[SI-LIST] Re: DC-blocking transmission-line

Orin,
We started this exercise some years ago and as you pointed out choosing the
right discrete capacitor was good enough, and we stopped working on this.
But the frequencies are creeping up and what was good enough some time ago
may need changes now or in the future. We understand that this may not be a
simple custom solution, but has to be adapted to the specific board design
or may be, if there is an available part the designers will consider it as a
starting point.
What I really wanted, was to start a discussion and see if the board design
community in large is interested in this...it looks that not yet.
There are also some problems related to the basics of this exercise. Of
course, we do model the behavior of the whole thing for a given
situation.... some intuitive representation will be nice..
Thanks for all the answers.
George

-----Original Message-----
From: olaney@xxxxxxxx [mailto:olaney@xxxxxxxx]
Sent: Friday, June 29, 2007 12:39 PM
To: gkorony@xxxxxxxxx
Cc: si-list@xxxxxxxxxxxxx
Subject: Re: [SI-LIST] DC-blocking transmission-line


I hope that last question is a rhetorical one, because the answer is yes.
 Some coding schemes guarantee a dozen or fewer zeroes in a row (and
ditto for the ones), while good old NRZ can provide an arbitrary
approximation to infinity.  So, there is no a priori optimum capacitor
value.  In fact, the solution to using the right capacitance, as such, is
to let the designer pick it and add it as discrete capacitors, which
makes the whole capacitance-blended-into-the-transmission-line moot.  The
only point of that exercise is if there is some advantage above and
beyond simply dropping in cheap and available discrete capacitors.
Properly implemented, any perturbation of impedance is small and
ignorable for digital signals.

In fact, when digital traces are implemented with mitered corners and
other microwave communication tricks that are needful only when you need
to squeeze out the last tenth of a dB, it's usually a sign that the
designer is ignorant.  The classic response to ignorance is overdesign,
and rules of thumb and cookbooking it are how that happens.  A good
designer knows what not to add.

As others have amply pointed out, this is a well picked over area, and
one plagued by the obvious.  That the USPTO can be gulled into granting
yet another patent in this area speaks more to the parlous quality of
patent examination than anything about the progress of technology.
Suffice to say, anything you market in this area will need to provide
information on the lower cutoff frequency and phase response, hopefully
in both the time and frequency domains.  The designer's choice will be
guided not only by matching that against the signal spectrum, but also by
how good the receiver circuitry is at recovering proper logic levels
(baseline wander compensation, DC restoration schemes, et al) after the
low frequencies are lopped off.  These are factors beyond your control.

Orin


On Fri, 29 Jun 2007 08:31:28 -0400 "George Korony" <gkorony@xxxxxxxxx>
writes:
> Content-Type: multipart/alternative;
>         boundary="----=_NextPart_002_000A_01C7BA27.E38566D0"
>
> ------=_NextPart_002_000A_01C7BA27.E38566D0
> Content-Type: text/plain;
>         charset="iso-8859-1"
> Content-Transfer-Encoding: 7bit
>
> BlankScott,
> The EM field propagates between the capacitors (or capacitor and
> GND?), but
> the capacitor's value is an interesting question. Of course to build
> a small
> and cheap device, lower the capacitor value better. But from the
> signal
> point of view the capacitor has to be big enough to transmit the
> lowest
> frequency-component of the signal in the given impedance
> environment. So the
> question is, what is the lowest frequency needed to reproduce
> accurately the
> signal at the receiver. I think this may be the same problem that
> was put on
> this forum before....how many adjacent zeros (or by the same means
> ones)
> does the signal contain?
> George
>
>
>
> ------=_NextPart_002_000A_01C7BA27.E38566D0
> Content-Type: text/html;
>         charset="iso-8859-1"
> Content-Transfer-Encoding: quoted-printable
>
> <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
> <HTML><HEAD><TITLE id=3DridTitle>Blank</TITLE>
> <META http-equiv=3DContent-Type content=3D"text/html; =
> charset=3Diso-8859-1">
> <STYLE>BODY {
>         MARGIN-TOP: 25px; FONT-SIZE: 10pt; MARGIN-LEFT: 25px; COLOR:
> #000000; =
> FONT-FAMILY: Arial, Helvetica
> }
> P.msoNormal {
>         MARGIN-TOP: 0px; FONT-SIZE: 10pt; MARGIN-LEFT: 0px; COLOR:
> #ffffcc; =
> FONT-FAMILY: Helvetica, "Times New Roman"
> }
> LI.msoNormal {
>         MARGIN-TOP: 0px; FONT-SIZE: 10pt; MARGIN-LEFT: 0px; COLOR:
> #ffffcc; =
> FONT-FAMILY: Helvetica, "Times New Roman"
> }
> </STYLE>
>
> <META content=3D"MSHTML 6.00.2800.1555" name=3DGENERATOR></HEAD>
> <BODY id=3DridBody background=3Dcid:970561812@29062007-1b79>
> <DIV><SPAN class=3D970561812-29062007>Scott, </SPAN></DIV>
> <DIV><SPAN class=3D970561812-29062007>The EM field propagates
> between =
> the=20
> capacitors (or capacitor and GND?), but the capacitor's value is an =
> interesting=20
> question. Of course to build a small and cheap device, lower the =
> capacitor value=20
> better. But from the signal point of view the capacitor has to be
> big =
> enough to=20
> transmit the lowest frequency-component of the signal in the given =
> impedance=20
> environment. So the question is, what is the lowest frequency
> needed=20
> to&nbsp;reproduce accurately the signal at the receiver. I think
> this =
> may be the=20
> same problem that was put on this forum before....how many adjacent
> =
> zeros (or by=20
> the same means ones) does the signal contain?</SPAN></DIV>
> <P><SPAN class=3D970561812-29062007>George</SPAN></P>
> <P>&nbsp;</P></BODY></HTML>
>
> ------=_NextPart_002_000A_01C7BA27.E38566D0--
>
>
> -- Binary/unsupported file stripped by Ecartis --
> -- Type: image/gif
> -- File: Blank Bkgrd.gif
>
>
>
> -- Binary/unsupported file stripped by Ecartis --
> -- Type: text/x-vcard
> -- File: George Korony.vcf
>
>
> ------------------------------------------------------------------
> 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
>
>
>
>

------------------------------------------------------------------
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

Other related posts: