[SI-LIST] Re: AW: Multi-Giga-Hertz Rigid-Flex Feasibility
- From: "Loyer, Jeff" <jeff.loyer@xxxxxxxxx>
- To: hirsh itzack <ihirshtal@xxxxxxxxxx>, si list freelist <si-list@xxxxxxxxxxxxx>
- Date: Wed, 2 Dec 2009 12:48:25 -0800
In my experience, both flex and coax have their pros/cons
Perhaps some flex can change its dimensions and hence impedance when bent
tightly, but I haven't experienced this (maybe because I keep the designs very
thin?). And most applications can (and should) limit this or form the bend
permanently.
RigidFlex (or flex) does take a lot of care to keep from introducing
inadvertent problems. Some that I've fought through include:
* too rigid - couldn't bend the thing without a hammer and vise. Stripline
that isn't hatched will do this, as will "book binding", in my experience (it
was supposed to make it more flexible, but failed miserably in my experience -
simple physics, as it turned out. You'd have to have the 2 parts of the
book-binding VERY close to each other, otherwise you've introduced a very thick
beam)
* be sure to not pass over severe bends at a right angle. Haven't experienced
failures when not doing this (always did it), but heard from experts that it's
something to avoid.
* impedance control over a hatched ground plane. Good luck modeling that one.
It's tough, and the design will probably have to be tweaked based on empirical
data. And, you'll have to be conscious of increased cross-talk, if routing
density is high.
* low loss dielectric covered with high-loss coverlay, ending up with more loss
than expected.
* you have to adjust trace width between the flex and rigid portions to keep Z0
constant. Check this carefully, beforehand and afterwards. I've had vendors
apply the same aperture to my entire trace, undoing my careful modeling.
In general, I'd say expect a spin or two to get it right. Plan on building
some test boards in your schedule.
But coax isn't without challenges, either.
* You must be sure your manufacturer really knows how to control skew between
signals, especially diff. pairs. You might want to plan on a verification
build to ensure this is done, or paying handsomely.
* At some point, you've got to attach that beautiful clean coax to your board,
and that attachment point deserves careful design, both for Signal Integrity
and strain relief reasons.
Again, I'd plan on a spin to verify your design if this is your first foray.
Regarding the original question (if I read it correctly), I think most vendors
will give you +/-10% for your design w/o too much cost adder. But you're
right, you'll have to treat the transition from flex to rigid-flex the same way
as a layer transition in your simulations - assume no correlation between
impedance of the two. But, at least there isn't a via. So, if the design
guide allows 2 or more layer transitions, you're golden. Else, you'll be in
the gray area.
My 2 cents - good luck,
Jeff Loyer
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On
Behalf Of Richard Jungert
Sent: Wednesday, December 02, 2009 12:01 PM
To: tom dogastino; hirsh itzack; Scott McMorrow; lee ritchey;
gert.havermann@xxxxxxxxxxx; si list freelist
Subject: [SI-LIST] Re: AW: Multi-Giga-Hertz Rigid-Flex Feasibility
Guys.
If you bend the flex cable the Zo changes too. BTW. Bend it some more while
observing the eye pattern on the scope and watch the eye pattern open and
close at higher speeds.
One can fabricate a flex with 1/4 oz copper. We did a 100 ohm differential, 12
inches long a couple of years ago with 1oz. copper and we could not bend it
without great force. Not good and the performance was not acceptable either.
Flex's are generally made for low speed apps like signals running inside disc
drives between motors.
I would recommend to anyone not to do a flex but to use minature coax's or
ribbonized coax's for high speed apps and signaling between rigid boards.
Richard Jungert
> From: tom@xxxxxxxxxxxxx
> To: ihirshtal@xxxxxxxxxx; scott@xxxxxxxxxxxxx; leeritchey@xxxxxxxxxxxxx
> CC: Gert.Havermann@xxxxxxxxxxx; si-list@xxxxxxxxxxxxx
> Subject: [SI-LIST] Re: AW: Multi-Giga-Hertz Rigid-Flex Feasibility
> Date: Wed, 2 Dec 2009 10:23:29 -0800
>
> Itzhak
>
> Think about what determines Zo. It is the relationship between the signal
> path and its return path(s). If the signal path and return path do not
> change from rigid to flex to rigid then Zo does no change. If you add extra
> metal or insulation at the boundaries that effect the fields then you will
> get changes in Zo. It all depends on how you design your system.
>
> Tom Dagostino
> Teraspeed(R) Labs
> 13610 SW Harness Lane
> Beaverton, OR 97008
> 503-430-1065
> 503-430-1285 FAX
> tom@xxxxxxxxxxxxx
> www.teraspeed.com
>
> Teraspeed Consulting Group LLC
> 121 North River Drive
> Narragansett, RI 02882
> 401-284-1827
>
>
> -----Original Message-----
> From: si-list-bounce@xxxxxxxxxxxxx [mailto:si-list-bounce@xxxxxxxxxxxxx] On
> Behalf Of Hirshtal Itzhak
> Sent: Wednesday, December 02, 2009 1:37 AM
> To: Scott McMorrow; Lee Ritchey
> Cc: Gert Havermann; si-list@xxxxxxxxxxxxx
> Subject: [SI-LIST] Re: AW: Multi-Giga-Hertz Rigid-Flex Feasibility
>
> Hi Scott
>
>
> As I've already told a few times in my emails, my main concern is impedance
> continuity across the rigid-flex-rigid combination.
