Browse si-list: This Month's Archive • Main Archive Page • Related posts • Previous by Date • Next by Date
[SI-LIST] Re: Testing method of differential intra-pair skew

From: "Loyer, Jeff" <jeff.loyer@xxxxxxxxx>
To: <changyifeng@xxxxxxxxxx>, "problem" <si-list@xxxxxxxxxxxxx>
Date: Fri, 2 Mar 2007 08:44:58 -0800
I've been able to use TDR/TDT to measure in the single picoseconds =
(though I haven't done a formal MCA - Measurement Capability Analysis).  =
Here are notes I put together for myself when I was doing this ~3 years =
ago - nothing formal, and no guarantees whatsoever.  And figures won't =
get through.  But, it might help.  It's written for the TEK TDR, but I =
don't think there's anything unique - similar technique should work for =
Agilent's.
Good luck.

___________________________________________________________
Differential, 4 Port TDR/TDT Deskew Procedure

TEK TDS8000 w/ 2 Dual TDR Heads
(4 channels)
Jeff Loyer

Disclaimer:
This procedure is provided "as-is", with no claims made to its accuracy =
or viability.

Board Design:
To facilitate these measurements, the board design should incorporate =
differential "thru" connections, including very short structures void of =
any skew (no fiberweave effects, etc.).  The launch structure should be =
designed to minimize the possibility of skew introduced by probe =
positioning ("probe" portion of launch structures minimized).

Instrument Settings:
Each waveform should be taken with an average of at least 32 =
measurements, with instrument beginning measurement when "start/stop" =
button is pressed, and stopping after those samples are completed.=20

Coarse Deskew:
Coarse deskew does not include deskew of probes (only cables and TDR =
sampling heads).   Nor does it account for skew differences related to =
even versus odd propagation mode.  The coarse skew can be accomplished =
with either mode of propagation (odd or even).  The final differential =
deskew is accomplished with only one mode of drive (odd or even =
propagation), and includes probe or final launch skew. =20


Step a: align ch1 and ch2's (with corresponding cables) TDR outputs, =
using ch3 as acquisition reference.
a1) Connect ch1 + ch1's cable to ch3, with ch1 as TDR, ch3 as TDT.  Save =
ch3 waveform as reference_a1
a2) Connect ch2 + ch2's cable to ch3, with ch2 as TDR, ch3 as TDT.  =
Adjust "TDR", "Manual Step Deskew", "C1_C2" such that ch3's waveform =
falls on top of reference_a1.
=20
ch1 + ch1's cable output is now aligned with ch2 + ch2's cable output.  =
The reference waveforms can be cleared.
=20
Step b: align ch2, ch3, and ch4's acquisition using ch1's TDR drive as =
waveform, ch2 as reference.
b1) Connect ch1 + ch1's cable + adapter to ch2's cable + ch2 with ch1 as =
TDR, ch2 as TDT.  Save ch2 waveform as reference_b1
b2) Connect ch1 + ch1's cable + adapter to ch3's cable + ch3 with ch1 as =
TDR, ch3 as TDT.  Adjust ch3's "Vertical", "Channel", "Deskew" to align =
ch3 waveform to reference_b1.
b3) Connect ch1 + ch1's cable + adapter to ch4's cable + ch4 with ch1 as =
TDR, ch4 as TDT.  Adjust ch4's "Vertical", "Channel", "Deskew" to align =
ch4 waveform to reference_b1.
NOTE: this may be an iterative process - you may have to coarsely align =
the waveforms and then recapture the reference_b1, since the reference =
waveform horizontal scale can't be adjusted.
ch 2, ch3, and ch4's acquisition are now aligned.  The reference =
waveforms can be cleared.
=20
Step c: align ch1's acquisition to ch2, using ch3's TDR as waveform ch2 =
as reference.
 c1) Connect ch3 + ch3's cable + adapter to ch2's cable + ch2 with ch3 =
as TDR, ch2 as TDT.  Save ch2 waveform as reference_c1.
c2) Connect ch3 + ch3's cable + adapter to ch1's cable + ch1 with ch3 as =
TDR, ch1 as TDT.  Adjust ch1's "Vertical", "Channel", "Deskew" to align =
ch1 waveform to reference_c1.
=20
ch1 & ch2's acquisitions are now aligned with cables only.  The =
reference waveforms can be cleared.



Final Differential Deskew:
Includes and compensates for probe skew and provides fine deskew when a =
4 port system is driven in odd mode with channel 1,2 driving and channel =
3,4 acquiring.  Compensation must be run first such that ch1 and ch2 =
amplitude is exactly the same, or within the measurement resolution of =
the TDS8000 screen ("compensation" calibrates the full scale output of =
the TDR driver and receiver.   Run "Utilities", "Compensation").

