Secrets of cross-hatched flex design
1) Do not let a layout person or automated CAD process design the cross-hatch.
2) Design the cross-hatch holistically w.r.t. the traces.
3) Single-ended traces follow the lengthwise copper strips in the cross-hatch,
keeping return path tight and continuous Isolate adjacent single-ended traces
with an additional lengthwise copper strip in between.
4) Design differential trace intrapair pitch to match cross-hatch lengthwise
strip pitch, and align pairs with strips. Differential pairs must be designed
to be routed symmetrically w.r.t. the cross-hatch pattern.
5) If you would like wider trace metal for lower losses, align SE-traces with
the center of the cross-hatch holes, so that the trace repetitively crosses
copper/hole/copper/hole. Center the trace so that the return path around the
hole is symmetric. For additional Xtalk isolation, keep at least one row of
holes between adjacent traces. Impedance of the trace will raise due to the
open hole, allowing for wider copper in the trace. A general 2D solver like
Ansys 2D can be used to determine the trace impedance across the hole, with the
left and right copper strips. This can be cascaded with a model of the section
that crosses over copper to determine the correct average impedance. This
forms a repetitive discontinuity filter (Bloch filter) that will have a
resonance. Keep holes small to push the resonance up.
6) Same technique can be used for differential pairs crossing holes, just align
the trace symmetrically with holes under each trace and the lengthwise copper
strip in the center. Compute the cross-sections, concatenate and generalize.
It's been a good 10 years since my last flex design (I've moved to twin-ax),
but the stuff works quite well if you actually engineer it, rather than phone
in the design with a flex layout house.
Just sayin'
Scott
Scott McMorrow, CTO Signal Integrity Group
Samtec
Office 401-284-1827 | +1-800-726-8329
www.samtec.com
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On ;
Behalf Of Kai Keskinen
Sent: Saturday, April 8, 2017 2:00 PM
To: alan@xxxxxxxxxxxxxxxxx; si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: flex crosshatch reference plane and LVDS differental
pairs
A long time ago when working for Nortel, we measured the loss of flex with
solid and cross-hatched ground planes and the loss curves diverged around a few
hundred MHz with of course the solid reference plane having significantly less
loss at even 500MHz.
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On ;
Behalf Of Alan Hilton-Nickel
Sent: Saturday, April 8, 2017 12:43 AM
To: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: flex crosshatch reference plane and LVDS differental
pairs
[Well replying from my phone did not work out at all. :-(] Are you running
multiple channels?
When you do your analysis, be careful to look at the effects of crosstalk on
adjacent lines. The cross hatching will distort the return currents, pushing
them closer together.
If you're running high enough data rates to be concerned about surface
roughness or fiber weave, you won't be happy with the cross hatch.
BTW, I'm no thermal expert, but how does the cross hatch improve thermal
issues? The hatch has less copper with a smaller, more restrictive profile, so
it should act as a thermal resistor. Any thermal problem I've come across
needed more copper, not less, but I suppose there could be instances where
you'd want to restrict heat transfer.
Alan Hilton-Nickel
On 4/7/2017 6:43 PM, Bert Simonovich (Redacted sender bertsimonovich for
DMARC) wrote:
I agree with Ryan and Al on using a good field solver for accurate~50GHz.
impedance and loss modeling over frequency. I recently benchmarked
both Polar Si9000 Transmission Line Field Solver and Simbeor by
applying my Cannonball modeling technique to the Huray roughness model
and compared results against de-embedded measured data from Al's CMP28
Wildriver modeling platform. Both tools gave excellent correlation to
using
Bert Simonovich
Signal/Power Integrity Practitioner | Backplane Specialist | Founder
LAMSIM Enterprises Inc.
Web Site: http://lamsimenterprises.com
Blog: http://blog.lamsimenterprises.com/
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx
[mailto:si-list-bounce@xxxxxxxxxxxxx] On Behalf Of Alfred P. Neves
Sent: 7-Apr-17 8:09 PM
To: rdawson16@xxxxxxxxxxx
Cc: si-list
Subject: [SI-LIST] Re: flex crosshatch reference plane and LVDS
differental pairs
Actually, the new Polar tools are pretty good at estimating impedance
over frequency, some of the most popular EDA tools that you may have
been
are horrid, however. The data was taken using our Channel Modelingaccurate.
platforms and speaks to the necessity of careful benchmarking even an
EDA package that costs upwards of $100K.
As Ryan stated, we are also using Simbeor due its cost/performance and
accurate loss modeling.
- Al
Products for the Signal Integrity Practitioner
Alfred P. Neves
Chief Technologist
Office: 503-679-2429
www.wildrivertech.com <http://www.wildrivertech.com/>
2015 Best In Design&Test Finalist
On Apr 7, 2017, at 4:26 PM, Randy Dawson <rdawson16@xxxxxxxxxxx> wrote:from a few engineers that impedance calculators (Polar) are not so
I am looking for SI articles on this and not finding much. I have
heard
I am also reading that I should expect some attenuation GHz
frequencies. I am forced into crosshatch vs solid due to thermal issues.
I do have the Sigrity field solver and will experiment and let youwhat I find.
know
What are you're experiences with this?
Randy
------------------------------------------------------------------
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
------------------------------------------------------------------
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
---
This email has been checked for viruses by AVG.
http://www.avg.com
------------------------------------------------------------------
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
