PXIe-6259 maximum update rate vs settling time

[howmean]

Hi,


Currently I'm using a PXIe-6259 to generate voltage through single channel AO.
According to the specification: 

 Maximum update rate
  1 channel..................................... 2.86 MS/s
  2 channels ................................... 2.00 MS/s
  3 channels ................................... 1.54 MS/s
  4 channels ................................... 1.25 MS/s

 Settling time, full scale step
 15 ppm (1 LSB)...............................2 μs

First Scenario:
The settings in the DAQ assistant are as follow:
1. Single channel AO: a0
2. Sampling rate: 2 MS/s (delta t between 2 samples would be 0.5e-6 s)
3. Samples to write: 2000 (this would give me a 1ms resolution based on 2 MS/s) 
4. Generation mode: Continuous
5. Output voltage range: -5 to 5 V

Soon after I started to generate according to the above configurations, I received a warning and the generation is stopped: 

"Warning 200040

Possible reason(s):
Measurements: Sample clock rate specified is so high that it violates the settling time requirements for the generation.
Reduce the sample clock rate, or the accuracy of the generated signal might be compromised."

Second Scenario:
So I changed the "Sampling rate" to 1MS/s and "Samples to write" to 1000. 
Delta t between 2 samples become 1e-6 s. The generation works without any "Settling Time" warning.


I'm rather confused here as in both scenarios, delta t between 2 samples are definitely lower than the Settling Time of 2e-6s as specified in the specification.
However, "Settling Time" warning only occurred in the First Scenario.

Did I interpret the Settling Time wrongly?

Could someone help me with this?


Thanks.


howmean

[Boon Chen]

Hi Howmean,

Your interpretation of the settling time is not wrong, but note that it's for 1 LSB.

National Instruments specifies the maximum rate at which an analog output device can change from one voltage to another. However, NI does not specify how big of a voltage change the device can make in-between updates. For example, if you output an array of voltages such as 2.3, 2.4, 2.5, 2.6… at an update rate of 2.86 MHz you will see the 200040 warning. However, because you’re only making small voltage changes of 0.1 V, you can ignore the error. If you output an array of voltages such as -10, 10, -10, 10… at an update rate of 2.86 MHz you cannot ignore the warning because of the large 20V swing in-between updates. The resulting signal will likely end up resembling -5.321, 4.465, -4.994, 5.012… rather than -10, 10, -10, 10… The device simply does not have enough time to change the voltage level before the next sample.

Therefore depending on the signal you are trying to output, you probably can ignore the warning.

Let me know if you have further questions.

Regards,
Boon Chen

[howmean]

Hi Boon Chen,

Thanks for your reply. 

However, I have another question about the Digital Output using PXIe-6259 instead.

Please refer to my attached VI.

For the "Daqmx Timing (Sample Clock).vi", I have tried with both the "100kHzTimebase" and "10MHzRefClock" as the sample clock source.

I try to observe the changes in the digital output waveform by changing the "Sampling rate".

However, it looks like it does not affect the duration of the output waveform at all.

Since changing "Sampling rate" doesn't affect duration of the output waveform, I modified the "number of samples" instead.

By doing this I saw some changes.

May I know why is that so? 

It is rather confusing because the relationship between Sampling rate and Number of samples in analog voltage generation case is always straight forward.

If I set the sampling rate to 100k S/s and number of samples to 10k. I expect to see an output waveform with duration of 100ms.

Thanks again.

Howmean

[Boon Chen]

Hi Howmean,

Did you change the rate & number of samples for the DAQmx timing VI or did you change that of the sub Generate Dig WDT Data.vi?

Since you are only looking at line0 of the digital waveform, it's simply an alternating waveform between 0 and 1. If you did not change the settings for the sub Generate Dig WDT Data.vi, I expect that you will see the same waveform being output. Normally we use the same number of samples input for both the DAQmx Timing VI and the sub Generate Dig WDT Data.vi.

For a clearer picture, I would suggest that you refer to our shipping example named Write Dig Chan - Ext Clk.vi

Regards,

Boon Chen