This reason is of course not trigger.
OK, a big part of that signal seems to be my horrible probing. I poked the bottom of the Arduino with the probe instead of clipping onto a the Dupont wire and it all got better.
Images have been removed to avoid wrong conclusions. If nobody steps up with a proper pulse generator+BNC then I might try again later with the little probe spring directly onto the AVR chip.
Well, I have not used those tools, but you're saying the result is the same? Data is shifted half a screen late and you missed half the pulse?
And what I see there, there is signs of corners wobbling (one form of aliasing) and perhaps tiny part of Gibbs
and maybe time for some of you to take on an electronics project for a change?
And what I see there, there is signs of corners wobbling (one form of aliasing) and perhaps tiny part of Gibbs
There's 130Mhz+ of signal going into something with Nyquist limit of 125Mhz. You can't expect perfection (or zero wobble!)
PS: What's going on at these corners?
This is turning into a "scopology" thread...
Time for me to unsubscribe from this thread,
Previous raw data points (Dot mode) plus persistence.
Only can be done with a waveform source with low jitter and in true real dots mode.
Previous raw data points (Dot mode) plus persistence.
Only can be done with a waveform source with low jitter and in true real dots mode.
OK, so a bit like this:
......Pic.............
PS: What's going on at these corners?
PS: What's going on at these corners?
I wasn't expecting such square corners. Is that what's expected? How close do we think these devices are representing the actual pulse waveform?
Here's the same thing with sin(x)/x off. Triggering is much tighter:
I'm not sure what conclusions can be drawn with my really crappy probing though.
This reason is of course not trigger.
OK, a big part of that signal seems to be my horrible probing. I poked the bottom of the Arduino with the probe instead of clipping onto a the Dupont wire and it all got better.
Images have been removed to avoid wrong conclusions. If nobody steps up with a proper pulse generator+BNC then I might try again later with the little probe spring directly onto the AVR chip.
This "horrible" probing is not at all what I look. I can filter these kind of things out from thinking and look important things what I think are trigger jitter and alising and fact that of course it can not do ideal Sinc reconstruction what is not even suitable in practical oscilloscope because reconstruction need be fast and only over some true sampled points. And what I see there, there is signs of corners wobbling (one form of aliasing) and perhaps tiny part of Gibbs but then (not my thinking but you told) most parts of overshoot is in your signal.
These example images are old and not at all for compare. These are from "theory and practice lesson" where I try simply way demonstrate things not at all if scope is good or bad etc but common principles. (and not even perfectly if think scalability to more speed without also BW scaling accordingly).
Yes these are made using 1us sample period. No matter. It can scale.. Reason was just that I can very accurate control rise and fall times etc and range is well below BW so that sure enough harmonics are there and I know enough sure what ADC really see.
Important is. Risetime 3.5us fall time 1us. Sampling period 1us. Of course fall have strong alias but also there is 3.5 sample in rising edge and also there can see some corner wobble.
If rise and fall is 1.5 sample period corners wobble reduce lot of.
I wasn't expecting such square corners. Is that what's expected? How close do we think these devices are representing the actual pulse waveform? Here is another sample.
Note that the sample rate in that image is only 1Msa/s and it's not a pulse (it's a slope). The reason the corners are so square is because of the huge bandwidth discrepancy between the analog bandwidth of that oscilloscope and the sampling rate being used (1000:1 ratio). This gives it the equivalent of a brickwall filter on the input.
ie. It's a sales demo, deliberately staged to fool people (and to be difficult to reproduce at home).
Related to sampling speed now analog BW before ADC is nearly like super wide so that ADC can see (and there is used it) up to very high over Nyquist. (also I have explained why I want ADC can see lot of over Nyquist...
Could what Fungus is doing considered as oversampling?
Otherwise it is quite weird that rise time stays ~const (on all Sinc=ON examples).
Isn't that what sinc filter is for?
If you look then in rf-loop example Sinc is ON yet averaging it could not result in true trace.
This is the reason the corners are nice and square compared to the other image that was posted.
(and to be difficult to reproduce at home)
How this can be difficult. There is nothing difficult at all. If this is difficult then...<advance censored>.
Just simple basic simple generator and basic simple scope, one cable and turn knobs.
This is why it is good to have opportunity to always turn all interpolations off (exept trigger engine internal "what ever"...). Just because only real data what we have are true sample points level and time and nothing else. Independent of what Keysight or who ever talks. There money talks.