-
|
I am confused by some behavior I am seeing. This is latest stable version of OpenHantek6022 in demo mode with only CH2 (the 500 Hz square wave) and its spectrum enabled. When I set the timebase to 10 ms/div everything looks good:
Nice square wave, and a spectrum with odd harmonics at multiples of 500 Hz. But if I increase the timebase to 20 ms/div things go wonky:
Specifically, the spectrum now shows everything happening as if the square wave were actually at 250 Hz, including odd harmonics at multiples of 250 Hz. The sample rate is 50k/s, which is 100x the base frequency though, being a square wave, I realize the harmonics go on forever. At first I thought this was a bug, but it may be that I just don't understand what is happening. Thanks for any insight! |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 2 replies
-
I do. At sampling rates < 100 kS/s, the frequency of the demo signal is reduced to show at least some recognisable traces instead of a blurred colour band - especially in "roll" mode. Also, at very high sampling rates > 10 MS/s, the channel 1 frequency starts to increase from 1000 Hz to 1200 and 1500 Hz, the same for the 500 Hz signal of channel 2 - if I remember correctly due to rounding during integer calculation. In any case, the frequency display at the bottom right is correct. BTW: If you wonder what this artifact could be - it is the histogram of the signal to the left. Do you intend to use a real Hantek6022? |
Beta Was this translation helpful? Give feedback.
-
|
Got it, that makes sense. Yes, I do have a real 6022 on order, just trying to familiarize myself with the software while waiting. Thanks for the quick reply, great application and support. Looking forward to doing some real measuring! |
Beta Was this translation helpful? Give feedback.
-
|
Great - if you like to do some HW improvements, just check the AC modification and the FG modification. |
Beta Was this translation helpful? Give feedback.
I do.
At sampling rates < 100 kS/s, the frequency of the demo signal is reduced to show at least some recognisable traces instead of a blurred colour band - especially in "roll" mode. Also, at very high sampling rates > 10 MS/s, the channel 1 frequency starts to increase from 1000 Hz to 1200 and 1500 Hz, the same for the 500 Hz signal of channel 2 - if I remember correctly due to rounding during integer calculation. In any case, the frequency display at the bottom right is correct.
I originally added this demo mode as a hack to create a reproducible test signal for function development. Take it as it is.
BTW: If you wonder what this artifact c…