Inputting PSD Curve
Inputting PSD Curve
(OP)
I'm inputting a PSD curve that Ansys is not liking very well and I'd like to know if my workout is proper. Here is a segment of the initial:
Hz G^2/Hz
5 0.047
7 0.047
7 0.18
9 0.18
9 0.026
13 0.026
13 0.16
27 0.16
This is the first 8 points given on the PSD curve. When I input as shown above Ansys prompts me with a message for each duplicate frequency and when I issue the psdgraph command it shows nothing like the actual. As a work around I've been inputting the above as this for instance:
Hz G^2/Hz
5 0.047
7 0.047
7.1 0.18
9 0.18
9.1 0.026
13 0.026
13.1 0.16
27 0.16
I notice that when I issue PSDGRAPH for the above the G^2/Hz value will occasionally fall to a zero value for the small range which the function is not defined for since it is not continuous.
My question is: Is my workaround an acceptable way of entering the PSD curve without affecting results? Or would a better way be to keep a continuous function but omit the second data point at any given frequency. For instance:
Hz G^2/Hz
5 0.047
7 0.047
9 0.18
13 0.026
27 0.16
As I understand dynamics this very closely approximates the original PSD. I'm just concerned that Ansys may not properly input the curve the way I have entered it now. Thankyou for reading and any help on this matter is much appreciated!
Best,
-Brian Craig
Hz G^2/Hz
5 0.047
7 0.047
7 0.18
9 0.18
9 0.026
13 0.026
13 0.16
27 0.16
This is the first 8 points given on the PSD curve. When I input as shown above Ansys prompts me with a message for each duplicate frequency and when I issue the psdgraph command it shows nothing like the actual. As a work around I've been inputting the above as this for instance:
Hz G^2/Hz
5 0.047
7 0.047
7.1 0.18
9 0.18
9.1 0.026
13 0.026
13.1 0.16
27 0.16
I notice that when I issue PSDGRAPH for the above the G^2/Hz value will occasionally fall to a zero value for the small range which the function is not defined for since it is not continuous.
My question is: Is my workaround an acceptable way of entering the PSD curve without affecting results? Or would a better way be to keep a continuous function but omit the second data point at any given frequency. For instance:
Hz G^2/Hz
5 0.047
7 0.047
9 0.18
13 0.026
27 0.16
As I understand dynamics this very closely approximates the original PSD. I'm just concerned that Ansys may not properly input the curve the way I have entered it now. Thankyou for reading and any help on this matter is much appreciated!
Best,
-Brian Craig





RE: Inputting PSD Curve
Your initial PSD curve is:
Hz G^2/Hz
5 0.047
7 0.047
7 0.18
9 0.18
9 0.026
13 0.026
13 0.16
27 0.16
i.e. two values of acceleration for 7, 9 and 13 hz? Are you sure this is what you want to do? You cannot have more than one magnitude for a single frequency for PSD. If you have two PSD curves you can input these separately using PSDFRQ and PSDVAL.
------------
See FAQ569-1083 for details on how to make best use of Eng-Tips.com