RE: [IBIS-Users] IBIS model for differential bipolar current mode logic output

From: Mirmak, Michael <michael.mirmak_at_.....>
Date: Thu Aug 09 2007 - 15:17:21 PDT
Jon,

Thanks for your questions about IBIS!  The Cookbook is intended to be
general-purpose, so the recommendations are crafted to be appropriate
for a variety of designs.

Though your drawing got a little mangled in transit, I believe I can
make a few recommendations on how your eventual IBIS model should look.

1) You are correct that an I/V curve should look like a 50 ohm resistor
referenced to VCC.  However, only one I-V table should look this way.

Based on the drawing and an assumption that one resistor is internal and
one is external, I believe your I-V tables should resemble the
following:
- [Pullup]:      omitted; no pullup devices drive, so this should not be
needed
- [POWER Clamp]: should show the effects of the resistor and any diode
structures connected to VCC but not VSS; should pass through "0" when
plotted VCC-relative, per the specification
- [GND Clamp]:   should NOT show the effects of the resistor, to avoid
double counting; should only show the effects of any diode structures
connected to VSS.  Will therefore also pass through "0" axis when
plotted VSS-relative
- [Pulldown]:    tricky.  This should NOT show the resistor's effects or
the diode effects.  In the operating region, I would expect the
transistor + current source I-V table will be fairly horizontal, at a
current corresponding to the output drive current target.  The less
ideal the source, the less perfectly horizontal the I-V curve.  The Ron
of the transistor will have some effect, of course.

2) Each "half" of the driver can be modeled independently, so that two
IBIS models (each consisting of the resistor and transistor/source
combination) can be "tied together" using the [Diff Pin] keyword.

The model could therefore be of type Open_sink, IO_open_sink or similar
variants.  V-t tables could include, at a minimum, 1 [Rising Waveform]
and 1 [Falling Waveform] into a VCC-connected load, to characterize the
pulldown section turning off and on, respectively.  Using the external
50 ohm load for your V-t tables should be fine.  

This should be enough to get you started and help resolve any issues you
may have been encountering so far.  Good luck!

- Michael Mirmak
  Intel Corp.
  Chair, EIA IBIS Open Forum

-----Original Message-----
From: owner-ibis-users@server.eda.org
[mailto:owner-ibis-users@server.eda.org] On Behalf Of Jon Read
Sent: Thursday, August 09, 2007 12:17 PM
To: ibis-users@server.eda.org
Subject: [IBIS-Users] IBIS model for differential bipolar current mode
logic output


IBIS users, (See schematic below).
           I am a new creator of IBIS models. The circuit that I need to

model are bipolar.
The first one I did was for s single ended AB driver with 50 ohm output 
impedance.
I have now been chalanged to create/extract a model for a
bipolar differential current mode logic output buffer, similar in 
principle as I have attempted
to draw below. So the differential extraction processed outlined in the 
IBIS cook book seems
rather complex.
1) I know there are no paths for differential currents to flow.
2) The I/V curve should look like a 50 Ohm resistor referenced to VCC.
The problem I had extracting the IV curves is that the IV curve should 
go to zero I
at some point and this is where the VT curves are tested to match. 
However, the actual circuit is
driven into a region it was not intended to operate in and the curves 
don't match. Another way of
saying this is the circuit simply will not operate unless driven into 
the 50 ohm pullup.
Does anyone have an idea how to get around this???

Jon Read


            Inside die/pkg        |       Outside die/pkg
                                  |
                 VCC              |

  VCC                      
                  |               |

   |            
              ____|___            |                                     
____|____        
             /        \           |                                    /

        \
   50 Ohms   \        / 50 Ohms   |                            50 Ohms \

        / 50 Ohms        
             /        \           |                                    /

        \
             \        /           |                                    
\         /
          QB 
|--------|-----------|-----=========================------|         | 
      
             |        
|-----------|-----=========================----------------|
             |        | Q         |     50 Ohm Transmission lines
          | /          \ |        |
          |/            \|        |
    D ___ |              |____DB  |
          |              |        |
          |<            >|        |
          | \           /|        |
             \         /          |
             |_________|          |
                  |               |
                                |                           |
                     Current Source

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.


--------------------------------------------------------------------
|For help or to subscribe/unsubscribe, e-mail majordomo@eda-stds.org
|with the appropriate command message(s) in the body:
|
|  help
|  subscribe   ibis       <optional e-mail address, if different>
|  subscribe   ibis-users <optional e-mail address, if different>
|  unsubscribe ibis       <optional e-mail address, if different>
|  unsubscribe ibis-users <optional e-mail address, if different>
|
|or e-mail a request to ibis-request@eda-stds.org.
|
|IBIS reflector archives exist under:
|
|  http://www.eda-stds.org/pub/ibis/email_archive/ Recent
|  http://www.eda-stds.org/pub/ibis/users_archive/ Recent
|  http://www.eda-stds.org/pub/ibis/email/         E-mail since 1993
Received on Thu Aug 9 15:17:41 2007

This archive was generated by hypermail 2.1.8 : Thu Aug 09 2007 - 15:19:42 PDT