subnanoseconds pulse
I'm working on designing subnanosecond pulse generator with limited of test equipment 100 MHz BW. (1 Gs) Tektronics TDS220 only.
Do you experts here have any tip and trick to estimate risetime of the generated pulse?
Thanks.
Well, you can use a sampler circuit clocked at a low frequency and recreate the much faster pulse. You form the pulse many times, and continue to sample it with a phase skew, eventually drawing out the average pulse shape on your slower scope screen. I think it is called "equivalent time sampling". Try picosecond pulse labs ap notes. You can either buy one or build the sampling front end yourself.
The other way is to borrow a spectrum analyzer and do an ifft to get back to the time domain rise time.
Added after 8 minutes:
You might be able to set up some sort of gigabit rate NAND gate-input flip-flop and a delay line to tell the pulse width. Send in the pulse, split it into two paths: one path straight to the nand gate input, the other path thru a coax line of known length then to the other nand input. Does the flip flop toggle with a pulse in? If not, then the initial pulse width was smaller than the delay line length PLUS the nand input latency. (ie, for the nand input to change the flip-flop state, both inputs had to be high. If one of the inputs was high but then went low before the other delayed input goes high, the flip flop never toggles).
Added after 8 minutes:
Check this out
Thank you very much biff44
Gated clock flip-flop with delay line technic looked very easy to deal with, I will try that.
Thanks again for your useful helps and infomation.