Arpad,
Exactly so. Based on your comment, onw would want to terminate all of the
not used ports with the Touchstone file reference impedance of the port,
not the high impedance as suggested in the BIRD. This is not so obvious if
the interconnect was an IBIS-ISS subckt.
In a crossing e-mail I asked Scott if interconnect sNp are measured with
all of the non-measured pins (and die pads) left unterminated, then is
this self-induce crosstalk already included in the s12. I just got Scott's
response which is repeated here:
Correct. Which is why for correlation to measurements this is the way to
go. Measurement are artificial, and we need to replicate the same
artificial environment in modeling as we do in measurements. In this
case, we'll see the traveling crosstalk wave from excitation of the open
circuit nets superimposed on the measurement.
This is one of the reasons I hate microprobes for measurement. Since it is
not easy to terminate unused lines, the measurements do not match
terminated modeling 1-to-1, unless you correct for the open circuit
conditions. Connectorized measurements are more faithful to the
transmission line environment, since we can terminate unused connectors.
But then we have the issue of removing the connector from the measurement.
Either way, we are dealing with the law of Conservation of Misery. -
Misery is neither created nor destroyed. It is merely transferred from one
place (or person) to another.
I do not think we can solve this problem because the answer will always be
"It Depends". I think the following statement:
Note that the terminals remain technically open, and terminations
connected by the EDA tool are intended to approximate open-circuit
conditions."
With something like:
Note that the terminals remain technically open, and the User/EDA tool may
choose an appropriate termination."
Walter
Walter Katz
<mailto:wkatz@xxxxxxxxxx> wkatz@xxxxxxxxxx
Phone 303.449-2308
Mobile 303.335-6156
From: ibis-interconn-bounce@xxxxxxxxxxxxx
[mailto:ibis-interconn-bounce@xxxxxxxxxxxxx] On Behalf Of Muranyi, Arpad
Sent: Friday, July 21, 2017 11:22 AM
To: IBIS-ATM <ibis-macro@xxxxxxxxxxxxx>; IBIS-Interconnect
<ibis-interconn@xxxxxxxxxxxxx>
Subject: [ibis-interconn] Re: [ibis-macro] Re: Re: Question about
references in BIRD189, and related comments about AMI Flows
Walter,
Regarding ".those pins that do not have connections on the board.", I read
the following
statement in the latest draft of BIRD189:
This optional subparameter defines the termination that is to be applied
by the EDA tool during simulation to the terminals of any IBIS-ISS
subcircuit or Touchstone network that is not being used in the
[Interconnect Model]/[End Interconnect Model] group.
Note that the highlighted text basically says that the purpose of this
parameter is to terminate
those terminals/ports which are not mentioned on any of the "terminal
lines" in the
[Interconnect Model] keyword. I believe that terminating ports for this
reason is a little
different from terminating ports in situations you are talking about (when
the pin of a device
is not connected to any traces on the board).
If the purpose of this termination is to make the interconnect model look
like as if that
terminal/port didn't exist, wouldn't you want to terminate it with the
"correct" termination
value to eliminate any reflections on that port, and/or potentially allow
to EDA tool to
numerically reduce the Touchstone file to a lower number of ports?
Thanks,
Arpad
=========================================================================
From: ibis-interconn-bounce@xxxxxxxxxxxxx
<mailto:ibis-interconn-bounce@xxxxxxxxxxxxx>
[mailto:ibis-interconn-bounce@xxxxxxxxxxxxx] On Behalf Of Walter Katz
Sent: Friday, July 21, 2017 9:01 AM
To: Scott@xxxxxxxxxxxxx <mailto:Scott@xxxxxxxxxxxxx> ; Muranyi, Arpad
<Arpad_Muranyi@xxxxxxxxxx <mailto:Arpad_Muranyi@xxxxxxxxxx> >; IBIS-ATM
<ibis-macro@xxxxxxxxxxxxx <mailto:ibis-macro@xxxxxxxxxxxxx> >;
IBIS-Interconnect <ibis-interconn@xxxxxxxxxxxxx
<mailto:ibis-interconn@xxxxxxxxxxxxx> >
Subject: [ibis-interconn] Re: [ibis-macro] Re: Re: Question about
references in BIRD189, and related comments about AMI Flows
Scott,
Thanks for the clarifications. Assuming most pins of a component are in
fact connected to something on the board, we need to be concerned about
used connections in the package that are adjacent (near) interconnect on
the package that are not connected to anything on the Board. What is the
EDA tool to do? What is the package model maker to do?
If a pin is marked NC in a component, and in fact, that pin has routing in
the package, then the most accurate thing to do is to include all these NC
pins along with the pins of interest in the terminals of the interconnect
mode, and then set the termination of all of the unused terminals to ~50
Ohms. An alternative is to make the interconnect model for a specific pin
include all of the significant aggressor pins. The EDA tool could chose to
terminate those aggressor pins that have connections on the board with ~50
Ohms, and those pins that do not have connections on the board (whether NC
or not) with 1Meg Ohm. Without this help from the interconnect model
maker, the EDA tool can either run with all 1 MegOhm terminations
(pessimistic), or all ~50 Ohm terminations (optimistic), and compare the
results.
Walter
Walter Katz
<mailto:wkatz@xxxxxxxxxx> wkatz@xxxxxxxxxx
Phone 303.449-2308
Mobile 303.335-6156
From: ibis-interconn-bounce@xxxxxxxxxxxxx
<mailto:ibis-interconn-bounce@xxxxxxxxxxxxx>
[mailto:ibis-interconn-bounce@xxxxxxxxxxxxx] On Behalf Of Muranyi, Arpad
Sent: Thursday, July 20, 2017 4:04 PM
To: IBIS-ATM <ibis-macro@xxxxxxxxxxxxx <mailto:ibis-macro@xxxxxxxxxxxxx>
; IBIS-Interconnect <ibis-interconn@xxxxxxxxxxxxx<mailto:ibis-interconn@xxxxxxxxxxxxx> >
