 2013 at 4:59 PM, Bruce Ravel bra...@bnl.gov wrote: On Tuesday, February 19, 2013 11:47:25 AM Devender wrote: I tried that too, went till 8 A, but it is still showing same error whereas R effective for second shell paths are 4.4 and 4.7 A. I have also attached athena file and artemis project file for my data for reference. Dev, I finally got to sit down and look at your problem in detail. I now appreciate why you are so confused. In Athena, you used the freeze group feature. That's good. Frozen groups are an excellent way of making sure that you do not accidentally change something you don't want changed. I am glad you are using that feature of Athena and I encourage you to continue doing so. Unfortunately, in Artemis, I did not take care to unset the flag that freezes a group. When you attempted to change the value of Rmax for the fit, as Darek, Chris and I were suggesting, it didn't work because I had forgotten to have Artemis unset that flag freezing the data group. Oops! I have already fixed this problem and this fix will be in the ne4xt release. As the immediate work-around, what you should do is open Athena, untick the freeze button for the data you want to analyze, save the Athena project file, then re-import the data into Artemis. Afterwards, I think you will find that changing the Rmax value will have the effect that we were talking about. Thank you for bringing this problem to my attention. It's a very serious problem that needs to be fixed. And now it will be! Cheers, B -- Bruce Ravel bra...@bnl.gov National Institute of Standards and Technology Synchrotron Methods Group at NSLS --- Beamlines U7A, X24A, X23A2 Building 535A Upton NY, 11973 Homepage:http://xafs.org/BruceRavel Software:https://github.com/bruceravel ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit -- Devender Graduate Student, Materials Science and Engineering Rensselaer Polytechnic Institute, Troy, NY Website https://sites.google.com/site/devendermaun/ ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
Re: [Ifeffit] Error in second shell fit
Hi, could you check if the same error appears if you give the Rmax of, let's say, 8A or even 6.5A? For higher R, the paths sometimes does not look like a single gaussian peak, but has some kind of satellites. You have to also remember that each path consists of the real and imaginary part. Regards kicaj W dniu 13-02-19 16:42, Devender pisze: Hi, I am trying to fit Bi2Te3 exafs data using artemis. I am able to fit first shell (screenshot of fit is attached, I have also attached 'atoms' and 'feff' files for reference) but when I am trying to include second shell, I am getting the error (attached file- Fit_error.log). Error reads that- The R effective for this path is well beyond the rmax value of its Data object. but I am already using Rmax of 6 A and Reff of these path are less than 5 A in my fit. I will appreciate if anyone can guide me or point me where I am making mistake. Regards, -- Devender Graduate Student, Materials Science and Engineering Rensselaer Polytechnic Institute, Troy, NY Website https://sites.google.com/site/devendermaun/ ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit ___ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit
