spectrum measurement node NaN error

[JAvela555]

I just noticed my input array has an NaN variable, that might be it. its at 2000.

[altenbach]

Yes, it is at index 2001. Value should be 1.

Still, even replacing that does not make the entire exercise more meaningful. What are you trying to get out of it?

Why do you do averaging if you only run it once?

Your windowing focuses on the data that is mostly zero.

Maybe some JTFA might be more informative. Do you have the advanced signal processing toolkit?

[JAvela555 · Accepted solution]

I use a split 1d array node to capture the data I want later on in the code. So i only get the data I need. I ended up just replacing it with one and it worked but I'm facing another issue at the moment where the graph saves previous data for some reason. I don't believe I have that toolkit. I'll look into it.

[altenbach · Accepted solution]

---

@JAvela555 wrote:

 I'm facing another issue at the moment where the graph saves previous data for some reason.

---

A graph does not have a memory. Maybe you are suing a chart instead of a graph? Fir a chart, you can clear the history data at program start).

If you have previous data on a graph, it might mean that you are building the data in an uninitialized shift register.

[JAvela555]

yeah, for some reason turning off the averaging box in the spectral measurements node did the trick. No previous memory saved.

[altenbach]

Or you can just wire a true in the first iteration (iteration terminal, =0 ) to the init terminal. Still, for a single run, averaging makes no sense anyway.

[billypeoples]

It sounds like you're having trouble with NaN errors when passing an array to the spectrum measurement node. One possible fix could be checking your data to ensure it's in the right format. Sometimes NaN errors happen when the data isn't properly formatted for full spectrum extraction. Try double-checking your array and the parameters you're using to see if everything matches up correctly.