Problem with XY Graph looking like a scribble

[MikeBowen]

Two questions:

1) The in range and coerce module, the wire that runs from the In-range? Output, does that delete data not in range?

2) I don't understand the use of imaginary values when constructing the plot.

Mike

[altenbach]

---

@MikeBowen wrote:

Two questions:

1) The in range and coerce module, the wire that runs from the In-range? Output, does that delete data not in range?

2) I don't understand the use of imaginary values when constructing the plot.

Mike

---

1. it will output the value coerced to the limits, but since the conditional tunnels skips all out of range data it does not matter either way.
2. XY plots understand a rich set of datatypes:

• An array of points (each "point" is a cluster of scalars x and y)
• A scalar cluster containing x and y 1D arrays
• A 1D complex array where RE=x and IM=y (my personal preference!)
• combinations of the above for multiple plots (but all plots must be of the same type!)
• etc.

[MikeBowen]

You lost me on the tunnel.

Mike

[altenbach]

Here's the light at the end of the tunnel... 😄

The autoindexing output tunnel skips elements where the boolean is false.

(right-click...conditional)

[MikeBowen]

The final for loop, does that code modify the curve?  The picture I got was very sinusoidal.  

[dsb@NI]

Recommendations:

 - Format your data file to be consistent.

   - Same number of columns in every row

   - Have a dedicated column for each sensor

   - Sort data by date.?

 - In your code, if you want to process and view the data chronologically, you should sort each sensor/channel by the date/time

 - In your display, you can just show points:

Edit: Sorry, I was staring at a cached version of the thread when I started responding, so I missed that Altenbach and OP have been active.

[altenbach]

---

@MikeBowen wrote:

The final for loop, does that code modify the curve?  The picture I got was very sinusoidal.  

---

No, it does not modify any data, it just unbundles the data as sorted by time.

I don't see anything "sinusoidal". Are you talking about the slow or fast oscillations?

[MikeBowen]

I took a look at the first rev code which plotted against time index, and it was sinusoidal in nature in that it was up and down.  This data has the same repetition, so it's my bad.

The other plots are intermittent in nature, which makes it hard to see that data.  I probably need to blow the data up and look at it slowly.  Any recomendations on how to look at the notes?  When I pull out the data I will add the notes and make that available, but it could involve a hundred or so points to look at.

Mike

[altenbach]

If the fast oscillations are noise, you could filter the data.

---

@MikeBowen wrote:

  Any recomendations on how to look at the notes?  When I pull out the data I will add the notes and make that available, but it could involve a hundred or so points to look at.

---

I would just add the note (I assume there is only one note for each time point and time points don't contain duplicates) as a string to the cluster. Each cluster will have a time, an array of values from the various sensors, and a string for notes.

[altenbach]

---

@altenbach wrote:

Each cluster will have a time, an array of values from the various sensors, and a string for notes.

---

Here's how that could look like....