开云体育

ctrl + shift + ? for shortcuts
© 2025 开云体育

measuring spectrum of a single pulse not feasible with TinySA?


 

I want to measure the frequency spread of a single pulse arising from capacitor discharge into a coil, but it sounds like from the discussion here on measuring radar pulses that this will not be feasible with TinySA. The sweep rate is too slow (?) I can guess what the fundamental frequency ought to be based on the LC formula and fish around that area in zero span mode (?), but I'd really like to see what kind of harmonics are contained in the pulse and what their relative amplitudes are compared to the fundamental.? Any ideas besides having to spend thousands more on another spectrum analyzer??


 

If you repeat the pulse and use MAX HOLD, the scanning will eventually capture the generated fundamental and harmonics.
Depending on the duty cycle of the generated output this "eventually" may take a long time.
--
Designer of the tinySA
For more info go to


 

Modern digital o'scopes contain many useful add-ons which are impossible with the older strictly analog approaches.? I'd also suggest capturing the discharge trace on a modern o'scope and using the FFT function to transform your time domain trace into the frequency domain.? Just an alternative to what the designer of the TinySA, Eric, suggested.? However, to capture most of the harmonics, the o'scope must be rather fast.? That can be determined by the time constant of your LC circuit.? Then, assume the o'scope BW must be rated to at least X10 that time constant.? And remember, the spec BW is the -3 dB (power) frequency.

Dave - W?LEV

On Tue, Apr 1, 2025 at 8:39?AM Erik Kaashoek via <erik=[email protected]> wrote:
If you repeat the pulse and use MAX HOLD, the scanning will eventually capture the generated fundamental and harmonics.
Depending on the duty cycle of the generated output this "eventually" may take a long time.
--
Designer of the tinySA
For more info go to



--
Dave - W?LEV



 

You can also do the FFT calculation off line in python if you can capture a data plot from the scope. Most modern scopes let you transfer them to a PC to do this even if they can't do the FFT internally properly.
?
Here's a you tube video with pointers to Python code that does exactly this.
?
?
Dave's comments about the needed scope bandwidth still apply, however.
?
I hope this helps without being too off topic...
?
M