Or you use a VNA and perform an IFFT to get the step response of your channel.
This also solves the problems with SNR, signal amplitude at high frequency,
linearity of your TDR pulse and reference plane calibration.
As an added bonus you get a defined and reproducible measurement uncertainty.
Chris
Sent from my mobile device
----- Message from si-list-bounce@xxxxxxxxxxxxx on Fri, 04 Mar 2016 00:04:00
GMT -----
From: si-list-bounce@xxxxxxxxxxxxx
To: gmanikandan20@xxxxxxxxx, "Bartlett, Josiah" <josiah.bartlett@xxxxxxxxxxxxx>
CC: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: TDR Rise time
The formula is saying that if you were to perform a Fourier transform on the
TDR step, then its 3-db down frequency
content is ~0.35/(10-90% Rise-time).
If the data rate is 28 Gb/s, then one Unit Interval is about 36 ps.
So, if you want to probe a channel (or a system) which is supposed to handle
36-ps pulses, then you'd want to hit it with a
TDR step that has a lot of energy at the 14 Ghz Nyquist. That would mean a DTR
rise-time of > "0.35/14 = 25 ps". But that
is barely enough, so personally, I would hit such a system with a TDR of at
least 10 ps, so I have a lot more energy in my
region of interest.
I reality, if you can de-convolve the incident signal out of the result
(output), then you can use a slower rise-time, but in order to
avoid having to know the exact shape and frequency content of your incident
wave, use a much faster rise than the formula gives
you at Nyquist, and make your life simple...if you can afford such a TDR :-)
To dig deeper, if you average the TDR signal a lot, and your Signal to noise
ratio is very high, then you can use a TDR rise/fall
slower than 25ps...But then again, why go into all that if you can afford a TDR
which is much faster than 25 ps!
Best Regards
Samie Samaan
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On ;
Behalf Of Mani
Sent: Thursday, March 03, 2016 3:44 PM
To: Bartlett, Josiah
Cc: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: TDR Rise time
Hi
Thank you all. I understand the theory behind TDR. But I am confused about
calculating the rise time to be used for simulation.
In the formula 0.35/BW. what is the bandwidth here? is it the data Nyquist
frequency?
Lets say the data rate is 28Gbps, how to calculate rise time based on 10%-90%
calculation and how does it change for 20%-80%?
Regards
Dan
On Thu, Mar 3, 2016 at 3:40 PM, Bartlett, Josiah <
josiah.bartlett@xxxxxxxxxxxxx> wrote:
Dan,
http://www.tek.com/document/fact-sheet/tdr-impedance-measurements-foun
dation-signal-integrity has some basics of TDR calculations and
applications that you may be interested in.
Best wishes,
Josiah Bartlett
Principal Engineer- High Performance Probes
Phone: 503.627.2946
www.tektronix.com
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx
[mailto:si-list-bounce@xxxxxxxxxxxxx]
On Behalf Of Mani
Sent: Thursday, March 03, 2016 3:20 PM
To: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] TDR Rise time
Hi,
Can I get help in understanding TDR rise time calculation? I came
across 10% to 90% calculation and 20% to 80% as well. Just wanted to
know what is the difference between them and how to calculate the TDR rise
time.
Thanks for the help.
Regards
Dan
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field
or to administer your membership from a web page, go to:
//www.freelists.org/webpage/si-list
For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field
List forum is accessible at:
http://tech.groups.yahoo.com/group/si-list
List archives are viewable at:
//www.freelists.org/archives/si-list
Old (prior to June 6, 2001) list archives are viewable at:
http://www.qsl.net/wb6tpu
