# "No X column is set" in Plot Data filter

**URL:** https://discourse.paraview.org/t/no-x-column-is-set-in-plot-data-filter/17706
**Category:** ParaView Support
**Created:** [August 29, 2026, 7:26pm UTC](https://discourse.paraview.org/t/no-x-column-is-set-in-plot-data-filter/17706 "2026-08-29T19:26:40Z")
**Posts on this page:** 3
**Page:** 1

<div class="post-metadata">

### Author: ![elynch](https://discourse.paraview.org/user_avatar/discourse.paraview.org/elynch/32/6418_2.png) [@elynch](https://discourse.paraview.org/u/elynch)
#### Post date: [August 29, 2026, 7:26pm UTC](https://discourse.paraview.org/t/no-x-column-is-set-in-plot-data-filter/17706/1 "2026-08-29T19:26:40Z")

</div>

I’m attempting to make a simple line plot of PressureCoeff\_AVG vs. Points\_X using the data in the attached file.

[avgTaps\_2.vtp](https://discourse.paraview.org/uploads/short-url/mOBKldeFVZtHgp8ELYAgxGsIvoC.vtp) (11.2 KB)

- Load data file
- Add a Plot Data filter, hit Apply
- Uncheck “Use Index For X Axis” and set X Array Name to Points\_X

I get an error that says "[paraview] vtkPlotPoints.cxx:560 ERR| vtkPlotLine (0xf320670): No X column is set (index 0).

I’ve seen this in both versions 6.0.1 and 6.1.0. The data I’m loading looks OK in the RenderView. Am I misusing this filter? The only workaround I’ve found is to insert a Python Calculator before the Plot Data filter that just copies points[:,0] to some other variable.

---

<div class="post-metadata">

### Author: ![mwestphal](https://discourse.paraview.org/user_avatar/discourse.paraview.org/mwestphal/32/17_2.png) [@mwestphal](https://discourse.paraview.org/u/mwestphal)
#### Post date: [August 31, 2026, 7:26am UTC](https://discourse.paraview.org/t/no-x-column-is-set-in-plot-data-filter/17706/2 "2026-08-31T07:26:23Z")

</div>

“Points” are not a true variable, which means it should not even be available, its a small bug.

You can work around it by adding the “Coordinates” filter before the PlotData.

---

<div class="post-metadata">

### Author: ![elynch](https://discourse.paraview.org/user_avatar/discourse.paraview.org/elynch/32/6418_2.png) [@elynch](https://discourse.paraview.org/u/elynch)
#### Post date: [August 31, 2026, 1:17pm UTC](https://discourse.paraview.org/t/no-x-column-is-set-in-plot-data-filter/17706/3 "2026-08-31T13:17:16Z")

</div>

I have a different data file containing the same geometry as the attached one, but with slightly different data. I do not run into this problem when plotting that data. I was puzzled about the difference between the two datasets, so I saved them as ascii to make comparison easier. The only difference I could see was that the “bad” data (the one I attached in my original post) has the points given as:

```auto
<DataArray type="Float64" Name="numpy_array" NumberOfComponents="3" format="ascii" RangeMin="85.5996210236873" RangeMax="90.13338330101483">

```

while the “good” data has the points given as:

```auto
<DataArray type="Float64" Name="Points" NumberOfComponents="3" format="ascii" RangeMin="85.5996210236873" RangeMax="90.13338330101483">

```

If I change “numpy\_array” to “Points,” I can plot without a problem.

So then I was curious about what led to “Points” vs. “numpy\_array.” Turns out, I had two versions of the same script. For the “good” data, I wrote the points using something like:

```auto
points = vtk.vtkPoints()
for i in range(n):
    points.InsertNextPoint(x[i], y[i], z[i])
poly = vtk.vtkPolyData()
poly.SetPoints(points)
...

```

For the “bad” data, I wrote the points using:

```auto
poly = dsa.WrapDataObject(vtk.vtkPolyData())
xyz = np.column_stack((x, y, z))
poly.SetPoints(xyz)

```