>
>
>
> If the tolerance of the 3 sections is independent of each other, and if the
> flex tolerance is +/-20%, assuming a standard +/-10% tolerance on the rigid
> boards, this could create a substantial impedance discontinuity, e.g:
> 120-Ohm on the Flex and 90-Ohm on the Rigids.
>
>
>
> I just wanted to know if and how it's practical, and over what flex-length,
> to transfer a few GHz (up to 3.125GHz) signals from one rigid board to the
> other
>
>
>
> Thanks
>
>
>
> Itzhak
>
>
>
> ________________________________
>
> From: Scott McMorrow [mailto:scott@xxxxxxxxxxxxx]
> Sent: Tuesday, December 01, 2009 4:05 PM
> To: Lee Ritchey
> Cc: Hirshtal Itzhak; Gert Havermann; si-list@xxxxxxxxxxxxx
> Subject: Re: [SI-LIST] Re: AW: Multi-Giga-Hertz Rigid-Flex Feasibility
>
>
>
> I agree with Lee, microstrip works just fine for flex circuits with a cover
> layer. High performance stripline is also achievable in a flexible assembly
> if you crosshatch the planes. To do this, you need to do a bit of
> engineering homework in order to achieve controlled impedance and low loss.
>
> scott
>
>
> Lee Ritchey wrote:
>
> Itzhtal,
>
> I use both polyamide based flex material and Liquid Crystal Polymer from
> Rogers. Both work well. LCP has lower loss.
>
> The reason that stripline is usually not used in flex circuits is that
> makes a 3 layer assembly which is not very flexible. There is nothing
> wrong with microstrip. Works fine.
>
> Lee Ritchey
>
>
>
>
>
>
> [Original Message]
> From: Hirshtal Itzhak <ihirshtal@xxxxxxxxxx>
> <mailto:ihirshtal@xxxxxxxxxx>
> To: Havermann, Gert <Gert.Havermann@xxxxxxxxxxx>
> <mailto:Gert.Havermann@xxxxxxxxxxx> ; <si-list@xxxxxxxxxxxxx>
> <mailto:si-list@xxxxxxxxxxxxx>
> Date: 11/29/2009 4:00:51 AM
> Subject: [SI-LIST] Re: AW: Multi-Giga-Hertz Rigid-Flex Feasibility
>
> Hello Gert
>
> First - thank you for your contribution.
>
> I have a few questions regarding your email:
>
> 1) You refer to a trace width of 0.8mm as "small". This is a 31.5mil
>
>
> width, which I hardly call "small"! I regularly use 3-4 mil trace widths in
> my designs. Is there something different in Flex circuits, which enforces
> me to use such LARGE traces (compared to my regular traces)?
>
>
> 2) I see the problem you mention for the Microstrip configuration.
> Is it
>
>
> common to use Stripline on a flex circuit? I was told this is impractical,
> although I'm not sure what the reason is.
>
>
> Thanks
>
> Itzhak Hirshtal
> Elta
>
> -----Original Message-----
> From: si-list-bounce@xxxxxxxxxxxxx
> [mailto:si-list-bounce@xxxxxxxxxxxxx]
>
>
> On Behalf Of Havermann, Gert
>
>
> Sent: Tuesday, November 24, 2009 4:19 PM
> To: si-list@xxxxxxxxxxxxx
> Subject: [SI-LIST] AW: Multi-Giga-Hertz Rigid-Flex Feasibility
>
> Hello Itzhak ,
>
> All Polimide (Kapton) materials will do the job from a loss and
>
>
> dispersion prospective. The tolerance you mention seems to be unreasonable
> high for an etching tolerance. I did Flex designs with impedance control,
> 100Ohm +/-10Ohm, 0.8mm tracewidth without problems (of corse I had to pay
> extra for the small trace width).
>
>
> If you refer to an impedance tolerance due to bending and
> surrounding
>
>
> material, thats another story. If you design it in microstrip (as you are
> planning it), then everithing touching the flex might change your
> impedance. Designing it in stripline decreases the flexibility of the flex
> and increases cost.
>
>
> Look for Dupont, Grace Electron or Shengyi Flex materials for more
> info.
