Why does data disappear between iterations

[Alv]

I'm continuously acquiring data from a accelerometer and displaying the data in a waveform graph (0 - 100Hz). Why does the data disapear between averages (I'm assuming between each iteration of the while loop).

I would like my data to remain in display until the averaging is complete.

I seen mention of this however cannot find it. Does anyone have a solution/example?

R/Alv

[chilly_charly]

Let's have a look to your code : build a simplified version of your vi (for instance removing the whole DAQ stuff, just leaving the loop, the graph, and a piece of data with the averaging function); see if the problem persists ; then post the vi... somebody will find the flaw.

CC

[Alv]

Charly,
Thanks. I enclosed a bmp of the code rather than re-wiring it again. I'll explain whats going on, basically I'm acquiring data from 2 transducers (accels) and coverting the data to displacement. From the code you can see that I'm displaying the frequency content as well as displacement. Give me thoughts.
R/

[Dennis_Knutson]

If you are using a graph like you say in your original post, then data disappears because a graph does not have a history buffer. That is what a chart is used for. If you want to use a graph, then you can create your own buffer by writing to a shift register.

[chilly_charly]

Sorry, may be I am tired after a busy Sunday, but I don't even see where the graph is on the bmp ! Don't know how I could help !!

CC

[Alv]

Dennis,
I'm using the shift registers on the while loop. I'm sure what your referencing to? I've included my code for reference.

[Alv]

Charly,
Icluded my code. R/

[Dennis_Knutson]

The only shift registers you have on the while loop are for the error cluster and the file reference. You need a shift register for each graph.

[chilly_charly]

Sorry Denis, but this has nothing to do with shift registers : a graph olds its data as long as you don't write anything else in it ! 😉

The flickering is due to the way the averaging is done, 2 subvi layers down. I don't have enough time to dig something out of there, but what can be done in the main vi is to add a case structure around the graph terminal, and wire the case input to a comparison node between the "magnitude" in the "power spectrum" cluster and an empty array.

See the attached vi, which is the simplified version I was asking for in my first answer.

CC

[Dennis_Knutson]

But he does a complete data acquisition every loop which would appear to write new data to the graph every time.