Error using CFC 600 filter class

[sirwin]

Hello,

I'm trying to filter some data using the CFC 600 filter in the Crash Analysis Toolkit. When I try to use the filter, I get the attached error message.

I know my sampling frequency is 10 kHz, so why does Diadem say it's only 10 Hz?

Am I using the filter wrong?

[cdaug]

Hi Sirwin,

Do you have any example files you used when making the CFC calculation? What channel types are time and 11TRRI01MIWSDSY0? I'm assuming numeric for both. The sampling frequency should be calculated based on the step in the X channel (or the x-part of a waveform channel).

Best,

Chris D. | Applications Engineer | National Instruments

[sirwin]

Both channels are numeric. Time is increasing monotonic from -100 ms to +500 ms in 100 us steps. The Y channel is sampled at the same rate as Time and is in millimeters.

I saw in the properties for the Time channel that the step width is showing as 0 (see attached). Is this why the calculation fails?

The data set is from a crash test and is confidential. I'll check into if I can post a sanitized version.

[cdaug]

Hi Sirwin,

Are you able to create a numeric channel that starts at -100, and is 6000 length with 100 us steps? This would replace the channel and reset the unit. I'm afraid the one now is going from -100 to 500 seconds instead of ms, so the 6000 steps are 0.1 s instead of 100 us. This obviously wouldn't be the final answer, I'd just like to see if you saw any different behavior.

Best,

Chris D. | Applications Engineer | National Instruments

[sirwin · Accepted solution]

I figured it out. It looks like the CFC filters expect time to be in seconds.

I first tried to create a numeric channel in ms, and got the same error. I created a second time channel in seconds, and got the filter to work.