How to generate two voltage waveform under the internal clock as analog output using PCIe6363

[johnd77]

Hello, I use the PCIe6363 and BNC2110 to generate the continued voltage waveform under the internal clock as my analog output.I have test the sample example "Cont Gen Volt Wfm-Int Clk" from the path C:\Users\Public\Documents\National Instruments\NI-DAQ\Examples\DAQmx ANSI C\Analog Out\Generate Voltage, it works well to generate the voltage waveform, but when I want to generate two analog outputs in the program such that the Dev1/ao0 and Dev1/ao1 under the internal clock at the same time, it returns error which is DAQmx Error: The specified resource is reserved, the operation could not be completed as specified.

In the program what I have done is just to copy the sample example and add another virtual channel, as follows:

// Configure the tasks
DAQmxErrChk (DAQmxCreateTask("",&TaskHandle0));
DAQmxErrChk (DAQmxCreateTask("",&TaskHandle1));

// Setting the two virtual channels
DAQmxErrChk(DAQmxCreateAOVoltageChan(TaskHandle0,"Dev1/ao0","",-10.0,10.0,DAQmx_Val_Volts,NULL));
DAQmxErrChk(DAQmxCreateAOVoltageChan(TaskHandle1,"Dev1/ao1","",-10.0,10.0,DAQmx_Val_Volts,NULL));
DAQmxErrChk(DAQmxCfgSampClkTiming(TaskHandle0,"",1000,DAQmx_Val_Rising,DAQmx_Val_ContSamps,1000));
DAQmxErrChk(DAQmxCfgSampClkTiming(TaskHandle1,"",1000,DAQmx_Val_Rising,DAQmx_Val_ContSamps,1000));

// Write code
DAQmxErrChk(DAQmxWriteAnalogF64(TaskHandle0,1000,0,10.0,DAQmx_Val_GroupByChannel,data,NULL,NULL));
DAQmxErrChk(DAQmxWriteAnalogF64(TaskHandle1,1000,0,10.0,DAQmx_Val_GroupByChannel,data,NULL,NULL));
// Start Code
DAQmxErrChk (DAQmxStartTask(TaskHandle0));
DAQmxErrChk (DAQmxStartTask(TaskHandle1));

So I want to what is problem in the program or what I should do to generate two analog outs in the same time under the internal clock.

Thanks!!!

[PTerro]

Hi johnd77,

You should try putting both analog output channels in the same task.  Your code would read:

// Configure the tasks
DAQmxErrChk (DAQmxCreateTask("",&TaskHandle0));

// Setting the two virtual channels 
DAQmxErrChk(DAQmxCreateAOVoltageChan(TaskHandle0,"Dev1/ao0:1","",-10.0,10.0,DAQmx_Val_Volts,NULL));
DAQmxErrChk(DAQmxCfgSampClkTiming(TaskHandle0,"",1000,DAQmx_Val_Rising,DAQmx_Val_ContSamps,1000));

// Write code
DAQmxErrChk(DAQmxWriteAnalogF64(TaskHandle0,1000,0,10.0,DAQmx_Val_GroupByChannel,data,NULL,NULL));
// Start Code
DAQmxErrChk (DAQmxStartTask(TaskHandle0));

Side note:
"Dev1/ao0:1" is the same as putting "Dev1/ao0, Dev1/ao1"

[johnd77]

Hi Peter,

I have try you solution but the program cannot compile correctely. It goes to the debug mode and also returns the Unhandled exception at 0x6dd5cc19:0x00000005: Access violation reading location 0x02437000 at the DAQmxWriteAnalogF64(taskHandler0,1000,0,10,DAQmx_Val_GroupByChannel,data,NULL,NULL),

I think it should be some error like the bad ptr and do you have some idea about this problem when I set the "Dev1/ao0:1"in the one task handle.   

[PTerro]

Hi Johnd77,

Did you input your taskHandler0 correctly?  It doesn't match up with what you had before which was "DAQmxErrChk(DAQmxWriteAnalogF64(TaskHandle0,1000,0,10.0,DAQmx_Val_GroupByChannel,data,NULL,NULL));" with "TaskHandle0" instead of "taskHandler0".  Can you send me a screenshot of the exception?  That error is a Windows error not a specific National Instruments error.

[johnd77]

Hi Peter,

You are right that I make some mistake with the "taskHandler0" and "TaskHandler0",but when I correct this mistake  and compile the program again, I get another mistake which is "Attempted writing analog data that is too large or too small" , Status Code:-200561.

I have no idea with this error because when I change the code and just use only one vitural channel like"Dev1/ao0"instead of"Dev1/ao0:1", the program works well.

I have attached the screen shot of the error and my main function here and I hope it will help you to figure out the problem.

[PTerro]

Hi johnd77,

So this error is more to do with writing over the max limit.  It is weird that you are getting a value that is -95.5 when you shouldn't get anything more than 9.95 with that equation.  Can you try running a variation on the addressing and see if it works, such as "Dev1/ao0, Dev1/ao1".  Here is our page on how to define the physical channels and they have a variety of ways you can do it:

http://zone.ni.com/reference/en-XX/help/370466V-01/mxcncpts/physchannames/

[johnd77]

Hi Peter,

I have try different ways to variate the addressing method

"Dev1\ao0:1"

"Dev1\ao0:Dev1\ao1"

"Dev1\ao0,Dev1\ao1"

reverse order

"Dev1\ao1:0"

"Dev1\ao1:Dev1\ao0"

"Dev1\ao1,Dev1\ao0"

But those all fail with the same error:"Attempted writing analog data that is too large or too small" and when I dedug the program

it fails at the line:

DAQmxErrChk (DAQmxWriteAnalogF64(TaskHandle0,1000,0,10.0,DAQmx_Val_GroupByChannel,data,NULL,NULL));

I also check the data array range (data max is 9.9499999 and the data min is -9.9499999 ). 

I feel so confused about this problem!

[johnd77]

Hi Peter,

I have figured out the problem by setting the data array to data[2000] and the sin wave form function is writeen as follows:

for(;i<2000;i++)
data[i] = 9.95*sin((double)i*2.0*PI/1000.0);

from above I can get two analog outputs (amplitude +/-9.95 and frequency 1Hz, sin waveform)in two virtural channels.

I have totally no idea with the reason why it works and do you know the trick inside?

But anyway it works now and I want to thank you very much for your time and help!

[PTerro]

Hi Johnd77,

I honestly have no clue why you were getting the error you were and why 2000 points vs 1000 points worked!  I went ahead and checked the values in Excel with the given inputs and you should not of been getting -95 for an entry.  I am glad to hear your program works though!!

[FoxSquirrel]

Hey Johnd77,

I was able to run the same code on my system with no issues.  The only things I could think of was that I initialized my data array (This is always a good idea).  Also, if 2000 is working for you, what I might suggest doing is to make it an array of 1001 and just initializing it correctly by declaring:

float64 data[1001] = {0};