Hi,
I am processing MASW data using geopsy. yesterday, it was fine. but the next day, the dispersion curve plot from active fk filter doesnt show up. I restart reprocessing with different database and it is still not showing. try to use different data, still not working.
here is the data and the log. the sampling rate is 512Hz by using tromino seismometer
MASW bug
MASW bug
- Attachments
-
- combine_2m.txt
- (924.7 KiB) Downloaded 1926 times
-
- log error geopsy.txt
- (15.03 KiB) Downloaded 1861 times
Re: MASW bug
Can you post a snapshot to help identifying the issue?
Re: MASW bug
sure. instead of snapshot, please check this video.
https://drive.google.com/drive/folders/ ... sp=sharing
Re: MASW bug
You can try to increase the minimum distance to 1. Did you try the conventional FK? A high resolution with a high damping is nearly equivalent.
Re: MASW bug
Re: MASW bug
Hi,
I was able to reproduce your error. The "step value" for velocity sampling is far too high. I did not notice it at first. Switch to "step count" to check the number of samples along the velocity axis. For frequency it is also relatively high. For a log scale, the default value of 1.025 is usually fine.
The generated samples for a log scale are: 1.025^i with i varying from N1 to N2. N1 and N2 are automatically adjusted so that the first sample is greater or equal to the minimum value and the last sample is lower or equal to the maximum value, respectively. The advantage of this formulation is that the samples are always the same whatever the minimum and maximum values. It is easier for comparing curves or results from various processing. Hence, with a step of 16, there is only one velocity sample at 256 m/s. A minimum of two samples is necessary to plot something, this is the reason why you got NaN values only.
Best regards,
Marc
I was able to reproduce your error. The "step value" for velocity sampling is far too high. I did not notice it at first. Switch to "step count" to check the number of samples along the velocity axis. For frequency it is also relatively high. For a log scale, the default value of 1.025 is usually fine.
The generated samples for a log scale are: 1.025^i with i varying from N1 to N2. N1 and N2 are automatically adjusted so that the first sample is greater or equal to the minimum value and the last sample is lower or equal to the maximum value, respectively. The advantage of this formulation is that the samples are always the same whatever the minimum and maximum values. It is easier for comparing curves or results from various processing. Hence, with a step of 16, there is only one velocity sample at 256 m/s. A minimum of two samples is necessary to plot something, this is the reason why you got NaN values only.
Best regards,
Marc