Need help: How to model a simulation of nano-machining with LAMMPS

Hello.
I am a new learner of LAMMPS.I am familiar with Linux, Now i am learning LAMMPS. This is my fiest letter for help about LAMMPS. My English is not good at the same time. Maybe my words fails to express my original meaning. I’ll try my best.
I know that I can make a simulation of nano-machining by LAMMPS, But I do not how to model it with LAMMPS, I have tried it a long time, But I feel that I still stand outside the door of LAMMPS. So I post this Mailing List, I just need some help. I think I need a simple example. Is there anyone who can email me a simple in file of LAMMPS as an nano-machining example?
Thanks in advance and thank you very much.
Best Regard from China.

Hello.
I am a new learner of LAMMPS.I am familiar with Linux, Now i am learning
LAMMPS. This is my fiest letter for help about LAMMPS. My English is not
good at the same time. Maybe my words fails to express my original meaning.
I'll try my best.
I know that I can make a simulation of nano-machining by LAMMPS, But I do
not how to model it with LAMMPS, I have tried it a long time, But I feel
that I still stand outside the door of LAMMPS. So I post this Mailing List,
I just need some help. I think I need a simple example. Is there anyone who
can email me a simple in file of LAMMPS as an nano-machining example?
Thanks in advance and thank you very much.

i think rather than examples, you need a (new? or second?) adviser or
a (senior) colleague with some experience. a mailing list is a very
bad replacement for that.

just an example input without much explanation is not really getting
you to the point where you want to go and it is not really needed to
have a specific example to find out how to build suitable input
scripts in LAMMPS. so you should go and study and understand the
examples that are already bundled with LAMMPS and also what you can
find elsewhere on the web as tutorials. even if they seem off-topic,
they will help you to get *fundamental understanding* in how to
construct an input in LAMMPS. it basically is the same as with
learning a language. if i would give you a book of great greek
literature (written in greek), it is not much help to you, if you
haven't learned reading greek first, right?

ciao from italy,
     axel.

Thanks for your reply very much.
I know that your advice is right. In fact I had read all the manual, about 1200 pages, especially some important commands, and a book ‘Molecular dynamics simulation Elementary methods’. Maybe this is not enough. The problem I faced is that there is no adviser or collegue around me. For my teacher this is a new region, too. Thanks for your reply. I will try my best to find someone who is familiar with LAMMPS, at the same time I hope someone in the List can help me.
About nano-machining, Now I can model it, but I has many other problems, for example, the atoms run off. So I just want to have a input file just as an example.
Maybe I need to read the LAMMPS’ Manual carefully, especially Important Commands, or which I would use.
Taking your time to reply my letter, I am very grateful. Best Regards.

Thanks for your reply very much.
I know that your advice is right. In fact I had read all the manual, about
1200 pages, especially some important commands, and a book 'Molecular
dynamics simulation Elementary methods'. Maybe this is not enough. The

what you should do is to first practice *elementary* MD and follow
tutorials not just try to get your specific science project going.
people make the same mistake all the time. they think they are wasting
time learning something that they don't need for their research
project, but they neglect that you have learn how to walk before you
can run. all the time spent getting a good understanding in your mind
of how the basic principles would, make it easier to cope with more
complex problems later. if you start with the more complex issues, you
will never know whether you have a problem that is due to a
fundamental misunderstanding of how MD works, just just they you are
using the keywords incorrectly.

so if you have a book about MD basics, look at the exercises that come
with it and try to implement them with LAMMPS. look at the examples
that are bundled with LAMMPS and the discussions of them in the manual
and elsewhere *and* experiment with them. come up with your own more
complex setups and see what errors can happen and why.

problem I faced is that there is no adviser or collegue around me. For my

people say this a lot, but i don't buy it. you just haven't asked the
right people.

teacher this is a new region, too. Thanks for your reply. I will try my best

then you have to make your teacher find an adviser, too. in my
personal opinion it is irresponsible behavior of an adviser to let a
student do work in a new field without having anybody at hand that can
provide suitable advice. this is bound to cause problems and lots of
wasted time through making avoidable mistakes and missing out on
important knowledge of the field. MD has been around for quite a
while, so there is a *huge* body of knowledge that you can tap into.

to find someone who is familiar with LAMMPS, at the same time I hope someone
in the List can help me.

please note that learning about LAMMPS and learning molecular dynamics
are two different issues. people often mix this up. so most of the
"science" problems are very independent of LAMMPS, so you don't need
to talk to a LAMMPS expert to get to know about this.

