All,
Figure 16 in IBIS describes in a very general way a Device Under Test
(DUT). It does not do a very good job of distinguishing between rail
voltages under test vs rail terminals for a Device In Action (DIA). And
the section is very confusing on how to hook up C_comp for compensation
for DUT data and how to hook up C_comp when DIA (rail voltages floating
and not constant).
The most common case in IBIS is
[Pulldown Reference]==[GND Clamp Reverence]==0.0V
[POWER Clamp Reference]==[Pullup Reference]==[Voltage Range]
The second most common case in IBIS is ECL
[GND Clamp Reverence]=="Data Book VEE"
[POWER Clamp Reference]==[Pullup Reference]== [Pulldown Reference]="Data
Book VCC"
Focusing on these two most most common cases, what does this look like
when measuring the device (DUT) to generate the model and using the model
for a device in action. I built a "Virtual Test Bench" to represent the
measurement one would do on a real test and operating system. I put probes
at location where one can actually make measurement in real hardware
(although it is much easier to put SPICE probes in a virtual test bench
then to instrument the real hardware). It is also possible to put probes
in a virtual test bench that are impossible to instrument in hardware and
are in fact meaningless measurements (e.g. a differential probe between a
die terminal and the reference voltage of a VRM 3 feet away).
Please review these schematics. These would be interesting to discuss in
ATM, Interconnect, GND Editorial or the Open Forum.
Walter
Walter Katz
<mailto:wkatz@xxxxxxxxxx> wkatz@xxxxxxxxxx
Phone 303.449-2308
Mobile 303.335-6156
Attachment:
image001.emz
Description: Binary data
Attachment:
oledata.mso
Description: Binary data
Attachment:
Legacy_DUT.JPG
Description: JPEG image
Attachment:
Legacy_DIA.JPG
Description: JPEG image
Attachment:
ECL_DUT.JPG
Description: JPEG image
Attachment:
ECL_DIA.JPG
Description: JPEG image