[SI-LIST] Re: Current Return Vias
- From: "Chris Cheng" <Chris.Cheng@xxxxxxxxxxxx>
- To: "Ken Cantrell" <Ken.Cantrell@xxxxxxxxxxxxxxxx>, <si-list@xxxxxxxxxxxxx>
- Date: Wed, 26 Jul 2006 08:29:44 -0700
Of course it address the same issue. Even in a few selected area where return
vias are missing in the package, the results can be significant and measurable.
A question was raised to show why return vias are needed and I presented my
case. You are welcome to disagree with me but all I am asking is put your money
where your mouth is and put your believe in practice. You can't on one hand say
"return vias are not needed, all you need is plane and decoupling capacitance"
and at the same time believe you need to contact those vias between package
planes.
________________________________
From: Ken Cantrell [mailto:Ken.Cantrell@xxxxxxxxxxxxxxxx]
Sent: Wed 7/26/2006 7:32 AM
To: Chris Cheng; Ken Cantrell; si-list@xxxxxxxxxxxxx
Subject: RE: [SI-LIST] Re: Current Return Vias
Chris -
That was my point, the example was extreme and didn't address the issue.
Ken
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx
[mailto:si-list-bounce@xxxxxxxxxxxxx]On Behalf Of Chris Cheng
Sent: Tuesday, July 25, 2006 6:05 PM
To: Ken Cantrell; si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: Current Return Vias
How stupid it is for the design below vs. statements like "all you need =
is the power/ground plane and decoupling capacitance for the return =
current path, no vias needed". Wait, aren't they the same stupid concept =
?
While the example is extremely, I have been called to fix package =
designs (not my own) that conveniently forgot to provide return vias =
between upper and lower power/gnd planes because package designers want =
to steal those large core via locations for their signal escapes to =
bottom layers. If you read Larry's reply carefully, you will see similar =
experiences he refers to. Some highspeed signals experience a huge =
glitch everytime some slow JTAG signals toggles hundreds of mils away on =
a package. As it turns out, the package designer forgot to provide the =
through vias for the reference planes and the return current decided to =
take the nearest return via as return path which is on the highspeed =
signal area hundreds of mils away.
At the end, talk is cheap. If you truly believe return via is useless =
and plane and decoupling capacitance is good enough, why border to drill =
those vias on packages. Put your money where your mouth is. And if you =
are too chicken to do so, you are not practicing what you preach.
-----Original Message-----
From: Ken Cantrell [mailto:Ken.Cantrell@xxxxxxxxxxxxxxxx]
Sent: Tuesday, July 25, 2006 6:49 AM
To: Chris Cheng; si-list@xxxxxxxxxxxxx
Subject: RE: [SI-LIST] Re: Current Return Vias
Chris,
Nobody said it was OK to do stupid stuff.
Ken
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx
[mailto:si-list-bounce@xxxxxxxxxxxxx]On Behalf Of Chris Cheng
Sent: Monday, July 24, 2006 5:15 PM
To: si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: Current Return Vias
Ok, I'll give you an evidence if you dare to take it up and try on your =
=3D
clients design and see if you can keep your job after that.
In a typical multi-layer BGA design, the signal fan out is almost =3D
entirely done on the top half of the package and reference to the planes =
=3D
on top. The power and ground planes are connected through blind and =3D
buried vias through the package to the PCB. As such a package design can =
=3D
therefore stop the via connection at any plane he/she desire.
My challenge to you is just connnect all those power/ground pin on the =
=3D
peripheral (not directly underneath the die) to the bottom half of the =
=3D
package only and not to the top half where the signals are reference to. =
=3D
Let's see how much ground/power bounce you will observe.
BTDT, have you ?
-----Original Message-----
From: si-list-bounce@xxxxxxxxxxxxx
[mailto:si-list-bounce@xxxxxxxxxxxxx]On Behalf Of Lee Ritchey
Sent: Monday, July 24, 2006 3:46 PM
To: Ken Cantrell; kenny_frohlich@xxxxxxxxx; si-list@xxxxxxxxxxxxx
Subject: [SI-LIST] Re: Current Return Vias
Ken,
Well put. When someone makes the statement that ground vias are =3D
required
to provide a path for return currents, it needs to be accompanied with =
=3D
some
evidence.
> [Original Message]
> From: Ken Cantrell <Ken.Cantrell@xxxxxxxxxxxxxxxx>
> To: <kenny_frohlich@xxxxxxxxx>; <leeritchey@xxxxxxxxxxxxx>;
<si-list@xxxxxxxxxxxxx>
> Date: 7/24/2006 6:51:24 AM
> Subject: [SI-LIST] Re: Current Return Vias
>
> Kenny,
> At 133MHz, and assuming your edge rates are > 500ps, you aren't going =
=3D
to
> need them, even if you have twice as many layers. It's an important
effect,
> but at 133MHz it's not relevant. What is important, as Lee indicated, =
=3D
is
> your PDS design.
>
> Ken
>
> -----Original Message-----
> From: si-list-bounce@xxxxxxxxxxxxx
> [mailto:si-list-bounce@xxxxxxxxxxxxx]On Behalf Of Kenny Frohlich
> Sent: Saturday, July 22, 2006 1:56 PM
> To: leeritchey@xxxxxxxxxxxxx; si-list@xxxxxxxxxxxxx
> Subject: [SI-LIST] Re: Current Return Vias
>
>
> Lee,
> I'm not talking about jumping from one layer to the next adjacent =
=3D
layer
> which shares the same referrence plane. In this case, I do not need a
> ground via. But I'm asking about jumping from one layer to another =
=3D
layer
> that has a differrent reference ground plane. For example, on an 8 =
=3D
layer
> PCB stackup where the two ground planes are layers 2 and 7, the =3D
signal
> jumps from the top layer (layer 1) to the bottom layer (layer 8).
>
> Thank you
> Kenny
> Lee Ritchey <leeritchey@xxxxxxxxxxxxx> wrote:
> Kenny,
>
> It is not true that you need a "return current" via next to each layer
> changing signal via. I continue to be amazed that engineers who are
> looked upon as SI experts say such things.
>
> Imagine you have a 4 layer PCB, such as the mother board in a PC, =3D
where
> there are only two planes, one Vdd and one ground, where would such =
=3D
vias
> connect? There have been billions of these made to date that work just
> fine and have very fast signals on them. The return currents you are
> concerned about find their way from plane to plane through the =3D
collection
> of decoupling capacitors and interplane capacitance that you had to
> engineer into the power delivery system in order to make it stable. =
=3D
Focus
> on this and the return currents take care of themselves. EMI is =3D
minimized
> he same way..
>
>
>
>
> > [Original Message]
> > From: Kenny Frohlich
> > To:
> > Date: 7/22/2006 6:45:56 AM
> > Subject: [SI-LIST] Current Return Vias
> >
> > Dear Experts,
> > I understand that I need to provide ground vias next to via =3D
explictly
> for the purpose of letting return currents jump between layers. I know
> it's a requirement for high speed signals, especially differrential
> signals. Is this also required for low speed single-ended signals =3D
(133Mhz
> or slower)?
> > If this is a requirement, what would be a good signal via to ground =
=3D
via
> ratio? For example, there are five signal vias within a 1 inch area, =
=3D
how
> many ground vias do I need?
> >
> > Thank you
> > Kenny
------------------------------------------------------------------
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 FAQ wiki page is located at:
http://si-list.org/wiki/wiki.pl?Si-List_FAQ
List technical documents are available at:
http://www.si-list.org
List archives are viewable at:
//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:
