In the video, I see a 60MHz signal on a 350MHz Siglent Oszi, which is very flat at the top. Shouldn't it look like the pictures?
Is a filter switched on?
You're on the right path - it has something to do with filtering:
The textbook shows a theoretical scenario, where a square wave is restricted to a certain number of harmonics. To achieve this in any practical scenario, we would need a brickwall filter, i.e. zero attenuation in the passband with no ripple together with an abrupt transition into the stopband, which in turn has indefinite attenuation. Needless to say, that such a filter cannot exist in the real world.
In practice, high order digital filters can come close to a brickwall characteristic. However, this is not the kind of filter characteristic you'd find in a genuine oscilloscope frontend, which is supposed to have Gaussian response, which is the exact opposite to a brickwall. For reasons not discussed here, high bandwidth DSOs often have a butterworth characteristic, which is still nothing even remotely close to a brickwall.
Long story short: use a steep digital filter to define the bandwidth, then you'll get something distorted close to what is shown the textbook. Use a genuine oscilloscope frontend with its Gaussian (or Butterworth) characteristic, and you get a relatively undistorted waveform with reasonably flat pulse top.
Look at the attached screenshot, where you can compare a 350 MHz bandwidth limit with Gaussian resonse to a digtital high order FIR filter with also 350 MHz...
Home
Help
Search
Login
Register
