Help with triggering and sampling rate on USB-6341 DAQ

[ianscandela]

Yes, that makes sense as far as proper acquisition timing for the AI. The one thing I don't understand is the AO sample clock and triggering. You wrote: 

"Configure AO to be triggered by the rising edge of the external pulse while using the falling edge as a sample clock."

I don't understand this. What is the difference between setting the "active edge" of the sample clock and the "edge" of the Start Digital Edge trigger, if they use the same clock source? It seems redundant to me. If I set the sample clock of the AO to the falling edge of my 2ms hardware signal, it will generate a sample and send it out of the AO channel. What, then, does setting the trigger to the rising edge of the 2ms signal do? Won't this override the falling active edge I set on the sample clock, and instead generate an AO sample on the rising edge?

[Kevin_Price]

I think you're right that the AO trigger I recommended is redundant.  It doesn't really accomplish anything.  (Note that it doesn't hurt either.  It wouldn't "override" the sample clock settings.  Triggers and sample clocks are very distinct for NI DAQ devices.)

Kudos for catching that!

-Kevin P

[pkp401]

I am using the same USB DAQ 6314. I am using a laser which can send a trigger each wavelength step and minimum step is 0.2 pm. I am using the trigger in PFI channel and capturing the photodetector output in analog input channel configured as finite sampling, one sample per trigger. This limits the resolution to 0.2 pm. I need higher resolution, which is possible as the wavelength sweep is linear and trigger can be interpolated to get higher wavelength resolution. However, I am not sure how to execute this. 
If I use trigger in PFI line as a clock source, I cannot higher resolution.

One option is to acquire the trigger in an analog input channel simultaneously with the photodetector signal. But I am not able to execute this, as the analog edge triggering is not supported in this daq.

What should I do to perform this acquisition? I am using python to runt the DAQ.

I am slowly learning DAQ, the previous responses are a bit advanced. Could you please help me with this?

[santo_13]

---

@pkp401 wrote:

I am using the same USB DAQ 6314. I am using a laser which can send a trigger each wavelength step and minimum step is 0.2 pm. I am using the trigger in PFI channel and capturing the photodetector output in analog input channel configured as finite sampling, one sample per trigger. This limits the resolution to 0.2 pm. I need higher resolution, which is possible as the wavelength sweep is linear and trigger can be interpolated to get higher wavelength resolution. However, I am not sure how to execute this. 
If I use trigger in PFI line as a clock source, I cannot higher resolution.

One option is to acquire the trigger in an analog input channel simultaneously with the photodetector signal. But I am not able to execute this, as the analog edge triggering is not supported in this daq.

What should I do to perform this acquisition? I am using python to runt the DAQ.

I am slowly learning DAQ, the previous responses are a bit advanced. Could you please help me with this?

---

If you plan to interpolate between triggers, ditch triggers on DAQ altogether, connect the LASER trigger out to a Analog Input along with your photodetector line, capture both analog inputs at highest rate possible. Once you have this high time resolution waveform, post process it to find the trigger locations and pick the photodetector value. Now you can extend this to even pick interpolated values.

[pkp401]

The DAQ seems not to support the analog edge triggering. Is there a way to synchronise the acquisition between analog input channels, I was thinking to use the trigger in ai channel as a trigger process at least to start the acquisition at the same time. 
Could you suggest a way that i will have a better synchronisation?

[santo_13]

---

@pkp401 wrote:

The DAQ seems not to support the analog edge triggering. Is there a way to synchronise the acquisition between analog input channels, I was thinking to use the trigger in ai channel as a trigger process at least to start the acquisition at the same time. 
Could you suggest a way that i will have a better synchronisation?

---

You don't need triggers between analog input lines, all that you need to do is put multiple AI channels in the same task and they are synchronized by hardware as they share the same timing engine.

[Kevin_Price]

---

@pkp401 wrote:

...I need higher resolution, which is possible as the wavelength sweep is linear and trigger can be interpolated to get higher wavelength resolution. However, I am not sure how to execute this.  If I use trigger in PFI line as a clock source, I cannot higher resolution.

1. Take the advice from santo_13 and capture both trigger and photodetector response at the highest feasible AI sample rate.  Then search the trigger data to find the pulse times and corresponding photodetector data.  This becomes your base data that will allow you to interpolate and estimate behavior between those triggers.

2. Keep in the back of your mind the following:  there IS a way to get slightly better timing resolution for those trigger times.  But it would fall into that realm of "more advanced" DAQ work that you shouldn't try to tackle before getting the current setup working.  Also, there's no certainty that you could even make good use of that extra precision.  But come back and revisit the question if/when you think you need it.

-Kevin P

[pkp401]

What should be the strategy to save the data after continuous sampling?
As there will be huge amount of data acquired. 

[Kevin_Price]

---

@pkp401 wrote:

What should be the strategy to save the data after continuous sampling?

The strategy should be to save the data *during* continuous sampling.  For starters, I'd recommend exploring the DAQmx capability for TDMS Logging.  There are other ways, but that's likely to be the simplest overall.

-Kevin P

[santo_13]

---

@Kevin_Price wrote:

---

@pkp401 wrote:

What should be the strategy to save the data after continuous sampling?

The strategy should be to save the data *during* continuous sampling.  For starters, I'd recommend exploring the DAQmx capability for TDMS Logging.  There are other ways, but that's likely to be the simplest overall.

-Kevin P

---

I also recommend leveraging DAQmx's TDMS logging to get the data, then develop code to read waveform from TDMS and process to extract your information of interest.