I'm implementing a microscope to image volumes. To do so, I'm using an electical tunable lens driven by a voltage to displace the vocal plane. Due to inertia it's beneficial to drive it with a sinusoidal signal. So I generate a data vector, consisting of 200 Samples (one period) and output it via Startbackground continuously (with listener rebuffering). Output frequency is 500 S/s = sinusoidalfrequency 2.5 Hz. Not very fast. But somehow, it's observable (using another NI DAQ-Device), that the sinusoidal voltage changes its frequency slightly or somewhere is a small delay after a certain time... This changes my results significantly and reconstruction of 3D images is getting bad. Let's assume the camera acquires constantly, the problem is somewhere in the NI device.
Currently I generate this signal:
Could it be that it's to long/short, to less samples etc? Does anybody have experience?