From Friday, April 19th (11:00 PM CDT) through Saturday, April 20th (2:00 PM CDT), 2024, ni.com will undergo system upgrades that may result in temporary service interruption.

We appreciate your patience as we improve our online experience.

Multifunction DAQ

cancel
Showing results for 
Search instead for 
Did you mean: 

Using external trigger for high frequency and high resolution analog sampling

Hello. I am attempting to use a NI USB-6259 to measure the spectrum of a high speed sample and hold circuit. I need to use an external clock as the analog input trigger so that I sample the S/H circuit's output at the appropriate time. I noticed a lot of distortion in the measurement so I tried measuring the output of a low distortion sine wave generator. When I use the on board clock to trigger sampling of the sine wave, I get a nice looking spectrum without harmonics. When I use an external clock as the analog input trigger, I have a wide "skirt" at the fundamental frequency rather than a narrow spike. I also see harmonics that don't show up when using the on board clock. It appears that the external clock is being synchronized with an on board clock in the DAQ. Because of this, the actual analog sampling rate is not perfectly periodic (like the external clock input or the on board clock). This causes errors in the output spectrum. I am not able to use an external clock as the analog input trigger and still make high frequency/ high resolution measurements. Is that correct?

It looks like I am forced to use the on board clock in order to get an accurate measurement. Is there a way for me to generate a 1 MHz digital output clock that is locked in frequency with the on board 1 MHz clock, but shifted in phase? If I can do this, I can use the DAQ to generate a sampling clock for my S/H circuit, and use the on board clock to trigger sampling. I could adjust the phase shift so that the S/H circuit's output is being sampled at the appropriate time.

Thanks,

Mark Hale
0 Kudos
Message 1 of 2
(2,651 Views)

Hello Mark,

Welcome to National Instruments forums. I would think that if you use an external clock coming from the S/H you would not need a start trigger since its already sampling at that time.

Now, the “skirt” like spectrum is also called non coherent Fourier transform and is that the frequency that you were trying to solve is not integer multiple of the sampling rate. This means that you will not have integer number of samples in an integer number of sine cycles. Internal clock or external, it should give the same results as long as the sample frequency is the same (and similar accuracy).

To generate two clocks that are locked to the internal reference, you can use two counter outputs (one CO task). One with an “initial delay” to make sure you sample at the correct time.

Hope this helps,

Gerardo O.
RF SW Engineering R&D
National Instruments
0 Kudos
Message 2 of 2
(2,633 Views)