Replies: 7 comments 6 replies
-
There are some errors here. Firstly you're missing the The main error is in the species blocks. When you run this code, you get the error message:
As discussed in the ionisation section of the species block documentation, you must supply the code with an atomic number for it to determine which atom/ion species you are modelling. I have included a corrected input deck at the end of this message. If there is contradictory information in the documentation, let me know! This code runs, and my 4-core laptop estimates the job to take ~3 hours to run to completion.
|
Beta Was this translation helpful? Give feedback.
-
Thanks Stuart,
but better to convert all the names to lower case then it will look even easier
Program received signal SIGSEGV: Segmentation fault - invalid memory reference. 2a) Do the code with ionization trying to performs some additional calls of "diagnostics" function compared to the case with no ionization? Because i output all needed variables i do not need it to do anything besides may be the option to restart the code for continuation. I removed these lines too but this did not help number_density = always + species If the ionization calls diagnostics subroutine for its purposes what keywords it uses to distinguish itself from other calls? I will try to ignore these calls based on this keyword 2b) Before for many years I was using completely different compiler and in my compiler when the error occurs i compile it with debugging switch and when the code crashes it starts the debugger and stops in it exactly on the offending line. How to debug this code in Linux? |
Beta Was this translation helpful? Give feedback.
-
One more question about EPOCH ionization model: does the electron which ionizes ion lose the portion of its energy equal to ionization potential of ion? In other words, is total energy conserved ( kinetic energy of electron plus ionization energy of ion are the same before and after collisional ionization act) Looks like Segmentation crash i mentioned in previous message was caused by these lines, which may conflicted with my addition to the diagnostics subroutine which doing for me the output of essential PIC code particle variables (xm, pm, weight, and ID)
|
Beta Was this translation helpful? Give feedback.
-
Stuart,
|
Beta Was this translation helpful? Give feedback.
-
Thanks Stuart. |
Beta Was this translation helpful? Give feedback.
-
Hey @DanRRRR, I mananged to find some time to put together a page of ionisation demos. Let me know if you run into any problems with these simple test cases. Cheers, |
Beta Was this translation helpful? Give feedback.
-
Thanks Stuart, i already saw that but had no time to investigate. The way EPOCH makes output of ionization data will cause problems with my codes. What it missing is the general way of output of ionization data without details what caused what. In short as an example -- EPOCH starts with two species, electrons and say gold and at the end can create additionally 78 (!!!) new species instead of just still having these two species with different Z. Mostly no one cares how they were created to give to new ionization stage some special name (if you include recombination for example then there will be huge amount of back and forth ionizations and recombinations so there often is no sense to know how and what created this particular ion to justify special name for it - just its degree of ionization is enough to know for 99.999% cases. With ID you can follow each particular ion with no problem). 78 is just the start. If you include photorecombination then there will be 78 more. If include three-body recombination 78 additionally. Auto-ionization will add its 78. The code will stop running and will never end a single step We discussed this with Chris couple years back and he suggested how to implement output in this simple way -- when you know about species only their initial name and degree of ionization -- but noted that this method was not debugged in his time when he was actively developing the code. Additionally there exist some way to speedup the code with ionization sometimes by an order of magnitude. We will discuss that when i will finally make through existing method of output |
Beta Was this translation helpful? Give feedback.
-
Here is the deck i was trying to create out of the one mentioned in other post which fails to start. That is maximum what is possible to do reading other decks and 100 times the manuals.
Where is the error and why it does not even start ?
Beta Was this translation helpful? Give feedback.
All reactions