TDS 8000 Set-up (must be maintained through steps d,e, and f) for Odd =
propagation mode:
        =09
                TDR Step        Polarity                Acq             Units
Channel 1       On              Up, rising              On              V
Channel 2       On              Down, falling           On              V
Channel 3       Off             n.a.                    On              V       
=09
Channel 4       Off             n.a.                    On              V

Deskew of Channel 1-2 and 3-4 can be more easily accomplished by =
multiplying the channel by (-1) in math mode and display this as M1, M2, =
etc.,   the following illustrates:

=20
Channel 1, 2 in Odd mode propagation with apparent skew.  All graphs are =
20psec/division horizontal scale.

=20
Channel 1, 2 in odd mode, with Channel*(1) using math function in =
TDS8000mainframe.   Measurement of skew can now be accomplished (skew is =
measured to be 16.4psec).




=20
After C1-C2 Manual Deskew, channels are deskewed although there is =
apparent asymmetry between magnitudes of TDR stimulus when run in =
different modes.   Running compensation before deskew minimizes this, =
however.  Skew is measured at the 50% threshold of all signals.


Step d:  Connect probes, compensate for probe skew from step a.    Align =
ch1 and ch2's (with cables and associated probe) TDR outputs, using ch3 =
as acquisition reference.
d1) Connect ch1 + ch1's probe to ch3 using a short Thru, with ch1 as =
TDR, ch3 as TDT.  Save ch3 waveform as reference_d1
d2) Connect ch2 + ch2's probe to ch3 using a short Thru, with ch2 as =
TDR, ch3 as TDT.  Adjust "TDR", "Manual Step Deskew", "C1_C2" such that =
ch3's waveform falls on top of reference_d1.
=20
ch1 + ch1's probe output is now aligned with ch2 + ch2's probe output.  =
The reference waveform can be cleared.

Step e:  With probes, align ch3, and ch4's acquisition using ch1's TDR =
drive as waveform, ch2 as reference.
e1) Connect ch1 probe to ch3's probe using short Thru, with ch1 as TDR =
and ch3 as TDT.  Adjust ch3's "Vertical", "Channel", "Deskew" to align =
ch3 waveform to reference_e1.
e2) Connect ch1 probe to ch4's probe using short Thru with ch1 as TDR =
and ch4 as TDT.  Adjust ch4's "Vertical", "Channel", "Deskew" to align =
ch4 waveform to reference_e1.


Step f: align ch1's acquisition to ch2, using ch3's TDR, both probes =
open, open reflected TDR
 f1) Raise ch1 and ch2 probe such that there is open reflected TDR only =
(not TDT) from the probe tips.
 f2)  Adjust either C1 or C2 vertical deskew.  Should be very small =
value (<several psec's) representing only probe skew.  ch1 & ch2's =
acquisitions are now aligned including the probes.


Final "Even Mode" Deskew:
Similar to differential deskew, but both channels drive in same =
direction.  The timing within the TDR head changes between odd and even =
modes, so deskew performed for the odd mode is not valid for the even =
mode.


Jeff Loyer

-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] =
On Behalf Of changyifeng
Sent: Wednesday, February 28, 2007 5:43 PM
To: problem
Subject: [SI-LIST] Testing method of differential intra-pair skew
Importance: High

Hi all,

=20

I am looking for accurate testing methods(or apparatus) of differential =
intra-pair skew.

=20

Because of the resolving power of TDR and the limited bandwidth of VNA, =
there are some defects in some methods such as TDR, TDT and VNA testing.

=20

What is the best solution when the intra-pair skew is between 1ps to =
5ps.(or 10ps)

=20

Thanks.



Regards



Chang Yifeng

Email=A3=BAchangyifeng@xxxxxxxxxx
*************************************************************************=
**************
=A1=A1=A1=A1This e-mail and its attachments contain confidential =
information from HUAWEI, which is intended only for the person or entity =
whose address is listed above. Any use of the information contained =
herein in any way (including, but not limited to, total or partial =
disclosure, reproduction, or dissemination) by persons other than the =
intended recipient(s) is prohibited. If you receive this e-mail in =
error, please notify the sender by phone or email immediately and delete =
it!
*************************************************************************=
**************
------------------------------------------------------------------
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:
http://www.freelists.org/webpage/si-list

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field


List technical documents are available at:
                http://www.si-list.net

List archives are viewable at:    =20
                http://www.freelists.org/archives/si-list
or at our remote archives:
                http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
                http://www.qsl.net/wb6tpu
 =20
------------------------------------------------------------------
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:
http://www.freelists.org/webpage/si-list

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field


List technical documents are available at:
                http://www.si-list.net

List archives are viewable at:     
                http://www.freelists.org/archives/si-list
or at our remote archives:
                http://groups.yahoo.com/group/si-list/messages
Old (prior to June 6, 2001) list archives are viewable at:
                http://www.qsl.net/wb6tpu
References:

[SI-LIST] Testing method of differential intra-pair skew
From: changyifeng
[SI-LIST] Re: Testing method of differential intra-pair skew

Other related posts:
