08-03-2023 09:42 AM
I save graph data this way:
Measurement is at 10000Hz, with a chart history length of 100000.
I read the graph data like this:
The read graph is also 100000 history length.
It is only 1000 measure points.
What do i do wrong?
Thank you in advance and for reading.
08-03-2023 10:05 AM
I tried to replicate your condition. I generated a waveform with the same number of points, it read well.
Can you confirm when you save the data, the number of points going to the file ?
08-03-2023 12:02 PM
We cannot really debug pictures. A good guess is that the local variable contains stale data.
Why exactly are you using charts instead of graphs?
Please attach some simplified code that demonstrates the problem.
08-04-2023 03:16 AM
We know now the problem is at the write part. This helps.
The spreadsheet file that is generated is also only 1000 measure points.
08-04-2023 03:23 AM - edited 08-04-2023 03:23 AM
Hi Sydney,
@SydneyAVDE wrote:
We know now the problem is at the write part. This helps.
The "write part" is not the problem.
The problem is (as told before) the data in your local variable: when your waveforms only contain 1000 samples then you cannot save more samples in your files…
As you use charts and rely on their history: using the local you only save the latest data of the chart, but not the whole history!
Why do you need to convert an array to a cluster just to index the (waveform) elements in the array? Why not use IndexArray???
08-04-2023 08:07 AM
@SydneyAVDE wrote:
This helps.
...
The spreadsheet file that is generated is also only 1000 measure points.
Not this does NOT help! You've been asked to attach your code because showing truncated pictures is completely useless. We don't even know where the terminal of that local variable is!!!
If that local variables is of the chart terminal (You never really specified that and it is not obvious from the picture!), it will only contain one reading (which I assumes is maybe 1000 points). You can use a property node to read the history data instead.
Once we see the entire code, we can give specific advice. Maybe you want to append to file whenever the chart terminal is written. Maybe you want to keep the history using your own code?
As has been said already, the array-to-cluster-unbundle-dance is just pain silly. I also strongly recommend to avoid string-to-path and path-to-string. You should process paths exclusively using build path and strip path, only the filename should be processed as string. Your current code would burst into flames as soon as you would run it on a different OS (e.g. Mac or linux).
08-04-2023 08:59 AM - edited 08-04-2023 09:07 AM
I have not have the time yet to make a simplified version. Programming in labview is not the main function of my job. I am trying what i can with what little i learned 5 years ago at school.
I started using charts instead of graphs because the graph that came with the DAQ example only showed x samples while chart showed me all samples appended. And appended is what i was looking for. I understand there must be a better way i have not yet discovered.
(this is noise and not an actual measurement.)
By making the simplified version, i found out what the problem is.
I was so focused at new challenges that i forgot the very beginning of my program.
And by using the chart variable, i hoped i could save all data that is shown in the chart, instead of the last x samples. Just like when you export picture. I have lots to learn.
A new challenge has risen now and i will have to re-write a part of my program. But that's for next week. Have a good weekend all and thank you for your time.
08-04-2023 09:22 AM
I tried the history node but it offered me the same result (at first glance).
Here is my code. Bare in mind that i'm a beginner and it is a lot of trial and error. There are parts that should be ignored for now.
What is important right now, is that i can call upon a file that is viewable in labview, so that we can zoom in on parts of the wave. Spreadsheet files and pictures are not ideal, but that's all i have been able to generate so far.
E.g.
The signals have to be measured at at least 5500Hz (to be tested), and maybe more, for 15 seconds.
Also, the part that you mention of the string-to-path and path-to-string... i am guessing you will malfunction when you see my code 😂 I will check in the build path and strip path as you suggested.
08-04-2023 10:02 AM
Sorry, I currently don't have access to LabVIEW 2023. Can you do a "save for previous" (2020 or below). and attach again.
You can build your data in a shift register and use a graph. (since t0 and dt are fixed, maybe keeping plain array data would be sufficient).
08-09-2023 03:59 AM
Hello
Here is the file as requested.