USB 6009: DAQmxSetAIConvRate causes error -200452

[MTR1a]

I don't see any spiking with one channel, and slowing down the sampling rate doesn't have any effect on spiking.  If you like, I can send you an executable program to try with another unit.

--Mike

[MickeyD]

Hi Mike,

Go ahead and build the executable if you could zip that and attach it I would be glad to run it with a 6009 here.  Make sure when you post the exe to include what the device name in MAX of your 6009 is, because I won't be able to change the exe so I will have to change my settings in MAX.  I will go ahead and grab the piece of hardware.

Have a great day,

Michael D
Applications Engineering
National Instruments

[MTR1a]

Michael, is there a way for me to send the attachment to just you?  I'd have to spend some time sanitizing the program before posting it for everyone to see.
 
--Mike

[MickeyD]

Hi Mike,

If you have a support contract with National Instruments, then you can call in or email in and just guide the person that gets your issue to this thread.  Then they will transfer the request to me, and we can converse by email.  If you do not have a support contract then I am going to be limited to interacting with you through the forumn method, in which case anything you post will be public.  Let me know how you want to proceed.  I can not get this to happen with a 6009, which implies yours is probably damaged in some way.  If all you are posting is an exe, then the block diagram is not accessible, so what would you have to change to put it on the forumn?  Let me know how you would like to proceed.

Michael D
Applications Engineering
National Instruments

[MTR1a]

Hi Michael,

I attached a zipped version of my executable.  I've tested it on Windows XP.  I'm using Dev1 for my usb6009.

When you run the program, a text window will pop open and also a graphics window.  You can use the left and right arrows in the graphics window to cycle through the channels (0-3).  The channel number isn't indicated anywhere, but the initial channel is 0.  Press 'Q' to quit the program.

The graphs show time series data taken in three-second snapshots.  The black trace is the raw data, and the red trace is filtered.  (If the signal is very clean, you might not see the black trace under the red one.)  This version of the program samples all eight channels as I described previously.  If I wire the same signal to channels 0 & 4, for example, the graphs for channel 0 look clean, but if I disconnect channel 4, then the black trace becomes spikey.

Thanks,
Mike

[MickeyD]

Hi Mike,

When I run the exe I get a message.  Warning wrong number of points read.  I assume I am not reading the proper amount of points to be updating the graph.  I just ran it with my USB 6009 plugged in as a Dev1   .   Is that a warning you have built into your code?

Thanks,

Michael D

[MTR1a]

That warning probably means there was a problem initializing the A/D system.  Have you confirmed that the USB 6009 is working using LabView or one of the logging utilities?

Thanks,
Mike

[MickeyD]

Hi Mike,

I can not find anything wrong with my USB 6009.  It is behaving as expected except when I call your exe.  I still get that same message, which is something your code is generating. I tried simulating the behavior you saw with my USB 6009 and did not get the channel spiking, so I would not be surprised if you device was broken.  I can not really do anything with the exe.  If you run one of our simple acquisition examples do you see the same behavior.  Just recording voltages on multiple channels.  You can find several acquisition examples by going to help find examples if you have LabVIEW.  The DAQmx examples are located under Hadware INput and Output » DAQmx.  Let me know if you see the same behavior with that.  If you want to run Visual C examples Go to the following folder.  C:\Program Files\National Instruments\NI-DAQ\Examples\DAQmx ANSI C\Analog In\Measure Voltage\Cont Acq-Int Clk

Let me know how you would like to proceed.

Have a great week,

Michael D
Applications Engineer
National Instruments

[MTR1a]

Thanks for trying, Michael.  It might not be worth it to diagnose the problem with my executable.  I did notice spiking behavior with the multichannel acquisition C example program (/NI-DAQmx Base/examples/ai/contAcquireNChan.c), so you might try that.  I modified the program slightly to print out more voltage readings (instead of just the first 10).  I had one channel wired to the 2.5V reference and the other to a potentiometer using the 5V reference.  The first channel occasionally showed values that were a volt or more different from 2.5V.

--Mike

[Brian F.]

Hello. 

I would recommend looking at the following knowledge base article and trying the steps described in it.  There are some very helpful tips on how to eliminate ghosting from your measurements.

How Do I Eliminate Ghosting From My Measurements?
http://digital.ni.com/public.nsf/allkb/73CB0FB296814E2286256FFD00028DDF

In the meantime, I will be testing a few of the DAQmx Base examples.  However, I don't suspect that I will see ghosting on the channels.  I believe that the issue is probably coming from one of the scenarios described in the linked article.

I will keep you updated of my progress.

Brian F
Applications Engineer
National Instruments