LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

Blinking graphs after several hour of run

I basically wrote a spectrum anylizer/logger. After several hours of running, my graphs starts blinking... in that it turns black then shows the graphs turns black again. It runs fine in the beginning.

I don't know where to begin to debug this issue. I checked the memory usage using windows task manager, but it seems to be fine. I thought I was writing out data to a too large of a file so I limited the file size having it start a new file, but it's doing the same thing. Any ideals?


Message Edited by d1sturbanc3 on 03-07-2008 08:00 AM
0 Kudos
Message 1 of 7
(2,966 Views)

You have not given us enough info to be of much use.

Standard follow-up question is "Can you post a stripped down example that demonstrates this?"

BUt lacking information I will guess. Smiley Surprised

Spectrum Analysis usually invlove large data sets. So how much data is being plotted how often?

Sometimes "defer FP updates" followed by "hide graph" then "apply new data", "unhide graph" then "Undefer FP updates" can help.

What type of graph are you using?

What are the spec of the machine?

What version of LV?

Still tring to help,

Ben

Retired Senior Automation Systems Architect with Data Science Automation LabVIEW Champion Knight of NI and Prepper LinkedIn Profile YouTube Channel
0 Kudos
Message 2 of 7
(2,960 Views)

Man your fast... I was switching machines so I can post my code and give you specs. Coming back to edit it. I've attached my code. The VI you need to run is "Front Planel.vi"

I'm running this on a sever 3.25 gb of ram. Quad cpu running at 3 ghz. Over 1.5 TB of harddrive space. I'm using a NI board PCI 4462. LV latest stable 8.5. Nothing else running in background.

It's basically 4 graphs. 1 SFTF spectrogram. 1 continuous updating waterfall. 1 Power average just a chart. 1 general XY. I have 2 outputs. One is just the raw data. The other is the XY chart.

I'm trying to run this with out write, but it's going to take several hours to see if it will do it again.


Message Edited by d1sturbanc3 on 03-07-2008 08:13 AM

Message Edited by d1sturbanc3 on 03-07-2008 08:15 AM

Message Edited by d1sturbanc3 on 03-07-2008 08:16 AM



Message Edited by d1sturbanc3 on 03-07-2008 08:20 AM
0 Kudos
Message 3 of 7
(2,954 Views)
oh yea.. stupid timeout on not allowing me edit
 
How often it's getting update: I'm reading in 1 pulse/sec trigger from a GPS unit. It's gathering data at 200KHZ.
0 Kudos
Message 4 of 7
(2,944 Views)
Hi d1sturbanc3,

The project you've attached contains alot of VIs. This makes it really hard for people to help you troubleshoot the issue. Is it possible for you to strip down your main VI (eg. removing sub-VIs, etc...) while still illustrating the problem?

Regards,
Stanley Hu
National Instruments
Applications Engineering
http://www.ni.com/support
0 Kudos
Message 5 of 7
(2,885 Views)
Thanks. I solved it (I think). I was in the process of stripping it down.... and tried to use a binary write instead of the express VI. The express VI was what was causing it.
0 Kudos
Message 6 of 7
(2,871 Views)
"The express VI was what was causing it."
 
The LAVA bumper sticker on my car reads "LAVA, no Express VI's required." Smiley Wink
 
Thanks for the update!
 
Ben
Retired Senior Automation Systems Architect with Data Science Automation LabVIEW Champion Knight of NI and Prepper LinkedIn Profile YouTube Channel
0 Kudos
Message 7 of 7
(2,866 Views)