About nano-machining, Now I can model it, but I has many other problems, for
example, the atoms run off. So I just want to have a input file just as an
example.

as i mentioned before, just having an input doesn't teach you
anything. any more complex inputs have to build in stages. even very
experienced (in fact, *particularly* very experienced) people do this.
this way you can figure out at each step, whether it worked as
expected. again, this is not very LAMMPS specific but just a general
approach that a researcher has to adopt to be successful.

Maybe I need to read the LAMMPS' Manual carefully, especially Important
Commands, or which I would use.

reading is only 10% of the work. you have to practice and come up with
simple examples that use those commands, so that you understand how
they work.

axel.

Are you asking how to construct the data file?
    (How to arrange the atoms in 3-D?)

    Or are you asking what kind of force-fields and methods to use?

    As Axel said, it is difficult to design a credible simulation
without -anyone- to help you set up the computer(s), and recommend
force-fields for your material. I suspect you will need to spend a
months using google and reading papers. You will get stuck.
Frequently. How much time do you have to run this simulation?
Perhaps you can email professors who can collaborate with you and your
advisor? I would be afraid to spend months running a simulation and
later discover that your computer model is totally inappropriate.

    If you post questions to the mailing list, nobody can help you
unless you post much more detail, including your input scripts, and
LAMMPS' error messages (and perhaps your data files).

  However, before you post questions there, I recommend to search this
web page first:
http://lammps.sandia.gov/doc/Section_errors.html#err_3
    I also recommend that you open the source file referred to in the
error message. ("pair_style.cpp:306" refers to line 306 of the file
"pair_style.cpp" located in LAMMPS "src" directory). If you still can
not understand the error, then try emailing the mailing list. Please
do not email the list and only say "it does not work".

    --- moltemplate examples ---

    Moltemplate has a "delete" command which can be used to cut
rectangle-shaped holes in crystals. (Overlapping rectangles are
okay.) I don't know if this would help you. Here are some sample
images, and links to the moltemplate (.lt) files which define them.

http://moltemplate.org/images/translocation/solvent.jpg
http://moltemplate.org/examples/translocation/solvent.lt
http://moltemplate.org/examples/translocation/solvent_single.lt

http://moltemplate.org/images/translocation/walls.jpg
http://moltemplate.org/examples/translocation/walls.lt
http://moltemplate.org/examples/translocation/wall_single.lt

http://moltemplate.org/images/sierpinski/lvl3.jpg
http://moltemplate.org/examples/sierpinski/sierpinski_cubes.lt
http://moltemplate.org/examples/sierpinski/al_cell.lt
http://moltemplate.org/examples/sierpinski/Al99.eam.alloy

http://lammps.sandia.gov/threads/jpgVaqPjUge0K.jpg
http://lammps.sandia.gov/threads/msg34435.html

http://moltemplate.org/images/nanotube+water/nanotube+walls+water_top_nopbc_t=0.jpg
http://moltemplate.org/examples/nanotube+water/graphene_walls.lt
http://moltemplate.org/examples/nanotube+water/nanotube.lt

   You can download complete examples include sample LAMMPS input
scripts, force-field data, images, and instructions for viewing the
trajectories at www.moltemplate.org. (All of the systems shown above
are included. However, the force-fields used in these examples are
typically not realistic.)

    If your system has explicit bonds which cross the periodic
boundaries of the unit cell, then moltemplate will not presently work.
(In that case you may have to write a program to generate those
bonds, angles, dihedrals, etc. yourself.)

    Of coarse, you can always write a short program in python, perl,
Java, or C (for example) with nested for loops to generate the
coordinates yourself. If you are not familiar with the LAMMPS data
file format, then there are other programs like "pizza.py" (or
"topotools", or "raw2data.py" distributed with moltemplate) to convert
these coordinates into a LAMMPS DATA file format. (Email back if need
want help using raw2data.py.)

    If you need to create a simple rectangular box or a cone
containing a metallic crystal, the easiest way is to use LAMMPS script
commands:

http://lammps.sandia.gov/doc/create_atoms.html
http://lammps.sandia.gov/doc/region.html
http://lammps.sandia.gov/doc/lattice.html

Thanks for you all.
I think I have knew what I should do now.