Why does pyDPF report negative harmonic indices

When listing model data pyDPF shows negative values for Harmonic Index?

Number of sets: 30
Cumulative Frequency (Hz) LoadStep Substep Harmonic index
1 670386.325235 1 1 0.000000
2 872361.424038 1 2 0.000000
3 1142526.525324 1 3 0.000000
4 1252446.741551 1 4 0.000000
5 1257379.552140 1 5 0.000000
6 1347919.358013 1 6 0.000000
7 679667.393214 2 1 1.000000
8 679667.393214 2 2 -1.000000
9 899321.218481 2 3 -1.000000
10 899321.218481 2 4 1.000000
11 1128387.049511 2 5 1.000000
12 1128387.049511 2 6 -1.000000
13 708505.071361 3 1 -2.000000
14 708505.071361 3 2 2.000000
15 966346.820117 3 3 2.000000
16 966346.820117 3 4 -2.000000
17 1031249.070606 3 5 -2.000000

It looks like for repeated modes it lists one as negative, but how does it determine this (it is not always the second repeated mode is negative).

Find more posts tagged with

PyDPF

Accepted answers

Oscar

@Jim Kosloski, the negative sign of the harmonic index is calculated by the MAPDL solver for the cases where CYCOPT,MSUP key is On. This sign is used to properly get the sign of the duplicated modes for the cyclic expansion since we don't have duplicate sector solutions in the RST for this case.
We just output the harmonic sign as reported by MAPDL.

For more information
https://ansyshelp.ansys.com/account/secured?returnurl=/Views/Secured/corp/v242/en/ans_cmd/Hlp_C_CYCOPT.html?q=CYCOPT

All comments

Oscar

For more information
https://ansyshelp.ansys.com/account/secured?returnurl=/Views/Secured/corp/v242/en/ans_cmd/Hlp_C_CYCOPT.html?q=CYCOPT

Andy M.

@Jim Kosloski It is as Oscar said. I will add to it that there is a relationship between the base and duplicate sector modes when there are modes at the same frequency. That is, the base_sector_of_mode_1 = (sign) duplicate_sector_of_mode_2, and base_sector_of_mode_2 = (sign) duplicate_sector_of_mode_1. There is an indeterminancy here, so the signs are not predetermined and it depends on how the solver responds to the operating system, hardware, frequency range, matrices, etc. This is why the signs aren't always in the same order.