Message by Fredrik Johansson: I’m having minor and major issues with the Langmuir Probe.
First of all, since there is no longer the possibility of a “fake” Langmuir probe which I used to assign the stub, I’m currently using the following setup
*S/C: node 0, no instrument support
*LP Stub: node 1, instrument support id 1
*LP: node 1, instrument support id 2
sweeping on instrument id 2, node 1.
So, doing a LP sweep on instrument support id 2, will sweep the potential of node 1 and thus eliminating the need for a ‘fake’ Langmuir probe
Major Issue 1: Impossible to get photoemission current from LP, emitted currents.txt only list emission for node 1 (i.e. stub and probe). Since this only depends on material properties, sun distance and geometry, I can just set up a dummy simulation with separate nodes everywhere to every individual emission current. Tedious.
Minor issue 1: How can I import LP instruments from previous simulations? Setting them up is tedious and prone to human errors
Major issue 2.1: Sweeping once and then changing instrument population generates errors
Major issue 2.2: Several LP instruments on the same node generates errors after the first sweep finishes
The same error happens for both situations, although the java.lang.ArrayIndexOutOfBoundsException number changes everytime (I’ve gotten ‘4’, ‘22’, ‘51’, etc):
java.lang.ArrayIndexOutOfBoundsException: 51
at spis.Top.Transition.LangmuirProbeTransition.nextTime(LangmuirProbeTransition.java:458)
at spis.Top.Transition.LangmuirProbeTransition.performTransition(LangmuirProbeTransition.java:402)
at spis.Top.Transition.TransitionObserver.notifyTransitions(TransitionObserver.java:106)
at spis.Top.Simulation.SimulationFromUIParams.integrate(SimulationFromUIParams.java:3710)
at spis.Top.Simulation.SimulationFromUIParams.integrate(SimulationFromUIParams.java:4001)
at spis.Top.Top.Scenario.integrate(Scenario.java:106)
at spis.Top.Top.NumTopFromUI.integrate(NumTopFromUI.java:176)
at org.spis.ui.simulation.command.LaunchSimulationCommand$1.run(LaunchSimulationCommand.java:93)
at java.lang.Thread.run(Thread.java:695)
Message by Fredrik Johansson:
Some format and grammatical errors, apologies.
I meant to write that I can get the photoemission current from the probe by creating a separate dummy simulation, but this is not an ideal workaround.
Message by Pierre Sarrailh:
Dear Fredrik,
1/ Concerning the “fake” Langmuir probes:
Normally, this functionnality still exist just by setting in the instrument a popName = “noPop” (warning: with some OS, it is case sensitive).
There was a bug in a previous version but normally it is corrected in the 5.1.6 version.
You are true, if you need the same potential sweep on the LP and on the Stub, you don’t have to create a “fake” LP.
2/ Major Issue 1: Impossible to get photoemission current from LP
The LP instrument have been created to compute accuratelly the current for populution for which we need to increase the statistic and the precision.
As the emitted current of photo-emission is fixed in the material properties, there were no need to use this kind of instrument to compute this current.
But, you are that it is something to improve because it would be good to have also the total current on the LP instrument.
That you can do in you situation is to set: > node 1: the stub > node 2: the LP
And to short circuit node 1 and 2 in the electrical circuit ( V 1 2 0.0).
In that case, you obtain in the “emittedCurrent.txt” file a different current for LP and Stub and, due to the fact you have shaorrt circuited the nodes, the LP sweep also applies to the node 1 and 2 (Stub + LP).
3/ Minor issue 1: How can I import LP instruments from previous simulations
Something to improve … there is no way, for the moment, to import an instrument from another simulation.
4/ Major issue 2.1: Sweeping once and then changing instrument population generates errors
It is not possible to change the population during the course of a simulation.
5/ Major issue 2.2: Several LP instruments on the same node generates errors after the first sweep finishes
It is not possible to add several LP on the same electrical node because it creates conflics between the differents potential sweep (it leads to the exception you obtain).
This has also to be improved.
For now you have two solutions: > to use the standard ouputs for the current not measured by the LP > if you need to increase the precision on more than 1 current, you can create ParticleDetector (in mode 1) for the other population (with the same measurement period as the LP).
This has really to be improved to enhance the user friendlyness.
Thank you for this feedback, it is always very interesting to have that from user in oder to have a maintenance plan for SPIS.
Pierre