[Moved] Jtter Measurement of DLL in Mentor Graphics software

[conviction]

Hi All,

Anyone can guide me as to how to go about measuring jitter for a DLL using Mentor Graphics?

Appreciate your kind help. Thanks!

 

[dick_freebird]

Re: Jtter Measurement of DLL in Mentor Graphics software

You can't measure anything. You can simulate, but that
is only as good as the "inputs" - your external and internal
noises, your layout parasitic couplings and so on. It is
likely, especially during your first several design / layout
passes, that individual device noise contributions are not
the most significant elements of jitter. Not to mention
that these sources will be invisible in the transient
analysis you need to run, to get meaningful large signal
DLL operation to be simulated. Perhaps a tnoise option
exists, or pnoise.

If you are only after the basics of running a simulation
and criticizing the results you get, for a number, then
I recommend you set up a "raster" sawtooth wave at the
reference oscillator frequency, and when you plot the
output waveform, do it against the "raster" voltage
rather than time. This will yield you an eye diagram of
sorts. But without all of the aggressors and couplings
represented, it is liable to be uselessly ideal and uninformative.

 

[conviction]

Re: Jtter Measurement of DLL in Mentor Graphics software

Thank you so much for your care and assistance...

Actually, I have a DLL circuit for which I want to measure the amount of the timing jitter in psrms with respect to the
achievable delay resolution using a voltage-controlled delay line. I'm not using an oscillator in my circuit, and I just want to know the steps or the procedure that can be followed to determine the amount of the jitter at the output of the DLL circuit (which consists of a delay line, a phase detector, a charge pump, and a loop filter).

Thanks again for your help!