Learning Forum

[Alisa Bannaron]

I created a simulation file of s triangular deflector in CHARGE and wanted to see the electrical field behaviour between each triangle (that has a separation). On y-axis, the simulation was possible, however, once I tried the simulation (2D z-normal), this error occurred. Could you please suggest what could be the cause and how to solve it?

 

Thank you!

[Amrita Pati]

Hi Alisa,

The error messages in CHARGE can be very generic sometimes. It can be a bit challenging to pinpoint the issue without running some tests, but we can look at some relatively simpler considerations and eliminate them:

1. Partition the simulation domain, this will enable the "Check" button. Click on "Check" to see if there are any error/warnings.
   
   If there are, can you send a screenshot?
2. Check that there is no gap in between the geometry objects. In my experience, the users sometimes leave a small gap between the objects (along one or more directions) which can cause errors. If there are any overlapping objects, make sure that their mesh order is properly set.
3. If you have extremely small features, make sure that the max and min edge length are small enough to resolve them. But at the same time, if the simulation domain is large and there are no fine features, but the edges are extremely small it can also cause error. To avoid this make the mesh coarser. 

Let me know what you find out from these.

Regards,
Amrita

[Alisa Bannaron]

Hi Amrita,

Thank you for your reply!

 

I have checked the points you suggested.

1. There are no warnings shown
2. My design has top triangular electrodes where each triangle has some separation (currently 3 um, but it showed the same error at 20 nm as well). Intentionally I created the gap between electrodes since I wanted to avoid the charge overlap between them in an actual device so in the simulation I want to see what the sufficient gap is. I attached the screenshot of my design here. Could this gap be the cause of error in 2D Z-normal simulation?

*On alternative triangles the electrical I set in BC is like +/-

3. As you mentioned about the mesh, here is how I set up the global mesh constraint now. the size of the triangle (from the tip to the base) is 20 um and the thickness is 0.6 um. I have played with the max and min a bit, but the error remains.

[Amrita Pati]

Hi Alisa,

Thanks for the additional details. Is there a material assigned in the gap? 

The min edge length seems extremely small to me, especially when the device is a few micrometers. Can you change the min edge length to 0.1 um and 0.5 um and see what happens?

Regards,
Amrita

[Alisa Bannaron]

Hi Amrita!

 

There is no material assigned in between triangles. With this setting, there was no problem running in the y-axis, only in the z-axis that I couldn't run the simulation. Should I add the other object and set as air to fill in between each triangles when simulating in z-axis?

I tried changing the min edge length while maintaining the max at 1 um, the same error still showed up. 

 

Regards,

Alisa

[Amrita Pati]

Hi Alisa,

In my experience unassgined simulation domains cause errors. I am surprised that the earlier simulation worked alright for you. But I will recommend you to assign materials to all the domains. You can add the material "Air" to your simulation. Then assign the background index as air:

After you do this, partition your simulation and check that all the domains have been assigned to a material as shown below:

Regards,
Amrita

[Alisa Bannaron]

Hi Amtira,

 

If you mean the materials for each element, they are assigned like the triangle is Au, and underneath is SiO2. So all domains already have their own material and the background material in the simulation region was None. 

 

 

I already changed the background material to air, but the error still occurred (boundary condition as close).

 

Regards,

Alisa

[Alisa Bannaron]

Hi!

 

Are there any updates on how can I fix the error?

 

Best regards,

Alisa