Replies : 4 Last Post : March 27, 2018 (Tue) 09:03:05

348

LMD energy

Posted on : March 27, 2018 (Tue) 06:40:53

by fati

Dear Administrator,

I got an error when i want to calculate the LMD energy.
the error:
***err in gengpt...nrpt too large

my fmg input file:
../data/CoMnTiSi 1 2 3 4
../data/CoMnTiSi_lmd 1 -1 1 -1 2 -2 2 -2 3 -3 4 -4

my input:
#---------------------------------- 0 ----------------------------------
# now # can be used as comment lines as well
#--------------------------------------------
go data/CoMnTiSi_lmd fcc 10.828129914557,1.0,1.0,90.0,90.0,90.0,
0.002 1.8 sra vwn mag 2nd
update 5 100 0.020
4
#--------------------------------------------
# name of type, number of components, rtr,
# field, lmax, atomic number, concentration
#--------------------------------------------
CoMn 4 0 0.0 2 27 48
27 48
25 2
25 2
MnCo 4 0 0.0 2 25 48
25 48
27 2
27 2
Ti 2 0 0.0 2 22 50
22 50
Si 2 0 0.0 2 14 50
14 50
4
0 0 0 CoMn
1/2 1/2 1/2 MnCo
1/4 1/4 1/4 Ti
3/4 3/4 3/4 Si

Thank you in advance.

 
 

352

[Re:04] LMD energy

Posted on : March 28, 2018 (Wed) 07:13:46

by fati

dear Administrator,
Thank you for your help but even though I did all these steps I find this problem:
Note: The following floating-point exceptions are signalling: IEEE_INVALID_FLAG IEEE_DIVIDE_BY_ZERO

 

351

[Re:03] LMD energy

Posted on : March 27, 2018 (Tue) 21:33:05

by Administrator

Dear Fati,

first, remove your data file
> rm data/CoMnTiSi_lmd
and start from scratch.

Best regards,

KKR admistrators

 
 

350

[Re:02] LMD energy

Posted on : March 27, 2018 (Tue) 21:19:49

by fati

thank you for your help but when i started the calculation i got this error:

--> ***** self-consistent iteration starts *****
CoMn MnCo Ti Si
***err in cstate...no convergence for nclr= 0 j= 1
NaN NaN NaN NaN NaN NaN
NaN NaN NaN NaN NaN NaN
NaN NaN NaN NaN NaN NaN
NaN NaN NaN NaN NaN NaN
NaN NaN NaN NaN NaN NaN
NaN NaN NaN NaN NaN NaN
NaN NaN NaN NaN
Note: The following floating-point exceptions are signalling: IEEE_INVALID_FLAG IEEE_DIVIDE_BY_ZERO IEEE_OVERFLOW_FLAG

Also when i Increase the value of nrpmx and the value of nk1 (for large bzqlty) i got this problem:
--> Erreur de segmentation

 

349

[Re:01] LMD energy

Posted on : March 27, 2018 (Tue) 09:03:05

by Administrator

Dear Fati,
two things have been noticed.
1) in your input,
CoMn 4 0 0.0 2 27 48
needs a blank at the 1st column. Otherwise, the line is regarded as a comment line.
2) In source/specx.f find the lines that look like
parameter
& (natmmx=6, ncmpmx=6, msizmx=96, mxlmx=4, nk1x=1250, nk3x=300,
& msex=201, ngmx=15, nrpmx=650, ngpmx=650, npmx=350, msr=400)
Increase the value of nrpmx, e.g. nrpmx=1000. Compile the code by the command "make" and try again.
3) However, the best way is use the latest version of AkaiKKR, cpa2002v010. The new version does not cause any this kind of trouble.

KKR administrators

 
 

POST NEW TOPIC

ID (Email)

Password

このたび、AkaiKKRのページをセキュリティ対策の一環として、システムを変更いたしましので、BBSへの書き込みはアカウント登録が必要です。
お手数かけて申し訳ございませんが、登録ページから新規登録をお願いいたします。

For security reasons, we have renewed the AkaiKKR website. Now, you need your account to send a message to BBS. If you do not have an account yet, please register.