Plotting a node value of point mesh in the time domain

[samwyss]

Hello all,

I apologize if a similar question has already been asked before that I missed while searching. If so, please feel free to link it below.

I have a simulation that uses the .silo data format to store time series potential difference values (across a membrane) as a function of theta. These potential values are stored as a Pointvar1 on top of a Pointmesh of the corresponding angle. At each time step the current potential for all angles is written to a .silo file with the Pointmesh and Pointvar (angle and potential respectively) timestamped using the DTIME option in a DBoptlist as is standard for the format.

In VisIt, I have been trying to view these membrane potential values in the time domain. So far, I have only been able to view this mesh evolve with time using the VCR window which plots the potential across all angles for a given time step.

I would like to make a plot of the potential of a single node (multiple nodes if possible later) for all time steps. So far I have tried using Controls->Pick->Time Pick->Do time curve with next pick which yielded the following warning

The Curve plot of variable "Scalar_Curves/potential" yielded no data.

As well as with Controls->Query->Pick->Do time query which was unable to find my data.

Is there something I am missing or should I be storing this data in another way that is more conducive to time series plotting?

Thank you!
Sam

[JustinPrivitera]

Correct me if I am misunderstanding - you have a single silo file for each timestep, and each one has DTIME. You can use the timeslider and associated buttons to view the mesh and variables at different timesteps. Is that right?

So your issue is with query over time? What variable are you specifying for time query? Does Scalar_Curves/potential exist across all timesteps? If you want, you can share your files with me and I can have a look. I am justin on LC or you can attach them here in a tarball or zip.

[samwyss]

Hi @JustinPrivitera ,

Sorry to keep bothering you about this. I have also attempted to use DBCurve's as well to no avail.

I attached another dataset below with a 1D time varying DBCurve named potential alongside a Quadvar1 named potential_field attached to a Quadmesh named mesh as an example. In this dataset, I have no issue querying any point in potential over time; however, cannot replicate this for 1D time varying data.

Based on my understanding, it seems like VisIt is attempting to query an (x,y) point (which might only exist for a single time step) over time for 1D datasets instead of querying the x location and plotting y vs t.

Do you have any other recommendations? The silo file format alongside VisIt is wonderful for representing the more "complex" (Fields/Mesh and from parallel IO) data from our simulations, I am just struggling to get the "simple" 1D data vs time plotting down. Worst case, I could write a python script to visualize these 1D datasets; however, I wanted to see if I could do this in VisIt prior to resorting to that option.

out2.zip

Thanks again!

[JustinPrivitera]

Hi @samwyss. It's no problem!

Again you could try the Curve File Format.

I'm confused about what you are trying to accomplish. Are you trying to take your curve and, for a single x value, plot how the y values for that x value change over time?

[samwyss]

Hi @JustinPrivitera ,

Thanks again for the quick response, and yes that is exactly what we are trying to do.

For context we are modeling biological membrane physics. With certain symmetries, the membrane of a cell is a pure function of an angle $\theta$. There are several quantities that only exist on/across said membrane (e.g., electric potential, tension, etc.) and not across the whole simulation domain thus it cannot be (to my knowledge) associated with the higher dimensional ($r$, $\theta$) mesh of the whole domain in the form of a 2D Quadmesh/Quadvar1. I was hoping there would be a way to easily view some property defined on the membrane, $\phi$ (as a placeholder variable name), which is in the form of a DBCurve or Quadvar1 defined on a 1D Quadmesh at some angular location $\theta$ as a function of time $t$ in the form of a $\phi$ vs $t$ plot.

I will take a look at the Curve file format post holiday, thanks for the recommendation! Feel free to close this thread or mark this as the answer in the meantime.

Thanks again for everything!

[JustinPrivitera]

I understand. I am having trouble querying your data, likely running into the same issues you have already seen. I will discuss with the other members of the team next week and get back to you. Enjoy Thanksgiving!

[samwyss]

Thank you and you as well @JustinPrivitera !