LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

fast AO response upon front panel change with PCI-6713

Hi,

 

I purchased the PCI-6713 analog output card. I tested some Labview examples for continuous wave generating and so on. In the example "function generator.vi" I can output for instance a square wave and change the amplitude on the front panel. I am very surprised however that there is a delay of about 6 seconds between a change on the front panel and the actual corresponding change of the analog output, which is of coarse much to slow to use in fast PID systems as is claimed in the brochure of the PCI-6713 card.

My question is how can I use the PCI-6713 card to continuously generate for instance a square wave, and change the amplitude in each cycle of the loop (corresponding to one period of the square wave) in less than -lets say- 10ms? (or at least in less than 6 seconds) ?

 

Kind regards,

Filip

0 Kudos
Message 1 of 6
(2,598 Views)
Hi Filip,

I can't find an example that has the name "function generator.vi". Perhaps because i am using an other labview version and it is named different.
Is it possible to provide me the vi so i can have a look at it?
Kind Regards,

Joris Donders
National Instruments
EMEIA GTM Lead for Semiconductor
www.ni.com/semiconductor
0 Kudos
Message 2 of 6
(2,561 Views)
Hi,
 
In attachment "function generator.vi". I use Labview 6.1
 
Filip
0 Kudos
Message 3 of 6
(2,555 Views)
Hi Filip,

About the 6 second delay. Did you changed anything in the "function generator.vi"?
And when you change the amplitude of your signal, is it first displayed on the front panel,
or does your output change first.

I am asking you this because i have trouble opening the example. All our pc's have Labview 8.5 installed
and 6.1 is just abit to long ago to be compeletly compatible. So i need to make an image of a pc then uninstall
Labview 8.5 and install an older version. I can open the vi with 8.5 but many subvi's are missing and i can't run
it so i can't see if i have a problem.

So let me know if you changed anything? From what i have seen there isn't anything wrong with the vi.

I think the problem is the following. When we select continuous generation we make the hardware fill a buffer with samples.
so the buffer can be read every time at the same moment and the signal appears continuous. But let's say we change the signal,
then it will only disapear after all the pervious buffer items are read out. So i think if you make the buffer smaller the delay should be smaller.
Perhaps you can take a look at that too.

Kind Regards,

Joris Donders
National Instruments
EMEIA GTM Lead for Semiconductor
www.ni.com/semiconductor
0 Kudos
Message 4 of 6
(2,545 Views)
Hi,
 
 
Maybe it is not important that you run this particular vi, i am pretty sure that there is nothing wrong with the vi! However, if I understand you correctly, the output of the newly written data in the buffer will not appear immediately but only the next time the buffer is read. This is not what I observe, namely that the old buffer is outputted many times (>20) before the new data is outputted.
A smaller buffer size indeed decreased the delay, but not as much as I had hoped for...
 
Would it help to use a double buffered approach? If so, how can I implement this?
 
Best regards,
 
Filip
0 Kudos
Message 5 of 6
(2,529 Views)
Hi Filip,

about the double buffered approach, this is possible. You can use the search on the web to find many explination for it.
one that could help you for example is this one:

http://zone.ni.com/devzone/cda/epd/p/id/2802

It's for digital output but well, it's just to show you how to implement it.

But also take a look at this article:

http://zone.ni.com/devzone/cda/tut/p/id/3993

Here they explain some stuff and talk about the automatic regeneration of the data, so perhaps this also could be a cause of your problem.
That's why i suggest you perhaps take a look at this example maybe this one does exactly what you want to do:

http://zone.ni.com/devzone/cda/epd/p/id/4872

It's about continuous reconfigurable waveform generation. And because it's the reconfigurable part we are having trouble with I think it can help.
Kind Regards,

Joris Donders
National Instruments
EMEIA GTM Lead for Semiconductor
www.ni.com/semiconductor
0 Kudos
Message 6 of 6
(2,514 Views)