pkd wrote on Feb 7th, 2010, 1:37pm:Say it gives a delay from input 50% to output 50% od d1.
What do you mean by "
od d1" ?
pkd wrote on Feb 7th, 2010, 1:37pm:Now if I change the square wave freq. to 120MHz then what sort of delay should I expect.
Will it be same or it will change.
Square wave pulse is composed of many frequencies.
For example, 90MHz pulse include 0Hz, 90MHz, 180MHz, 270MHz,...., 900MHz,...
If cable have loss, transmission characteristics show frequency dispersion.
Here phase velocity have frequency dependency.
So you will observe waveform distortion(or wave packet spread) at output of cable compared to waveform at input.
Although your cable loss is negligible small for 120MHz, you can not ignore frequency dispersion effects for high frequency components, e.g. 1200MHz.
So answer to your question is that delay of 120MHz pulse is almost same as delay of 90MHz pulse.
Strictly speaking, delays of center of wave packet are almost same.
You have to study phase velocity, phase delay, group velocity, group delay, frequency dispersion, etc.
pkd wrote on Feb 7th, 2010, 1:37pm:How can I simulate this in spice/Cadence?
Use U-element or W-element in HSPICE, although I don't know what you mean by "
spice/Cadence".
You can find out many examples of coaxial cable using U-element in HSPICE manuals.
If you have Agilent RFDE(ADS), use following.
http://edocs.soco.agilent.com/display/ads2009/COAX+MDS+%28Coaxial+Cable%29
Home
Pages: 1
modelling a coaxial cable as a delay element for a wide freq. range of sq. wave (Read 4792 times)