>
> BR
> Gert
>
>
>
> --------------------------------------------------------------------------
> Absender ist HARTING Electronics GmbH & Co. KG; Sitz der
> Gesellschaft:
>
>
> Espelkamp; Registergericht: Bad Oeynhausen; Register-Nr.: HRA 5596;
> persönlich haftende Gesellschafterin: HARTING Electronics Management GmbH;
> Sitz der Komplementär-GmbH: Espelkamp; Registergericht der
> Komplementär-GmbH: Bad Oeynhausen; Register-Nr. der Komplementär-GmbH: HRB
> 8808; Geschäftsführer: Torsten Ratzmann
>
>
> -----Ursprüngliche Nachricht-----
>
> Von: si-list-bounce@xxxxxxxxxxxxx
> [mailto:si-list-bounce@xxxxxxxxxxxxx]
>
>
> Im Auftrag von Hirshtal Itzhak
>
>
> Gesendet: Dienstag, 24. November 2009 14:21
> An: si-list@xxxxxxxxxxxxx
> Betreff: [SI-LIST] Multi-Giga-Hertz Rigid-Flex Feasibility
>
> Hello All
>
>
> My project manager intends to use a rigid-flex design in our next
>
>
> high-speed board.
>
>
>
>
> He wants to route a few pairs of 3.125GHz from a rigid section to
> another
>
>
> one through a flex section.
>
>
>
>
> Does anyone know if this is feasible? I found out that the 2
> sections,
>
>
> although spec'd to be 100-Ohm diff impedance, can still differ
> substantially - as much as 30%, because the tolerance of the flex section
> is +/-20%. Can such a design work for a substantial trace length? Should I
> restrict the trace length on the flex section to be no more than an inch or
> so, in order to reduce the impact of the supposed non-uniform impedance?
>
>
>
>
> I intend to use a 2-layer flexible laminate with the pairs on one
> layer
>
>
> and a Ground reference on the other one.
>
>
>
>
> Any advice on this matter would be helpful
>
>
>
> Thanks
>
>
>
> Itzhak hirshtal
>
>
>
>
>
> The information contained in this communication is proprietary to
> Israel
>
>
> Aerospace Industries Ltd., ELTA Systems Ltd.
>
>
> and/or third parties, may contain classified or privileged
> information,
>
>
> and is intended only for the use of the intended addressee thereof. If you
> are not the intended addressee, please be aware that any use, disclosure,
> distribution and/or copying of this communication is strictly prohibited.
>
>
> If you receive this communication in error, please notify the sender
>
>
> immediately and delete it from your computer. Thank you.
>
>
>
> This message is processed by the PrivaWall Email Security Server.
>
>
>
>
> ------------------------------------------------------------------
> 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 technical documents are available at:
> http://www.si-list.net
>
> 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
>
>
>
>
> http://www.mesago.de/de/SPS/main.htm
> Besuchen Sie uns auf der SPS/IPC/Drives 2009
> vom 24. - 26. November 2009 in Nürnberg, in Halle 10 am Stand 130.
>
> ------------------------------------------------------------------
> 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 technical documents are available at:
> http://www.si-list.net
>
> 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 message is processed by the PrivaWall Email Security Server.
>
>
>
>
> The information contained in this communication is proprietary to
> Israel
>
>
> Aerospace Industries Ltd., ELTA Systems Ltd.
>
>
> and/or third parties, may contain classified or privileged
> information,
>
>
> and is intended only for
>
>
> the use of the intended addressee thereof. If you are not the
> intended
>
>
> addressee, please be aware
>
>
> that any use, disclosure, distribution and/or copying of this
>
>
> communication is strictly prohibited.
>
>
> If you receive this communication in error, please notify the sender
>
>
> immediately and delete it from
>
>
> your computer. Thank you.
>
>
> This message is processed by the PrivaWall Email Security Server.
>
>
>
> ------------------------------------------------------------------
> 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 technical documents are available at:
> http://www.si-list.net
>
> 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 technical documents are available at:
> http://www.si-list.net
>
> 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
>
>
>
>
>
>
>
>
>
> --
> Scott McMorrow
> Teraspeed Consulting Group LLC
> 121 North River Drive
> Narragansett, RI 02882
> (401) 284-1827 Business
> (401) 284-1840 Fax
>
> http://www.teraspeed.com
>
> Teraspeed® is the registered service mark of
> Teraspeed Consulting Group LLC
>
>
>
> This message is processed by the PrivaWall Email Security Server.
>
>
>
> The information contained in this communication is proprietary to Israel
> Aerospace Industries Ltd., ELTA Systems Ltd.
> and/or third parties, may contain classified or privileged information, and
> is intended only for
> the use of the intended addressee thereof. If you are not the intended
> addressee, please be aware
> that any use, disclosure, distribution and/or copying of this communication
> is strictly prohibited.
> If you receive this communication in error, please notify the sender
> immediately and delete it from
> your computer. Thank you.
>
>
> This message is processed by the PrivaWall Email Security Server.
>
>
>
> ------------------------------------------------------------------
> 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 technical documents are available at:
> http://www.si-list.net
>
> 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 technical documents are available at:
> http://www.si-list.net
>
> 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
>
>
_________________________________________________________________
Get gifts for them and cashback for you. Try Bing now.
http://www.bing.com/shopping/search?q=xbox+games&scopeÊshback&form=MSHYCB&publ=WLHMTAG&crea=TEXT_MSHYCB_Shopping_Giftsforthem_cashback_1x1
------------------------------------------------------------------
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 technical documents are available at:
http://www.si-list.net
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 technical documents are available at:
http://www.si-list.net
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
Other related posts:
