Altair RISE

A program to recognize and reward our most engaged community members

Nominate Yourself Now!

How to resolve the error 31027 for this file?[fEKO v 2013.1]

half_arc_a_46.98_final_probe_4by1_array_0.46_lambda_spacing_final_coordinate.cfx

hi,

In some super computer I am getting result for this model without any error, but whenever I try to simulate this file in my own PC , once it showed license error, next time it showed memory error, and now it is showing error 31027 with the warnings 31096 that -Adaptfeko has reached minimum frequency stepping, results might not have converged". How can I get the simulation done for this file?

Find more posts tagged with

Feko

v2023.1

Accepted answers

Torben Voigt

Hi @sswh2 ,

I see you set the Convergence Accuracy to "High". Try with default or low.

If that doesn't help, it would be best to use discrete frequency range in this case.

Best regards,
Torben

Torben Voigt

Hi @sswh2,

Please have a look at this model: https://securefiletransfer.altair.de/link/jXqCOO4bQOo1tCUkfZcbtu

This is how I would do it. There are some frequencies where the s-parameter results look a bit suspicious, which could possible improved by further mesh refinement (but not sure!).

Note I deactivated StandardConfiguration1 and added a farfield request to the SParameterConfiguration. This way you'll get the individual farfields for each port. If you want to have the farfield with all ports active, you would have to re-activate StandardConfuguration1 again.

Best regards,
Torben

Torben Voigt

Hi @sswh2,

Yes, of course you can do that. But perhaps you should consider whether you really need the far field over the entire frequency range, or perhaps only for individual frequencies. That would of course save a lot of time.

Best regards,
Torben

All comments

Torben Voigt

Hi @sswh2 ,

I see you set the Convergence Accuracy to "High". Try with default or low.

If that doesn't help, it would be best to use discrete frequency range in this case.

Best regards,
Torben

sswh2

Okay thanks, I will try now with the default option and will let you know if it is running without error or not.

Torben Voigt

Hi @sswh2 ,

When using the adaptive frequency range, the solver basically calculates individual frequencies and uses a fitting algorithm to determine how many additional frequencies are required in the immediate vicinity. In rare cases, this can lead to an extremely large number of frequencies being calculated around one frequency. It helps to either limit the total number of frequencies (you have already limited this to 501 or 1001, which is a lot), or to determine a minimum increment between the frequencies. I would recommend the latter here (e.g. 100 MHz):

Alternatively, as mentioned, discrete frequency range.

Best regards,
Torben

Torben Voigt

Hi @sswh2,

Please have a look at this model: https://securefiletransfer.altair.de/link/jXqCOO4bQOo1tCUkfZcbtu

This is how I would do it. There are some frequencies where the s-parameter results look a bit suspicious, which could possible improved by further mesh refinement (but not sure!).

Best regards,
Torben

sswh2

@Torben Voigt sir thanks for your explanation! I need the radiation pattern when all ports are active, so I have requested the far-field by activating both the standard configuration and S-parameter. Is that okay? Can I get the actual radiation pattern when all ports are active by using this method?

Torben Voigt

Hi @sswh2,

Best regards,
Torben

sswh2

@Torben Voigt Thank you so much sir